validation_models.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "cc4e91cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "from data_wrangle import data_wrangle as dw\n",
    "from datetime import datetime, date, timedelta\n",
    "import time\n",
    "import itertools\n",
    "import pickle\n",
    "\n",
    "# Basics\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import yfinance as yf \n",
    "\n",
    "\n",
    "#Visualization\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Optimization part\n",
    "from cvxopt import matrix\n",
    "from cvxopt.solvers import qp\n",
    "from sklearn.covariance import LedoitWolf\n",
    "from pypfopt import risk_models, expected_returns, plotting, EfficientFrontier\n",
    "\n",
    "# sclearn RandomForest - Returns, ML Metrics and tools\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import RandomizedSearchCV, cross_val_score, GridSearchCV\n",
    "from sklearn.pipeline import make_pipeline\n",
    "\n",
    "# GARCH - Volatility\n",
    "from arch import arch_model\n",
    "from statsmodels.graphics.tsaplots import plot_acf, plot_pacf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "54b38d00",
   "metadata": {},
   "outputs": [],
   "source": [
    "# class model_validation:\n",
    "    \n",
    "#     def __init__(self, tickers, start, end, features_tickers):\n",
    "#         self.tickers = tickers\n",
    "#         self.start = start\n",
    "#         self.end = end\n",
    "\n",
    "#         self.features = yf.download(features_tickers, start=self.start, end=self.end, \n",
    "#                                     interval='1wk').fillna(method='ffill')\n",
    "#         self.ohlc = yf.download(self.tickers, start=self.start, end=self.end, interval='1wk')\n",
    "    \n",
    "   \n",
    "#     def tech_features(self):\n",
    "#         inst = dw()\n",
    "#         output = pd.DataFrame()\n",
    "#         for i in self.ohlc.columns.levels[1]:\n",
    "\n",
    "#             tech = inst.get_tech_features(close=self.ohlc['Adj Close'][i],\n",
    "#                                               volume=self.ohlc['Volume'][i],\n",
    "#                                               low=self.ohlc['Low'][i],\n",
    "#                                               high=self.ohlc['High'][i],\n",
    "#                                               concat=True).dropna()\n",
    "#             tech.columns = pd.MultiIndex.from_product([[i], tech.columns])\n",
    "\n",
    "#             output = pd.concat([output, tech], axis=1)\n",
    "            \n",
    "#         self.ohlc = self.ohlc.loc[output.index]\n",
    "#         self.features = self.features['Adj Close'].loc[output.index]\n",
    "#         self.tech = output\n",
    "#         print('Done!')\n",
    "        \n",
    "\n",
    "#     def binary_classificator(self, x):\n",
    "#         if x > 0:\n",
    "#             return 1\n",
    "#         else:\n",
    "#             return -1\n",
    "        \n",
    "    \n",
    "#     def binary_map(self, prices):\n",
    "#         output = pd.DataFrame()\n",
    "#         for i in prices.columns:\n",
    "#             # binary classification of returns\n",
    "            \n",
    "#             up_down = pd.DataFrame(prices[i].pct_change().dropna())\n",
    "#             up_down[i] = up_down[i].map(model_validation.binary_classificator)\n",
    "\n",
    "\n",
    "#             # concatenate all companies in dataframe\n",
    "#             output = pd.concat([output, up_down], axis=1)\n",
    "\n",
    "#         return output\n",
    "    \n",
    "    \n",
    "    \n",
    "#     def train_test_split(self, window, val_size, test=False, test_size=False):\n",
    "#         self.window = window\n",
    "#         self.val_size = val_size\n",
    "\n",
    "        \n",
    "#         self.ohlc_val = self.ohlc[:self.window + self.val_size]\n",
    "#         self.tech_val = self.tech[:self.window + self.val_size]\n",
    "#         self.feat_val = self.features[:self.window + self.val_size]\n",
    "#         self.binary_val = model_validation.binary_map(self.ohlc['Adj Close'])[:self.window + self.val_size - 1]\n",
    "#         self.returns_val = 100 * self.ohlc['Adj Close'].pct_change().dropna()[:self.window + self.val_size - 1]\n",
    "\n",
    "#         print('Initial training window is included')\n",
    "#         print(f'ohlc val {len(self.ohlc_val)}, tech val {len(self.tech_val)}')\n",
    "#         print(f'binary_val {len(self.binary_val)}, returns_val {len(self.returns_val)}')\n",
    "#         assert len(self.ohlc_val) == len(self.tech_val)\n",
    "#         assert len(self.binary_val) == len(self.returns_val)\n",
    "#         assert len(self.ohlc_val) == len(self.binary_val) + 1\n",
    "#         assert self.ohlc_val.index[1] == self.binary_val.index[0]\n",
    "        \n",
    "#         assert self.feat_val.index[0] == self.tech_val.index[0]\n",
    "\n",
    "#         if test:\n",
    "            \n",
    "#             self.ohlc_test = self.ohlc[self.window + self.val_size:]\n",
    "#             self.tech_test = self.tech[self.window + self.val_size:]\n",
    "#             self.feat_test = self.features[self.window + self.val_size:]\n",
    "#             self.binary_test = model_validation.binary_map(self.ohlc['Adj Close'])[self.window + self.val_size :]\n",
    "#             self.returns_test = 100 * self.ohlc['Adj Close'].pct_change().dropna()[self.window + self.val_size :]\n",
    "#             self.test_size = len(self.ohlc_test) - self.window\n",
    "            \n",
    "#             print('Initial training window is included')\n",
    "#             print(f'ohlc test {len(self.ohlc_test)}, tech test {len(self.tech_test)}')\n",
    "#             print(f'binary_test {len(self.binary_test)}, returns_test {len(self.returns_test)}')\n",
    "            \n",
    "#             assert len(self.ohlc_test) == len(self.tech_test)\n",
    "#             assert len(self.binary_test) == len(self.returns_test)\n",
    "#             assert len(self.ohlc_test) == len(self.binary_test) + 1\n",
    "#             assert self.ohlc_test.index[1] == self.binary_test.index[0]\n",
    "#             assert self.feat_test.index[0] == self.tech_test.index[0]\n",
    "        \n",
    "        \n",
    "#     def classify_next_bar(self, X, y, RF_params=(10, None, None), features=None):\n",
    "#         \"\"\"\n",
    "#         X, y - features and target-vector based on a shape of rolling window\n",
    "#         X - tech features and external data for the date t-1\n",
    "#         y - binary target - [1 - up, -1 - down] for time t\n",
    "#         X_out - 1 day forward features to make prediction\n",
    "#         Make assert statement to ensure that\n",
    "        \n",
    "#         \"\"\"\n",
    "#         assert X.shape[0] - 1 == y.shape[0]\n",
    "#         assert len(X) == len(features)\n",
    "#         X_data = pd.concat([X, features], axis=1)\n",
    "#         X_train_data = X_data.shift().dropna()\n",
    "        \n",
    "#         assert X_train_data.shape[0] == y.shape[0]\n",
    "        \n",
    "#         X_forecast = pd.DataFrame(X_data.iloc[-1, :]).T\n",
    "        \n",
    "#         # Fit the model - returns classifier\n",
    "#         model = RandomForestRegressor(n_estimators=RF_params[0], \n",
    "#                                        max_depth=RF_params[1], \n",
    "#                                        max_leaf_nodes=RF_params[2], \n",
    "#                                        random_state=42,\n",
    "#                                        n_jobs=-1)\n",
    "#         model.fit(X_train_data, y)\n",
    "#         bar = model.predict(X_forecast)\n",
    "        \n",
    "#         return bar\n",
    "    \n",
    "#     def GARCH_model(self, returns, GARCH_params=(1, 1), summary=False):\n",
    "#         assert type(returns) == pd.Series\n",
    "#         model = arch_model(\n",
    "#             returns,\n",
    "#             p=GARCH_params[0],\n",
    "#             q=GARCH_params[1],\n",
    "#             rescale=False\n",
    "#         )\n",
    "\n",
    "#         model_res = model.fit(disp=0)\n",
    "#         next_pred = model_res.forecast(horizon=1, reindex=False).variance.iloc[0,0] ** 0.5\n",
    "        \n",
    "#         if summary:\n",
    "#             aic = model_res.aic\n",
    "#             return next_pred, aic\n",
    "\n",
    "#         return next_pred\n",
    "    \n",
    "    \n",
    "    \n",
    "#     def rolling_window(self, tech, binary, window, test_size, returns, \n",
    "#                        val_model='Both', RF_params=(10, None, None), \n",
    "#                        GARCH_params=(1, 1), summary_garch=False, features=None):\n",
    "#         \"\"\"\n",
    "#         val_model: what model to validate. Relevant parameters: 'RF', 'GARCH', 'Both'\n",
    "        \n",
    "#         \"\"\"\n",
    "#         if val_model == 'Both':\n",
    "\n",
    "\n",
    "#             prediction_ret = pd.DataFrame(columns=binary.columns)\n",
    "#             prediction_std = pd.DataFrame(columns=returns.columns)\n",
    "\n",
    "#             total_time = time.time()\n",
    "\n",
    "#             for i in range(-test_size, 0):\n",
    "#                 start_time = time.time()\n",
    "#                 # Create rolling window to train the model\n",
    "#                 # 100 - window size, -1 correction for lag of tech features\n",
    "#                 X_train = tech.iloc[-window-1+i:i]\n",
    "#                 y_train = binary.iloc[-window+i:i]\n",
    "\n",
    "#                 y_train_ret = returns.iloc[-window+i:i]\n",
    "\n",
    "#                 # RETURNS PART\n",
    "#                 #list for returns predictions\n",
    "#                 pred_list = np.array([])\n",
    "\n",
    "#                 #VOLATILITY PART\n",
    "#                 pred_list_std = np.array([])\n",
    "\n",
    "#                 for company in binary.columns:\n",
    "#                     # RETURNS PART\n",
    "#                     # Train a model\n",
    "#                     #print(f' Start {company}')\n",
    "\n",
    "#                     next_pred = model_validation.classify_next_bar(X_train[company], y_train[company], RF_params)\n",
    "#                     # Collect results and append them\n",
    "#                     pred_list = np.append(pred_list, next_pred)\n",
    "\n",
    "\n",
    "#                     #VOLATILITY PART\n",
    "#                     # Train a model\n",
    "#                     next_pred_std = model_validation.GARCH_model(y_train_ret[company], GARCH_params, summary=summary_garch)\n",
    "#                     # Collect results and append them\n",
    "#                     pred_list_std = np.append(pred_list_std, next_pred_std)\n",
    "\n",
    "#                     #print(f' End {company}')\n",
    "\n",
    "#                 pred_list = pd.DataFrame(pred_list, index=binary.columns).T\n",
    "#                 prediction_ret = pd.concat([prediction_ret, pred_list], axis=0)\n",
    "\n",
    "#                 pred_list_std = pd.DataFrame(pred_list_std, index=returns.columns).T\n",
    "#                 prediction_std = pd.concat([prediction_std, pred_list_std], axis=0)\n",
    "#                 elapsed_time = round(time.time() - start_time, 2)\n",
    "#                 print(f\"Test circle {i} trained in {elapsed_time} seconds.\")\n",
    "\n",
    "\n",
    "\n",
    "#             prediction_ret.index = binary.iloc[-test_size:, :].index\n",
    "#             prediction_std.index = returns.iloc[-test_size:, :].index\n",
    "\n",
    "#             elapsed_total_time = round(time.time() - total_time, 2)\n",
    "#             print(f\"Model trained in {elapsed_total_time} seconds.\")\n",
    "\n",
    "#             return prediction_ret, prediction_std\n",
    "        \n",
    "#         if val_model == 'RF':\n",
    "\n",
    "#             prediction_ret = pd.DataFrame(columns=binary.columns)\n",
    "\n",
    "#             total_time = time.time()\n",
    "\n",
    "#             for i in range(-test_size, 0):\n",
    "#                 start_time = time.time()\n",
    "#                 # Create rolling window to train the model\n",
    "#                 # 100 - window size, -1 correction for lag of tech features\n",
    "#                 X_train = tech.iloc[-window-1+i:i]\n",
    "#                 if features is None:\n",
    "#                     pass\n",
    "#                 else:\n",
    "#                     X_features = features.iloc[-window-1+i:i]\n",
    "#                     assert len(X_train) == len(X_features)\n",
    "                \n",
    "#                 y_train = returns.iloc[-window+i:i]\n",
    "\n",
    "#                 # RETURNS PART\n",
    "#                 #list for returns predictions\n",
    "#                 pred_list = np.array([])\n",
    "\n",
    "#                 for company in binary.columns:\n",
    "#                     # RETURNS PART\n",
    "#                     # Train a model\n",
    "#                     if features is None:\n",
    "#                         next_pred = model_validation.classify_next_bar(X_train[company], y_train[company], \n",
    "#                                                                    RF_params)\n",
    "#                     else:\n",
    "#                         next_pred = model_validation.classify_next_bar(X_train[company], y_train[company], \n",
    "#                                                                    RF_params, features=X_features)\n",
    "                    \n",
    "#                     # Collect results and append them\n",
    "#                     pred_list = np.append(pred_list, next_pred)\n",
    "\n",
    "#                 pred_list = pd.DataFrame(pred_list, index=binary.columns).T\n",
    "#                 prediction_ret = pd.concat([prediction_ret, pred_list], axis=0)\n",
    "\n",
    "#                 elapsed_time = round(time.time() - start_time, 2)\n",
    "#                 print(f\"Test circle {i} trained in {elapsed_time} seconds.\")\n",
    "\n",
    "#             prediction_ret.index = binary.iloc[-test_size:, :].index\n",
    "\n",
    "#             elapsed_total_time = round(time.time() - total_time, 2)\n",
    "#             print(f\"Model trained in {elapsed_total_time} seconds.\")\n",
    "\n",
    "#             return prediction_ret\n",
    "        \n",
    "        \n",
    "#         if val_model == 'GARCH':\n",
    "#             if summary_garch:\n",
    "#                 aic = {}\n",
    "            \n",
    "#             prediction_std = pd.DataFrame(columns=returns.columns)\n",
    "\n",
    "#             total_time = time.time()\n",
    "\n",
    "#             for i in range(-test_size, 0):\n",
    "#                 start_time = time.time()\n",
    "#                 # Create rolling window to train the model\n",
    "\n",
    "#                 y_train_ret = returns.iloc[-window+i:i]\n",
    "\n",
    "#                 #VOLATILITY PART\n",
    "#                 pred_list_std = np.array([])\n",
    "\n",
    "#                 for company in binary.columns:\n",
    "\n",
    "#                     #VOLATILITY PART\n",
    "#                     # Train a model\n",
    "#                     if summary_garch:\n",
    "#                         next_pred_std, aic[(i, company)] = model_validation.GARCH_model(y_train_ret[company], \n",
    "#                                                                      GARCH_params, \n",
    "#                                                                      summary=True)\n",
    "#                     else:\n",
    "#                         next_pred_std = model_validation.GARCH_model(y_train_ret[company], \n",
    "#                                                                      GARCH_params, \n",
    "#                                                                      summary=False)\n",
    "                    \n",
    "                    \n",
    "                    \n",
    "#                     # Collect results and append them\n",
    "#                     pred_list_std = np.append(pred_list_std, next_pred_std)\n",
    "                    \n",
    "#                 pred_list_std = pd.DataFrame(pred_list_std, index=returns.columns).T\n",
    "#                 prediction_std = pd.concat([prediction_std, pred_list_std], axis=0)\n",
    "#                 elapsed_time = round(time.time() - start_time, 2)\n",
    "#                 print(f\"Test circle {i} trained in {elapsed_time} seconds.\")\n",
    "\n",
    "#             prediction_std.index = returns.iloc[-test_size:, :].index\n",
    "\n",
    "#             elapsed_total_time = round(time.time() - total_time, 2)\n",
    "#             print(f\"Model trained in {elapsed_total_time} seconds.\")\n",
    "            \n",
    "#             if summary_garch:\n",
    "#                 return prediction_std, aic\n",
    "            \n",
    "#             return prediction_std            \n",
    "              \n",
    "        \n",
    "        \n",
    "#     def get_return_predictions(self, tech, binary, window, test_size, returns):\n",
    "#         direction, standard_dev = model_validation.rolling_window(tech, binary, window, test_size, returns)\n",
    "#         returns_prediction = direction * standard_dev\n",
    "\n",
    "#         return returns_prediction / 100, direction\n",
    "        \n",
    "        \n",
    "    \n",
    "    \n",
    "\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d3cab5ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "class model_validation:\n",
    "    \n",
    "    def __init__(self, tickers, start, end, features_tickers):\n",
    "        self.tickers = tickers\n",
    "        self.start = start\n",
    "        self.end = end\n",
    "\n",
    "        self.features = yf.download(features_tickers, start=self.start, end=self.end, \n",
    "                                    interval='1wk').fillna(method='ffill')\n",
    "        self.ohlc = yf.download(self.tickers, start=self.start, end=self.end, interval='1wk')\n",
    "    \n",
    "   \n",
    "    def tech_features(self):\n",
    "        inst = dw()\n",
    "        output = pd.DataFrame()\n",
    "        for i in self.ohlc.columns.levels[1]:\n",
    "\n",
    "            tech = inst.get_tech_features(close=self.ohlc['Adj Close'][i],\n",
    "                                              volume=self.ohlc['Volume'][i],\n",
    "                                              low=self.ohlc['Low'][i],\n",
    "                                              high=self.ohlc['High'][i],\n",
    "                                              concat=True).dropna()\n",
    "            tech.columns = pd.MultiIndex.from_product([[i], tech.columns])\n",
    "\n",
    "            output = pd.concat([output, tech], axis=1)\n",
    "            \n",
    "        self.ohlc = self.ohlc.loc[output.index]\n",
    "        self.features = self.features['Adj Close'].loc[output.index]\n",
    "        self.tech = output\n",
    "        print('Done!')\n",
    "        \n",
    "\n",
    "    def binary_classificator(self, x):\n",
    "        if x > 0:\n",
    "            return 1\n",
    "        else:\n",
    "            return -1\n",
    "        \n",
    "    \n",
    "    def binary_map(self, prices):\n",
    "        output = pd.DataFrame()\n",
    "        for i in prices.columns:\n",
    "            # binary classification of returns\n",
    "            \n",
    "            up_down = pd.DataFrame(prices[i].pct_change().dropna())\n",
    "            up_down[i] = up_down[i].map(self.binary_classificator)\n",
    "\n",
    "\n",
    "            # concatenate all companies in dataframe\n",
    "            output = pd.concat([output, up_down], axis=1)\n",
    "\n",
    "        return output\n",
    "    \n",
    "    \n",
    "    \n",
    "    def train_test_split(self, window, val_size, test=False, test_size=False):\n",
    "        self.window = window\n",
    "        self.val_size = val_size\n",
    "\n",
    "        \n",
    "        self.ohlc_val = self.ohlc[:self.window + self.val_size]\n",
    "        self.tech_val = self.tech[:self.window + self.val_size]\n",
    "        self.feat_val = self.features[:self.window + self.val_size]\n",
    "        self.binary_val = self.binary_map(prices=self.ohlc['Adj Close'])[:self.window + self.val_size - 1]\n",
    "        self.returns_val = 100 * self.ohlc['Adj Close'].pct_change().dropna()[:self.window + self.val_size - 1]\n",
    "\n",
    "        print('Initial training window is included')\n",
    "        print(f'ohlc val {len(self.ohlc_val)}, tech val {len(self.tech_val)}')\n",
    "        print(f'binary_val {len(self.binary_val)}, returns_val {len(self.returns_val)}')\n",
    "        assert len(self.ohlc_val) == len(self.tech_val)\n",
    "        assert len(self.binary_val) == len(self.returns_val)\n",
    "        assert len(self.ohlc_val) == len(self.binary_val) + 1\n",
    "        assert self.ohlc_val.index[1] == self.binary_val.index[0]\n",
    "        \n",
    "        assert self.feat_val.index[0] == self.tech_val.index[0]\n",
    "\n",
    "        if test:\n",
    "            \n",
    "            self.ohlc_test = self.ohlc[self.window + self.val_size:]\n",
    "            self.tech_test = self.tech[self.window + self.val_size:]\n",
    "            self.feat_test = self.features[self.window + self.val_size:]\n",
    "            self.binary_test = self.binary_map(self.ohlc['Adj Close'])[self.window + self.val_size :]\n",
    "            self.returns_test = 100 * self.ohlc['Adj Close'].pct_change().dropna()[self.window + self.val_size :]\n",
    "            self.test_size = len(self.ohlc_test) - self.window\n",
    "            \n",
    "            print('Initial training window is included')\n",
    "            print(f'ohlc test {len(self.ohlc_test)}, tech test {len(self.tech_test)}')\n",
    "            print(f'binary_test {len(self.binary_test)}, returns_test {len(self.returns_test)}')\n",
    "            \n",
    "            assert len(self.ohlc_test) == len(self.tech_test)\n",
    "            assert len(self.binary_test) == len(self.returns_test)\n",
    "            assert len(self.ohlc_test) == len(self.binary_test) + 1\n",
    "            assert self.ohlc_test.index[1] == self.binary_test.index[0]\n",
    "            assert self.feat_test.index[0] == self.tech_test.index[0]\n",
    "        \n",
    "        \n",
    "    def classify_next_bar(self, X, y, RF_params=(10, None, None), features=None):\n",
    "        \"\"\"\n",
    "        X, y - features and target-vector based on a shape of rolling window\n",
    "        X - tech features and external data for the date t-1\n",
    "        y - binary target - [1 - up, -1 - down] for time t\n",
    "        X_out - 1 day forward features to make prediction\n",
    "        Make assert statement to ensure that\n",
    "        \n",
    "        \"\"\"\n",
    "        assert X.shape[0] - 1 == y.shape[0]\n",
    "        assert len(X) == len(features)\n",
    "        X_data = pd.concat([X, features], axis=1)\n",
    "        X_train_data = X_data.shift().dropna()\n",
    "        \n",
    "        assert X_train_data.shape[0] == y.shape[0]\n",
    "        \n",
    "        X_forecast = pd.DataFrame(X_data.iloc[-1, :]).T\n",
    "        \n",
    "        # Fit the model - returns classifier\n",
    "        model = RandomForestClassifier(n_estimators=RF_params[0], \n",
    "                                       max_depth=RF_params[1], \n",
    "                                       max_leaf_nodes=RF_params[2], \n",
    "                                       random_state=42,\n",
    "                                       n_jobs=-1)\n",
    "        model.fit(X_train_data, y)\n",
    "        bar = model.predict(X_forecast)\n",
    "        \n",
    "        return bar\n",
    "\n",
    "    def regress_next_bar(self, X, y, RF_params=(10, None, None), features=None):\n",
    "        \"\"\"\n",
    "        X, y - features and target-vector based on a shape of rolling window\n",
    "        X - tech features and external data for the date t-1\n",
    "        y - binary target - [1 - up, -1 - down] for time t\n",
    "        X_out - 1 day forward features to make prediction\n",
    "        Make assert statement to ensure that\n",
    "        \n",
    "        \"\"\"\n",
    "        assert X.shape[0] - 1 == y.shape[0]\n",
    "        assert len(X) == len(features)\n",
    "        X_data = pd.concat([X, features], axis=1)\n",
    "        X_train_data = X_data.shift().dropna()\n",
    "        \n",
    "        assert X_train_data.shape[0] == y.shape[0]\n",
    "        \n",
    "        X_forecast = pd.DataFrame(X_data.iloc[-1, :]).T\n",
    "        \n",
    "        # Fit the model - returns classifier\n",
    "        model = RandomForestRegressor(n_estimators=RF_params[0], \n",
    "                                       max_depth=RF_params[1], \n",
    "                                       max_leaf_nodes=RF_params[2], \n",
    "                                       random_state=42,\n",
    "                                       n_jobs=-1)\n",
    "        model.fit(X_train_data, y)\n",
    "        bar = model.predict(X_forecast)\n",
    "        \n",
    "        return bar    \n",
    "    \n",
    "    \n",
    "    \n",
    "    \n",
    "    \n",
    "    def GARCH_model(self, returns, GARCH_params=(1, 1), summary=False):\n",
    "        assert type(returns) == pd.Series\n",
    "        model = arch_model(\n",
    "            returns,\n",
    "            p=GARCH_params[0],\n",
    "            q=GARCH_params[1],\n",
    "            rescale=False\n",
    "        )\n",
    "\n",
    "        model_res = model.fit(disp=0)\n",
    "        next_pred = model_res.forecast(horizon=1, reindex=False).variance.iloc[0,0] ** 0.5\n",
    "        \n",
    "        if summary:\n",
    "            aic = model_res.aic\n",
    "            return next_pred, aic\n",
    "\n",
    "        return next_pred\n",
    "    \n",
    "    \n",
    "    def expanding_window(self, tech, binary, window, test_size, returns,\n",
    "                        val_model='RF', RF_params=(10, None, None),\n",
    "                        GARCH_params=(1, 1), summary_garch=False, features=None):\n",
    "        \n",
    "        if val_model == 'RF':\n",
    "\n",
    "            prediction_ret = pd.DataFrame(columns=binary.columns)\n",
    "\n",
    "            total_time = time.time()\n",
    "\n",
    "            for i in range(-test_size, 0):\n",
    "                start_time = time.time()\n",
    "                # Create rolling window to train the model\n",
    "                # 100 - window size, -1 correction for lag of tech features\n",
    "                X_train = tech.iloc[-window-1+i:i]\n",
    "                if features is None:\n",
    "                    pass\n",
    "                else:\n",
    "                    X_features = features.iloc[-window-1+i:i]\n",
    "                    assert len(X_train) == len(X_features)\n",
    "                \n",
    "                y_train = returns.iloc[-window+i:i]\n",
    "\n",
    "                # RETURNS PART\n",
    "                #list for returns predictions\n",
    "                pred_list = np.array([])\n",
    "\n",
    "                for company in binary.columns:\n",
    "                    # RETURNS PART\n",
    "                    # Train a model\n",
    "                    if features is None:\n",
    "                        next_pred = self.regress_next_bar(X_train[company], y_train[company], \n",
    "                                                                   RF_params)\n",
    "                    else:\n",
    "                        next_pred = self.regress_next_bar(X_train[company], y_train[company], \n",
    "                                                                   RF_params, features=X_features)\n",
    "                    \n",
    "                    # Collect results and append them\n",
    "                    pred_list = np.append(pred_list, next_pred)\n",
    "\n",
    "                pred_list = pd.DataFrame(pred_list, index=binary.columns).T\n",
    "                prediction_ret = pd.concat([prediction_ret, pred_list], axis=0)\n",
    "\n",
    "                elapsed_time = round(time.time() - start_time, 2)\n",
    "                print(f\"Test circle {i} trained in {elapsed_time} seconds.\")\n",
    "\n",
    "            prediction_ret.index = binary.iloc[-test_size:, :].index\n",
    "\n",
    "            elapsed_total_time = round(time.time() - total_time, 2)\n",
    "            print(f\"Model trained in {elapsed_total_time} seconds.\")\n",
    "\n",
    "            return prediction_ret        \n",
    "    \n",
    "    \n",
    "    \n",
    "    def rolling_window(self, tech, binary, window, test_size, returns, \n",
    "                       val_model='Both', RF_params=(10, None, None), \n",
    "                       GARCH_params=(1, 1), summary_garch=False, features=None):\n",
    "        \"\"\"\n",
    "        val_model: what model to validate. Relevant parameters: 'RF', 'GARCH', 'Both'\n",
    "        \n",
    "        \"\"\"\n",
    "        if val_model == 'Both':\n",
    "\n",
    "\n",
    "            prediction_ret = pd.DataFrame(columns=binary.columns)\n",
    "            prediction_std = pd.DataFrame(columns=returns.columns)\n",
    "\n",
    "            total_time = time.time()\n",
    "\n",
    "            for i in range(-test_size, 0):\n",
    "                start_time = time.time()\n",
    "                # Create rolling window to train the model\n",
    "                # 100 - window size, -1 correction for lag of tech features\n",
    "                X_train = tech.iloc[-window-1+i:i]\n",
    "                y_train = binary.iloc[-window+i:i]\n",
    "\n",
    "                y_train_ret = returns.iloc[-window+i:i]\n",
    "\n",
    "                # RETURNS PART\n",
    "                #list for returns predictions\n",
    "                pred_list = np.array([])\n",
    "\n",
    "                #VOLATILITY PART\n",
    "                pred_list_std = np.array([])\n",
    "\n",
    "                for company in binary.columns:\n",
    "                    # RETURNS PART\n",
    "                    # Train a model\n",
    "                    #print(f' Start {company}')\n",
    "\n",
    "                    next_pred = self.classify_next_bar(X_train[company], y_train[company], RF_params)\n",
    "                    # Collect results and append them\n",
    "                    pred_list = np.append(pred_list, next_pred)\n",
    "\n",
    "\n",
    "                    #VOLATILITY PART\n",
    "                    # Train a model\n",
    "                    next_pred_std = self.GARCH_model(y_train_ret[company], GARCH_params, summary=summary_garch)\n",
    "                    # Collect results and append them\n",
    "                    pred_list_std = np.append(pred_list_std, next_pred_std)\n",
    "\n",
    "                    #print(f' End {company}')\n",
    "\n",
    "                pred_list = pd.DataFrame(pred_list, index=binary.columns).T\n",
    "                prediction_ret = pd.concat([prediction_ret, pred_list], axis=0)\n",
    "\n",
    "                pred_list_std = pd.DataFrame(pred_list_std, index=returns.columns).T\n",
    "                prediction_std = pd.concat([prediction_std, pred_list_std], axis=0)\n",
    "                elapsed_time = round(time.time() - start_time, 2)\n",
    "                print(f\"Test circle {i} trained in {elapsed_time} seconds.\")\n",
    "\n",
    "\n",
    "\n",
    "            prediction_ret.index = binary.iloc[-test_size:, :].index\n",
    "            prediction_std.index = returns.iloc[-test_size:, :].index\n",
    "\n",
    "            elapsed_total_time = round(time.time() - total_time, 2)\n",
    "            print(f\"Model trained in {elapsed_total_time} seconds.\")\n",
    "\n",
    "            return prediction_ret, prediction_std\n",
    "        \n",
    "        if val_model == 'RF':\n",
    "\n",
    "            prediction_ret = pd.DataFrame(columns=binary.columns)\n",
    "\n",
    "            total_time = time.time()\n",
    "\n",
    "            for i in range(-test_size, 0):\n",
    "                start_time = time.time()\n",
    "                # Create rolling window to train the model\n",
    "                # 100 - window size, -1 correction for lag of tech features\n",
    "                X_train = tech.iloc[-window-1+i:i]\n",
    "                if features is None:\n",
    "                    pass\n",
    "                else:\n",
    "                    X_features = features.iloc[-window-1+i:i]\n",
    "                    assert len(X_train) == len(X_features)\n",
    "                \n",
    "                y_train = binary.iloc[-window+i:i]\n",
    "\n",
    "                # RETURNS PART\n",
    "                #list for returns predictions\n",
    "                pred_list = np.array([])\n",
    "\n",
    "                for company in binary.columns:\n",
    "                    # RETURNS PART\n",
    "                    # Train a model\n",
    "                    if features is None:\n",
    "                        next_pred = self.classify_next_bar(X_train[company], y_train[company], \n",
    "                                                                   RF_params)\n",
    "                    else:\n",
    "                        next_pred = self.classify_next_bar(X_train[company], y_train[company], \n",
    "                                                                   RF_params, features=X_features)\n",
    "                    \n",
    "                    # Collect results and append them\n",
    "                    pred_list = np.append(pred_list, next_pred)\n",
    "\n",
    "                pred_list = pd.DataFrame(pred_list, index=binary.columns).T\n",
    "                prediction_ret = pd.concat([prediction_ret, pred_list], axis=0)\n",
    "\n",
    "                elapsed_time = round(time.time() - start_time, 2)\n",
    "                print(f\"Test circle {i} trained in {elapsed_time} seconds.\")\n",
    "\n",
    "            prediction_ret.index = binary.iloc[-test_size:, :].index\n",
    "\n",
    "            elapsed_total_time = round(time.time() - total_time, 2)\n",
    "            print(f\"Model trained in {elapsed_total_time} seconds.\")\n",
    "\n",
    "            return prediction_ret\n",
    "        \n",
    "        \n",
    "        if val_model == 'GARCH':\n",
    "            if summary_garch:\n",
    "                aic = {}\n",
    "            \n",
    "            prediction_std = pd.DataFrame(columns=returns.columns)\n",
    "\n",
    "            total_time = time.time()\n",
    "\n",
    "            for i in range(-test_size, 0):\n",
    "                start_time = time.time()\n",
    "                # Create rolling window to train the model\n",
    "\n",
    "                y_train_ret = returns.iloc[-window+i:i]\n",
    "\n",
    "                #VOLATILITY PART\n",
    "                pred_list_std = np.array([])\n",
    "\n",
    "                for company in binary.columns:\n",
    "\n",
    "                    #VOLATILITY PART\n",
    "                    # Train a model\n",
    "                    if summary_garch:\n",
    "                        next_pred_std, aic[(i, company)] = self.GARCH_model(y_train_ret[company], \n",
    "                                                                     GARCH_params, \n",
    "                                                                     summary=True)\n",
    "                    else:\n",
    "                        next_pred_std = self.GARCH_model(y_train_ret[company], \n",
    "                                                                     GARCH_params, \n",
    "                                                                     summary=False)\n",
    "                    \n",
    "                    \n",
    "                    \n",
    "                    # Collect results and append them\n",
    "                    pred_list_std = np.append(pred_list_std, next_pred_std)\n",
    "                    \n",
    "                pred_list_std = pd.DataFrame(pred_list_std, index=returns.columns).T\n",
    "                prediction_std = pd.concat([prediction_std, pred_list_std], axis=0)\n",
    "                elapsed_time = round(time.time() - start_time, 2)\n",
    "                print(f\"Test circle {i} trained in {elapsed_time} seconds.\")\n",
    "\n",
    "            prediction_std.index = returns.iloc[-test_size:, :].index\n",
    "\n",
    "            elapsed_total_time = round(time.time() - total_time, 2)\n",
    "            print(f\"Model trained in {elapsed_total_time} seconds.\")\n",
    "            \n",
    "            if summary_garch:\n",
    "                return prediction_std, aic\n",
    "            \n",
    "            return prediction_std            \n",
    "              \n",
    "        \n",
    "        \n",
    "    def get_return_predictions(self, tech, binary, window, test_size, returns):\n",
    "        direction, standard_dev = self.rolling_window(tech, binary, window, test_size, returns)\n",
    "        returns_prediction = direction * standard_dev\n",
    "\n",
    "        return returns_prediction / 100, direction\n",
    "        \n",
    "        \n",
    "    \n",
    "    \n",
    "\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ddd3e7ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Validation\n",
    "tickers = ['AMZN', 'TM', 'HD', 'GOOGL', 'VZ', 'CMCSA',\n",
    "          'WMT', 'PG', 'KO', 'XOM', 'CVX', 'SHEL',\n",
    "          'SPGI', 'WFC', 'JPM', 'UPS', 'RTX', 'HON',\n",
    "          'UNH', 'LLY', 'JNJ', 'BHP', 'LIN', 'RIO',\n",
    "          'PLD', 'AMT', 'EQIX', 'TSM', 'ORCL', 'CSCO',\n",
    "          'NEE', 'SO', 'DUK']\n",
    "\n",
    "\n",
    "# Test set\n",
    "# tickers = ['MCD', 'NKE', 'SBUX', 'DIS', 'T', 'NFLX',\n",
    "#           'PEP', 'COST', 'UL', 'TTE', 'BP', 'EQNR',\n",
    "#           'BAC', 'HSBC', 'MS', 'BA', 'LMT', 'GE',\n",
    "#           'AZN', 'MRK', 'PFE', 'APD', 'VALE', 'SHW',\n",
    "#           'PSA', 'CCI', 'O', 'AAPL', 'MSFT', 'NVDA',\n",
    "#           'PCG', 'SRE', 'AEP']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9f740cb0",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "features = [\"EURUSD=X\", '^IRX', '^FVX', 'CL=F', 'GC=F', 'NG=F', 'RB=F']\n",
    "start = '2005-01-01'\n",
    "end = '2022-12-31'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "91e44d7d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[*********************100%***********************]  7 of 7 completed\n",
      "[*********************100%***********************]  33 of 33 completed\n"
     ]
    }
   ],
   "source": [
    "model_validation = model_validation(tickers, start, end, features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "ae95360a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "939\n"
     ]
    }
   ],
   "source": [
    "print(len(model_validation.ohlc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "450b530b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done!\n"
     ]
    }
   ],
   "source": [
    "model_validation.tech_features()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "bec0f1d1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "181\n",
      "725\n"
     ]
    }
   ],
   "source": [
    "window = int(len(model_validation.tech) * 0.2)\n",
    "val_size = len(model_validation.tech) - window\n",
    "print(window)\n",
    "print(val_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "aee93393",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Initial training window is included\n",
      "ohlc val 906, tech val 906\n",
      "binary_val 905, returns_val 905\n"
     ]
    }
   ],
   "source": [
    "model_validation.train_test_split(window=window, val_size=val_size, test=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aaa048c3",
   "metadata": {},
   "source": [
    "# Random Forest Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b1e5b7e2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Random forest params\n",
    "RF_params = []\n",
    "param_grid_random_forest = {\n",
    "    'n_estimators': [25, 50],\n",
    "    'max_depth': [3, 6],\n",
    "    'max_leaf_nodes': [3, 6],\n",
    "}\n",
    "# param_grid_random_forest = {\n",
    "#     'n_estimators': [50],\n",
    "#     'max_depth': [3],\n",
    "#     'max_leaf_nodes': [None]\n",
    "# }\n",
    "\n",
    "for i in itertools.product(param_grid_random_forest['n_estimators'],\n",
    "                           param_grid_random_forest['max_depth'],\n",
    "                           param_grid_random_forest['max_leaf_nodes']):\n",
    "    \n",
    "    RF_params.append(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "612d1a7d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "=========================================================================================\n",
      "Model (25, 3, 3) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 1.51 seconds.\n",
      "Test circle -724 trained in 1.54 seconds.\n",
      "Test circle -723 trained in 1.55 seconds.\n",
      "Test circle -722 trained in 1.53 seconds.\n",
      "Test circle -721 trained in 1.59 seconds.\n",
      "Test circle -720 trained in 1.54 seconds.\n",
      "Test circle -719 trained in 1.57 seconds.\n",
      "Test circle -718 trained in 1.65 seconds.\n",
      "Test circle -717 trained in 1.57 seconds.\n",
      "Test circle -716 trained in 1.54 seconds.\n",
      "Test circle -715 trained in 1.57 seconds.\n",
      "Test circle -714 trained in 1.73 seconds.\n",
      "Test circle -713 trained in 1.64 seconds.\n",
      "Test circle -712 trained in 1.63 seconds.\n",
      "Test circle -711 trained in 1.63 seconds.\n",
      "Test circle -710 trained in 1.63 seconds.\n",
      "Test circle -709 trained in 1.54 seconds.\n",
      "Test circle -708 trained in 1.59 seconds.\n",
      "Test circle -707 trained in 1.59 seconds.\n",
      "Test circle -706 trained in 1.54 seconds.\n",
      "Test circle -705 trained in 1.55 seconds.\n",
      "Test circle -704 trained in 1.56 seconds.\n",
      "Test circle -703 trained in 1.61 seconds.\n",
      "Test circle -702 trained in 1.59 seconds.\n",
      "Test circle -701 trained in 1.57 seconds.\n",
      "Test circle -700 trained in 1.54 seconds.\n",
      "Test circle -699 trained in 1.57 seconds.\n",
      "Test circle -698 trained in 1.58 seconds.\n",
      "Test circle -697 trained in 1.61 seconds.\n",
      "Test circle -696 trained in 1.6 seconds.\n",
      "Test circle -695 trained in 1.57 seconds.\n",
      "Test circle -694 trained in 1.64 seconds.\n",
      "Test circle -693 trained in 1.59 seconds.\n",
      "Test circle -692 trained in 1.62 seconds.\n",
      "Test circle -691 trained in 1.69 seconds.\n",
      "Test circle -690 trained in 1.56 seconds.\n",
      "Test circle -689 trained in 1.67 seconds.\n",
      "Test circle -688 trained in 1.59 seconds.\n",
      "Test circle -687 trained in 1.63 seconds.\n",
      "Test circle -686 trained in 1.67 seconds.\n",
      "Test circle -685 trained in 1.65 seconds.\n",
      "Test circle -684 trained in 1.56 seconds.\n",
      "Test circle -683 trained in 1.68 seconds.\n",
      "Test circle -682 trained in 1.65 seconds.\n",
      "Test circle -681 trained in 1.59 seconds.\n",
      "Test circle -680 trained in 1.57 seconds.\n",
      "Test circle -679 trained in 1.56 seconds.\n",
      "Test circle -678 trained in 1.56 seconds.\n",
      "Test circle -677 trained in 1.59 seconds.\n",
      "Test circle -676 trained in 1.61 seconds.\n",
      "Test circle -675 trained in 1.57 seconds.\n",
      "Test circle -674 trained in 1.66 seconds.\n",
      "Test circle -673 trained in 1.58 seconds.\n",
      "Test circle -672 trained in 1.63 seconds.\n",
      "Test circle -671 trained in 1.6 seconds.\n",
      "Test circle -670 trained in 1.57 seconds.\n",
      "Test circle -669 trained in 1.57 seconds.\n",
      "Test circle -668 trained in 1.56 seconds.\n",
      "Test circle -667 trained in 1.6 seconds.\n",
      "Test circle -666 trained in 1.6 seconds.\n",
      "Test circle -665 trained in 1.63 seconds.\n",
      "Test circle -664 trained in 1.56 seconds.\n",
      "Test circle -663 trained in 1.56 seconds.\n",
      "Test circle -662 trained in 1.6 seconds.\n",
      "Test circle -661 trained in 1.66 seconds.\n",
      "Test circle -660 trained in 1.63 seconds.\n",
      "Test circle -659 trained in 1.57 seconds.\n",
      "Test circle -658 trained in 1.56 seconds.\n",
      "Test circle -657 trained in 1.56 seconds.\n",
      "Test circle -656 trained in 1.59 seconds.\n",
      "Test circle -655 trained in 1.62 seconds.\n",
      "Test circle -654 trained in 1.57 seconds.\n",
      "Test circle -653 trained in 1.65 seconds.\n",
      "Test circle -652 trained in 1.56 seconds.\n",
      "Test circle -651 trained in 1.63 seconds.\n",
      "Test circle -650 trained in 1.6 seconds.\n",
      "Test circle -649 trained in 1.57 seconds.\n",
      "Test circle -648 trained in 1.57 seconds.\n",
      "Test circle -647 trained in 1.57 seconds.\n",
      "Test circle -646 trained in 1.66 seconds.\n",
      "Test circle -645 trained in 1.63 seconds.\n",
      "Test circle -644 trained in 1.57 seconds.\n",
      "Test circle -643 trained in 1.56 seconds.\n",
      "Test circle -642 trained in 1.57 seconds.\n",
      "Test circle -641 trained in 1.58 seconds.\n",
      "Test circle -640 trained in 1.6 seconds.\n",
      "Test circle -639 trained in 1.64 seconds.\n",
      "Test circle -638 trained in 1.56 seconds.\n",
      "Test circle -637 trained in 1.56 seconds.\n",
      "Test circle -636 trained in 1.57 seconds.\n",
      "Test circle -635 trained in 1.63 seconds.\n",
      "Test circle -634 trained in 1.69 seconds.\n",
      "Test circle -633 trained in 1.57 seconds.\n",
      "Test circle -632 trained in 1.57 seconds.\n",
      "Test circle -631 trained in 1.57 seconds.\n",
      "Test circle -630 trained in 1.6 seconds.\n",
      "Test circle -629 trained in 1.63 seconds.\n",
      "Test circle -628 trained in 1.57 seconds.\n",
      "Test circle -627 trained in 1.57 seconds.\n",
      "Test circle -626 trained in 1.57 seconds.\n",
      "Test circle -625 trained in 1.61 seconds.\n",
      "Test circle -624 trained in 1.61 seconds.\n",
      "Test circle -623 trained in 1.58 seconds.\n",
      "Test circle -622 trained in 1.57 seconds.\n",
      "Test circle -621 trained in 1.56 seconds.\n",
      "Test circle -620 trained in 1.58 seconds.\n",
      "Test circle -619 trained in 1.63 seconds.\n",
      "Test circle -618 trained in 1.61 seconds.\n",
      "Test circle -617 trained in 1.57 seconds.\n",
      "Test circle -616 trained in 1.57 seconds.\n",
      "Test circle -615 trained in 1.59 seconds.\n",
      "Test circle -614 trained in 1.66 seconds.\n",
      "Test circle -613 trained in 1.75 seconds.\n",
      "Test circle -612 trained in 1.6 seconds.\n",
      "Test circle -611 trained in 1.61 seconds.\n",
      "Test circle -610 trained in 1.6 seconds.\n",
      "Test circle -609 trained in 1.67 seconds.\n",
      "Test circle -608 trained in 1.72 seconds.\n",
      "Test circle -607 trained in 1.62 seconds.\n",
      "Test circle -606 trained in 1.62 seconds.\n",
      "Test circle -605 trained in 1.64 seconds.\n",
      "Test circle -604 trained in 1.7 seconds.\n",
      "Test circle -603 trained in 1.69 seconds.\n",
      "Test circle -602 trained in 1.65 seconds.\n",
      "Test circle -601 trained in 1.64 seconds.\n",
      "Test circle -600 trained in 1.65 seconds.\n",
      "Test circle -599 trained in 1.71 seconds.\n",
      "Test circle -598 trained in 1.72 seconds.\n",
      "Test circle -597 trained in 1.66 seconds.\n",
      "Test circle -596 trained in 1.66 seconds.\n",
      "Test circle -595 trained in 1.66 seconds.\n",
      "Test circle -594 trained in 1.73 seconds.\n",
      "Test circle -593 trained in 1.81 seconds.\n",
      "Test circle -592 trained in 1.69 seconds.\n",
      "Test circle -591 trained in 1.7 seconds.\n",
      "Test circle -590 trained in 1.7 seconds.\n",
      "Test circle -589 trained in 1.79 seconds.\n",
      "Test circle -588 trained in 1.81 seconds.\n",
      "Test circle -587 trained in 1.79 seconds.\n",
      "Test circle -586 trained in 1.82 seconds.\n",
      "Test circle -585 trained in 1.77 seconds.\n",
      "Test circle -584 trained in 1.8 seconds.\n",
      "Test circle -583 trained in 1.74 seconds.\n",
      "Test circle -582 trained in 1.75 seconds.\n",
      "Test circle -581 trained in 1.75 seconds.\n",
      "Test circle -580 trained in 1.8 seconds.\n",
      "Test circle -579 trained in 1.84 seconds.\n",
      "Test circle -578 trained in 1.77 seconds.\n",
      "Test circle -577 trained in 1.78 seconds.\n",
      "Test circle -576 trained in 1.75 seconds.\n",
      "Test circle -575 trained in 1.89 seconds.\n",
      "Test circle -574 trained in 1.84 seconds.\n",
      "Test circle -573 trained in 2.19 seconds.\n",
      "Test circle -572 trained in 2.02 seconds.\n",
      "Test circle -571 trained in 1.92 seconds.\n",
      "Test circle -570 trained in 1.71 seconds.\n",
      "Test circle -569 trained in 1.51 seconds.\n",
      "Test circle -568 trained in 1.52 seconds.\n",
      "Test circle -567 trained in 1.52 seconds.\n",
      "Test circle -566 trained in 1.56 seconds.\n",
      "Test circle -565 trained in 1.56 seconds.\n",
      "Test circle -564 trained in 1.52 seconds.\n",
      "Test circle -563 trained in 1.53 seconds.\n",
      "Test circle -562 trained in 1.53 seconds.\n",
      "Test circle -561 trained in 1.6 seconds.\n",
      "Test circle -560 trained in 1.56 seconds.\n",
      "Test circle -559 trained in 1.56 seconds.\n",
      "Test circle -558 trained in 1.6 seconds.\n",
      "Test circle -557 trained in 1.54 seconds.\n",
      "Test circle -556 trained in 1.6 seconds.\n",
      "Test circle -555 trained in 1.58 seconds.\n",
      "Test circle -554 trained in 1.64 seconds.\n",
      "Test circle -553 trained in 1.55 seconds.\n",
      "Test circle -552 trained in 1.57 seconds.\n",
      "Test circle -551 trained in 1.58 seconds.\n",
      "Test circle -550 trained in 1.62 seconds.\n",
      "Test circle -549 trained in 1.6 seconds.\n",
      "Test circle -548 trained in 1.55 seconds.\n",
      "Test circle -547 trained in 1.56 seconds.\n",
      "Test circle -546 trained in 1.55 seconds.\n",
      "Test circle -545 trained in 1.6 seconds.\n",
      "Test circle -544 trained in 1.57 seconds.\n",
      "Test circle -543 trained in 1.56 seconds.\n",
      "Test circle -542 trained in 1.55 seconds.\n",
      "Test circle -541 trained in 1.56 seconds.\n",
      "Test circle -540 trained in 1.6 seconds.\n",
      "Test circle -539 trained in 1.59 seconds.\n",
      "Test circle -538 trained in 1.56 seconds.\n",
      "Test circle -537 trained in 1.57 seconds.\n",
      "Test circle -536 trained in 1.61 seconds.\n",
      "Test circle -535 trained in 1.63 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -534 trained in 1.67 seconds.\n",
      "Test circle -533 trained in 1.67 seconds.\n",
      "Test circle -532 trained in 1.56 seconds.\n",
      "Test circle -531 trained in 1.55 seconds.\n",
      "Test circle -530 trained in 1.58 seconds.\n",
      "Test circle -529 trained in 1.6 seconds.\n",
      "Test circle -528 trained in 1.62 seconds.\n",
      "Test circle -527 trained in 1.56 seconds.\n",
      "Test circle -526 trained in 1.58 seconds.\n",
      "Test circle -525 trained in 1.58 seconds.\n",
      "Test circle -524 trained in 1.64 seconds.\n",
      "Test circle -523 trained in 1.59 seconds.\n",
      "Test circle -522 trained in 1.56 seconds.\n",
      "Test circle -521 trained in 1.58 seconds.\n",
      "Test circle -520 trained in 1.56 seconds.\n",
      "Test circle -519 trained in 1.61 seconds.\n",
      "Test circle -518 trained in 1.61 seconds.\n",
      "Test circle -517 trained in 1.56 seconds.\n",
      "Test circle -516 trained in 1.57 seconds.\n",
      "Test circle -515 trained in 1.61 seconds.\n",
      "Test circle -514 trained in 1.73 seconds.\n",
      "Test circle -513 trained in 1.58 seconds.\n",
      "Test circle -512 trained in 1.57 seconds.\n",
      "Test circle -511 trained in 1.56 seconds.\n",
      "Test circle -510 trained in 1.53 seconds.\n",
      "Test circle -509 trained in 1.55 seconds.\n",
      "Test circle -508 trained in 1.59 seconds.\n",
      "Test circle -507 trained in 1.55 seconds.\n",
      "Test circle -506 trained in 1.67 seconds.\n",
      "Test circle -505 trained in 1.71 seconds.\n",
      "Test circle -504 trained in 1.61 seconds.\n",
      "Test circle -503 trained in 1.56 seconds.\n",
      "Test circle -502 trained in 1.55 seconds.\n",
      "Test circle -501 trained in 1.55 seconds.\n",
      "Test circle -500 trained in 1.55 seconds.\n",
      "Test circle -499 trained in 1.56 seconds.\n",
      "Test circle -498 trained in 1.61 seconds.\n",
      "Test circle -497 trained in 1.56 seconds.\n",
      "Test circle -496 trained in 1.58 seconds.\n",
      "Test circle -495 trained in 1.58 seconds.\n",
      "Test circle -494 trained in 1.66 seconds.\n",
      "Test circle -493 trained in 1.58 seconds.\n",
      "Test circle -492 trained in 1.58 seconds.\n",
      "Test circle -491 trained in 1.58 seconds.\n",
      "Test circle -490 trained in 1.58 seconds.\n",
      "Test circle -489 trained in 1.61 seconds.\n",
      "Test circle -488 trained in 1.6 seconds.\n",
      "Test circle -487 trained in 1.6 seconds.\n",
      "Test circle -486 trained in 1.61 seconds.\n",
      "Test circle -485 trained in 1.61 seconds.\n",
      "Test circle -484 trained in 1.96 seconds.\n",
      "Test circle -483 trained in 1.61 seconds.\n",
      "Test circle -482 trained in 1.61 seconds.\n",
      "Test circle -481 trained in 1.62 seconds.\n",
      "Test circle -480 trained in 1.63 seconds.\n",
      "Test circle -479 trained in 1.74 seconds.\n",
      "Test circle -478 trained in 1.63 seconds.\n",
      "Test circle -477 trained in 1.64 seconds.\n",
      "Test circle -476 trained in 1.66 seconds.\n",
      "Test circle -475 trained in 1.62 seconds.\n",
      "Test circle -474 trained in 1.7 seconds.\n",
      "Test circle -473 trained in 1.61 seconds.\n",
      "Test circle -472 trained in 1.62 seconds.\n",
      "Test circle -471 trained in 1.61 seconds.\n",
      "Test circle -470 trained in 1.62 seconds.\n",
      "Test circle -469 trained in 1.61 seconds.\n",
      "Test circle -468 trained in 1.61 seconds.\n",
      "Test circle -467 trained in 1.61 seconds.\n",
      "Test circle -466 trained in 1.64 seconds.\n",
      "Test circle -465 trained in 1.61 seconds.\n",
      "Test circle -464 trained in 1.62 seconds.\n",
      "Test circle -463 trained in 1.61 seconds.\n",
      "Test circle -462 trained in 1.63 seconds.\n",
      "Test circle -461 trained in 1.62 seconds.\n",
      "Test circle -460 trained in 1.6 seconds.\n",
      "Test circle -459 trained in 1.61 seconds.\n",
      "Test circle -458 trained in 1.61 seconds.\n",
      "Test circle -457 trained in 1.63 seconds.\n",
      "Test circle -456 trained in 1.61 seconds.\n",
      "Test circle -455 trained in 1.61 seconds.\n",
      "Test circle -454 trained in 1.69 seconds.\n",
      "Test circle -453 trained in 1.6 seconds.\n",
      "Test circle -452 trained in 1.62 seconds.\n",
      "Test circle -451 trained in 1.6 seconds.\n",
      "Test circle -450 trained in 1.61 seconds.\n",
      "Test circle -449 trained in 1.61 seconds.\n",
      "Test circle -448 trained in 1.61 seconds.\n",
      "Test circle -447 trained in 1.61 seconds.\n",
      "Test circle -446 trained in 1.61 seconds.\n",
      "Test circle -445 trained in 1.62 seconds.\n",
      "Test circle -444 trained in 1.62 seconds.\n",
      "Test circle -443 trained in 1.6 seconds.\n",
      "Test circle -442 trained in 1.6 seconds.\n",
      "Test circle -441 trained in 1.59 seconds.\n",
      "Test circle -440 trained in 1.7 seconds.\n",
      "Test circle -439 trained in 1.73 seconds.\n",
      "Test circle -438 trained in 1.71 seconds.\n",
      "Test circle -437 trained in 1.61 seconds.\n",
      "Test circle -436 trained in 1.61 seconds.\n",
      "Test circle -435 trained in 1.61 seconds.\n",
      "Test circle -434 trained in 1.69 seconds.\n",
      "Test circle -433 trained in 1.61 seconds.\n",
      "Test circle -432 trained in 1.61 seconds.\n",
      "Test circle -431 trained in 1.61 seconds.\n",
      "Test circle -430 trained in 1.61 seconds.\n",
      "Test circle -429 trained in 1.61 seconds.\n",
      "Test circle -428 trained in 1.61 seconds.\n",
      "Test circle -427 trained in 1.61 seconds.\n",
      "Test circle -426 trained in 1.61 seconds.\n",
      "Test circle -425 trained in 1.64 seconds.\n",
      "Test circle -424 trained in 1.63 seconds.\n",
      "Test circle -423 trained in 1.61 seconds.\n",
      "Test circle -422 trained in 1.61 seconds.\n",
      "Test circle -421 trained in 1.62 seconds.\n",
      "Test circle -420 trained in 1.6 seconds.\n",
      "Test circle -419 trained in 1.6 seconds.\n",
      "Test circle -418 trained in 1.61 seconds.\n",
      "Test circle -417 trained in 1.6 seconds.\n",
      "Test circle -416 trained in 1.6 seconds.\n",
      "Test circle -415 trained in 1.69 seconds.\n",
      "Test circle -414 trained in 1.61 seconds.\n",
      "Test circle -413 trained in 1.61 seconds.\n",
      "Test circle -412 trained in 1.61 seconds.\n",
      "Test circle -411 trained in 1.61 seconds.\n",
      "Test circle -410 trained in 1.6 seconds.\n",
      "Test circle -409 trained in 1.61 seconds.\n",
      "Test circle -408 trained in 1.61 seconds.\n",
      "Test circle -407 trained in 1.62 seconds.\n",
      "Test circle -406 trained in 1.6 seconds.\n",
      "Test circle -405 trained in 1.6 seconds.\n",
      "Test circle -404 trained in 1.6 seconds.\n",
      "Test circle -403 trained in 1.66 seconds.\n",
      "Test circle -402 trained in 1.62 seconds.\n",
      "Test circle -401 trained in 1.6 seconds.\n",
      "Test circle -400 trained in 1.6 seconds.\n",
      "Test circle -399 trained in 1.61 seconds.\n",
      "Test circle -398 trained in 1.61 seconds.\n",
      "Test circle -397 trained in 1.61 seconds.\n",
      "Test circle -396 trained in 1.69 seconds.\n",
      "Test circle -395 trained in 1.6 seconds.\n",
      "Test circle -394 trained in 1.6 seconds.\n",
      "Test circle -393 trained in 1.61 seconds.\n",
      "Test circle -392 trained in 1.61 seconds.\n",
      "Test circle -391 trained in 1.61 seconds.\n",
      "Test circle -390 trained in 1.61 seconds.\n",
      "Test circle -389 trained in 1.61 seconds.\n",
      "Test circle -388 trained in 1.6 seconds.\n",
      "Test circle -387 trained in 1.65 seconds.\n",
      "Test circle -386 trained in 1.6 seconds.\n",
      "Test circle -385 trained in 1.61 seconds.\n",
      "Test circle -384 trained in 1.59 seconds.\n",
      "Test circle -383 trained in 1.6 seconds.\n",
      "Test circle -382 trained in 1.61 seconds.\n",
      "Test circle -381 trained in 1.6 seconds.\n",
      "Test circle -380 trained in 1.61 seconds.\n",
      "Test circle -379 trained in 1.6 seconds.\n",
      "Test circle -378 trained in 1.6 seconds.\n",
      "Test circle -377 trained in 1.69 seconds.\n",
      "Test circle -376 trained in 1.6 seconds.\n",
      "Test circle -375 trained in 1.61 seconds.\n",
      "Test circle -374 trained in 1.6 seconds.\n",
      "Test circle -373 trained in 1.6 seconds.\n",
      "Test circle -372 trained in 1.6 seconds.\n",
      "Test circle -371 trained in 1.91 seconds.\n",
      "Test circle -370 trained in 1.66 seconds.\n",
      "Test circle -369 trained in 1.6 seconds.\n",
      "Test circle -368 trained in 1.62 seconds.\n",
      "Test circle -367 trained in 1.6 seconds.\n",
      "Test circle -366 trained in 1.69 seconds.\n",
      "Test circle -365 trained in 1.65 seconds.\n",
      "Test circle -364 trained in 1.62 seconds.\n",
      "Test circle -363 trained in 1.61 seconds.\n",
      "Test circle -362 trained in 1.59 seconds.\n",
      "Test circle -361 trained in 1.62 seconds.\n",
      "Test circle -360 trained in 1.61 seconds.\n",
      "Test circle -359 trained in 1.61 seconds.\n",
      "Test circle -358 trained in 1.62 seconds.\n",
      "Test circle -357 trained in 1.6 seconds.\n",
      "Test circle -356 trained in 1.68 seconds.\n",
      "Test circle -355 trained in 1.71 seconds.\n",
      "Test circle -354 trained in 1.6 seconds.\n",
      "Test circle -353 trained in 1.61 seconds.\n",
      "Test circle -352 trained in 1.62 seconds.\n",
      "Test circle -351 trained in 1.63 seconds.\n",
      "Test circle -350 trained in 1.63 seconds.\n",
      "Test circle -349 trained in 1.6 seconds.\n",
      "Test circle -348 trained in 1.62 seconds.\n",
      "Test circle -347 trained in 1.61 seconds.\n",
      "Test circle -346 trained in 1.66 seconds.\n",
      "Test circle -345 trained in 1.73 seconds.\n",
      "Test circle -344 trained in 1.85 seconds.\n",
      "Test circle -343 trained in 1.8 seconds.\n",
      "Test circle -342 trained in 1.64 seconds.\n",
      "Test circle -341 trained in 1.63 seconds.\n",
      "Test circle -340 trained in 1.6 seconds.\n",
      "Test circle -339 trained in 1.64 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -338 trained in 1.69 seconds.\n",
      "Test circle -337 trained in 1.72 seconds.\n",
      "Test circle -336 trained in 1.81 seconds.\n",
      "Test circle -335 trained in 1.66 seconds.\n",
      "Test circle -334 trained in 1.67 seconds.\n",
      "Test circle -333 trained in 1.7 seconds.\n",
      "Test circle -332 trained in 1.74 seconds.\n",
      "Test circle -331 trained in 1.76 seconds.\n",
      "Test circle -330 trained in 1.94 seconds.\n",
      "Test circle -329 trained in 1.78 seconds.\n",
      "Test circle -328 trained in 1.79 seconds.\n",
      "Test circle -327 trained in 1.67 seconds.\n",
      "Test circle -326 trained in 1.66 seconds.\n",
      "Test circle -325 trained in 1.68 seconds.\n",
      "Test circle -324 trained in 1.69 seconds.\n",
      "Test circle -323 trained in 1.65 seconds.\n",
      "Test circle -322 trained in 1.63 seconds.\n",
      "Test circle -321 trained in 1.63 seconds.\n",
      "Test circle -320 trained in 1.73 seconds.\n",
      "Test circle -319 trained in 1.72 seconds.\n",
      "Test circle -318 trained in 1.63 seconds.\n",
      "Test circle -317 trained in 1.61 seconds.\n",
      "Test circle -316 trained in 1.76 seconds.\n",
      "Test circle -315 trained in 1.73 seconds.\n",
      "Test circle -314 trained in 1.72 seconds.\n",
      "Test circle -313 trained in 1.62 seconds.\n",
      "Test circle -312 trained in 1.62 seconds.\n",
      "Test circle -311 trained in 1.61 seconds.\n",
      "Test circle -310 trained in 1.62 seconds.\n",
      "Test circle -309 trained in 1.65 seconds.\n",
      "Test circle -308 trained in 1.74 seconds.\n",
      "Test circle -307 trained in 1.61 seconds.\n",
      "Test circle -306 trained in 1.62 seconds.\n",
      "Test circle -305 trained in 1.64 seconds.\n",
      "Test circle -304 trained in 1.66 seconds.\n",
      "Test circle -303 trained in 1.66 seconds.\n",
      "Test circle -302 trained in 1.67 seconds.\n",
      "Test circle -301 trained in 1.64 seconds.\n",
      "Test circle -300 trained in 1.69 seconds.\n",
      "Test circle -299 trained in 1.68 seconds.\n",
      "Test circle -298 trained in 1.65 seconds.\n",
      "Test circle -297 trained in 1.65 seconds.\n",
      "Test circle -296 trained in 1.69 seconds.\n",
      "Test circle -295 trained in 1.77 seconds.\n",
      "Test circle -294 trained in 1.77 seconds.\n",
      "Test circle -293 trained in 1.72 seconds.\n",
      "Test circle -292 trained in 1.71 seconds.\n",
      "Test circle -291 trained in 1.68 seconds.\n",
      "Test circle -290 trained in 1.63 seconds.\n",
      "Test circle -289 trained in 1.63 seconds.\n",
      "Test circle -288 trained in 1.72 seconds.\n",
      "Test circle -287 trained in 1.72 seconds.\n",
      "Test circle -286 trained in 1.69 seconds.\n",
      "Test circle -285 trained in 1.71 seconds.\n",
      "Test circle -284 trained in 1.65 seconds.\n",
      "Test circle -283 trained in 1.65 seconds.\n",
      "Test circle -282 trained in 1.62 seconds.\n",
      "Test circle -281 trained in 1.62 seconds.\n",
      "Test circle -280 trained in 1.66 seconds.\n",
      "Test circle -279 trained in 1.68 seconds.\n",
      "Test circle -278 trained in 1.61 seconds.\n",
      "Test circle -277 trained in 1.65 seconds.\n",
      "Test circle -276 trained in 1.64 seconds.\n",
      "Test circle -275 trained in 1.78 seconds.\n",
      "Test circle -274 trained in 1.71 seconds.\n",
      "Test circle -273 trained in 1.64 seconds.\n",
      "Test circle -272 trained in 1.69 seconds.\n",
      "Test circle -271 trained in 1.69 seconds.\n",
      "Test circle -270 trained in 1.85 seconds.\n",
      "Test circle -269 trained in 1.88 seconds.\n",
      "Test circle -268 trained in 1.7 seconds.\n",
      "Test circle -267 trained in 1.69 seconds.\n",
      "Test circle -266 trained in 1.66 seconds.\n",
      "Test circle -265 trained in 1.67 seconds.\n",
      "Test circle -264 trained in 1.68 seconds.\n",
      "Test circle -263 trained in 1.69 seconds.\n",
      "Test circle -262 trained in 1.75 seconds.\n",
      "Test circle -261 trained in 1.84 seconds.\n",
      "Test circle -260 trained in 1.7 seconds.\n",
      "Test circle -259 trained in 1.68 seconds.\n",
      "Test circle -258 trained in 1.79 seconds.\n",
      "Test circle -257 trained in 1.69 seconds.\n",
      "Test circle -256 trained in 1.75 seconds.\n",
      "Test circle -255 trained in 1.65 seconds.\n",
      "Test circle -254 trained in 1.65 seconds.\n",
      "Test circle -253 trained in 1.63 seconds.\n",
      "Test circle -252 trained in 1.65 seconds.\n",
      "Test circle -251 trained in 1.63 seconds.\n",
      "Test circle -250 trained in 1.62 seconds.\n",
      "Test circle -249 trained in 1.61 seconds.\n",
      "Test circle -248 trained in 1.6 seconds.\n",
      "Test circle -247 trained in 1.67 seconds.\n",
      "Test circle -246 trained in 1.64 seconds.\n",
      "Test circle -245 trained in 1.65 seconds.\n",
      "Test circle -244 trained in 1.68 seconds.\n",
      "Test circle -243 trained in 1.73 seconds.\n",
      "Test circle -242 trained in 1.62 seconds.\n",
      "Test circle -241 trained in 1.67 seconds.\n",
      "Test circle -240 trained in 1.69 seconds.\n",
      "Test circle -239 trained in 1.79 seconds.\n",
      "Test circle -238 trained in 1.8 seconds.\n",
      "Test circle -237 trained in 1.64 seconds.\n",
      "Test circle -236 trained in 1.78 seconds.\n",
      "Test circle -235 trained in 1.63 seconds.\n",
      "Test circle -234 trained in 1.63 seconds.\n",
      "Test circle -233 trained in 1.61 seconds.\n",
      "Test circle -232 trained in 1.64 seconds.\n",
      "Test circle -231 trained in 1.65 seconds.\n",
      "Test circle -230 trained in 1.67 seconds.\n",
      "Test circle -229 trained in 1.76 seconds.\n",
      "Test circle -228 trained in 1.74 seconds.\n",
      "Test circle -227 trained in 1.67 seconds.\n",
      "Test circle -226 trained in 1.67 seconds.\n",
      "Test circle -225 trained in 1.8 seconds.\n",
      "Test circle -224 trained in 1.92 seconds.\n",
      "Test circle -223 trained in 1.72 seconds.\n",
      "Test circle -222 trained in 1.67 seconds.\n",
      "Test circle -221 trained in 1.64 seconds.\n",
      "Test circle -220 trained in 1.62 seconds.\n",
      "Test circle -219 trained in 1.63 seconds.\n",
      "Test circle -218 trained in 1.64 seconds.\n",
      "Test circle -217 trained in 1.64 seconds.\n",
      "Test circle -216 trained in 1.69 seconds.\n",
      "Test circle -215 trained in 1.62 seconds.\n",
      "Test circle -214 trained in 1.66 seconds.\n",
      "Test circle -213 trained in 1.61 seconds.\n",
      "Test circle -212 trained in 1.6 seconds.\n",
      "Test circle -211 trained in 1.67 seconds.\n",
      "Test circle -210 trained in 1.61 seconds.\n",
      "Test circle -209 trained in 1.6 seconds.\n",
      "Test circle -208 trained in 1.64 seconds.\n",
      "Test circle -207 trained in 1.61 seconds.\n",
      "Test circle -206 trained in 1.6 seconds.\n",
      "Test circle -205 trained in 1.61 seconds.\n",
      "Test circle -204 trained in 1.61 seconds.\n",
      "Test circle -203 trained in 1.6 seconds.\n",
      "Test circle -202 trained in 1.61 seconds.\n",
      "Test circle -201 trained in 1.6 seconds.\n",
      "Test circle -200 trained in 1.6 seconds.\n",
      "Test circle -199 trained in 1.62 seconds.\n",
      "Test circle -198 trained in 1.61 seconds.\n",
      "Test circle -197 trained in 1.6 seconds.\n",
      "Test circle -196 trained in 1.59 seconds.\n",
      "Test circle -195 trained in 1.69 seconds.\n",
      "Test circle -194 trained in 1.59 seconds.\n",
      "Test circle -193 trained in 1.61 seconds.\n",
      "Test circle -192 trained in 1.68 seconds.\n",
      "Test circle -191 trained in 1.62 seconds.\n",
      "Test circle -190 trained in 1.63 seconds.\n",
      "Test circle -189 trained in 1.62 seconds.\n",
      "Test circle -188 trained in 1.6 seconds.\n",
      "Test circle -187 trained in 1.62 seconds.\n",
      "Test circle -186 trained in 1.71 seconds.\n",
      "Test circle -185 trained in 1.75 seconds.\n",
      "Test circle -184 trained in 1.62 seconds.\n",
      "Test circle -183 trained in 1.67 seconds.\n",
      "Test circle -182 trained in 1.67 seconds.\n",
      "Test circle -181 trained in 1.7 seconds.\n",
      "Test circle -180 trained in 1.63 seconds.\n",
      "Test circle -179 trained in 1.67 seconds.\n",
      "Test circle -178 trained in 1.61 seconds.\n",
      "Test circle -177 trained in 1.6 seconds.\n",
      "Test circle -176 trained in 1.68 seconds.\n",
      "Test circle -175 trained in 1.6 seconds.\n",
      "Test circle -174 trained in 1.6 seconds.\n",
      "Test circle -173 trained in 1.62 seconds.\n",
      "Test circle -172 trained in 1.72 seconds.\n",
      "Test circle -171 trained in 1.72 seconds.\n",
      "Test circle -170 trained in 1.64 seconds.\n",
      "Test circle -169 trained in 1.66 seconds.\n",
      "Test circle -168 trained in 1.67 seconds.\n",
      "Test circle -167 trained in 1.67 seconds.\n",
      "Test circle -166 trained in 1.62 seconds.\n",
      "Test circle -165 trained in 1.61 seconds.\n",
      "Test circle -164 trained in 1.61 seconds.\n",
      "Test circle -163 trained in 1.67 seconds.\n",
      "Test circle -162 trained in 1.64 seconds.\n",
      "Test circle -161 trained in 1.61 seconds.\n",
      "Test circle -160 trained in 1.61 seconds.\n",
      "Test circle -159 trained in 1.61 seconds.\n",
      "Test circle -158 trained in 1.6 seconds.\n",
      "Test circle -157 trained in 1.59 seconds.\n",
      "Test circle -156 trained in 1.6 seconds.\n",
      "Test circle -155 trained in 1.7 seconds.\n",
      "Test circle -154 trained in 1.61 seconds.\n",
      "Test circle -153 trained in 1.61 seconds.\n",
      "Test circle -152 trained in 1.61 seconds.\n",
      "Test circle -151 trained in 1.61 seconds.\n",
      "Test circle -150 trained in 1.7 seconds.\n",
      "Test circle -149 trained in 1.64 seconds.\n",
      "Test circle -148 trained in 1.62 seconds.\n",
      "Test circle -147 trained in 1.61 seconds.\n",
      "Test circle -146 trained in 1.6 seconds.\n",
      "Test circle -145 trained in 1.62 seconds.\n",
      "Test circle -144 trained in 1.63 seconds.\n",
      "Test circle -143 trained in 1.62 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -142 trained in 1.61 seconds.\n",
      "Test circle -141 trained in 1.61 seconds.\n",
      "Test circle -140 trained in 1.61 seconds.\n",
      "Test circle -139 trained in 1.61 seconds.\n",
      "Test circle -138 trained in 1.61 seconds.\n",
      "Test circle -137 trained in 1.82 seconds.\n",
      "Test circle -136 trained in 1.91 seconds.\n",
      "Test circle -135 trained in 1.71 seconds.\n",
      "Test circle -134 trained in 1.69 seconds.\n",
      "Test circle -133 trained in 1.62 seconds.\n",
      "Test circle -132 trained in 1.67 seconds.\n",
      "Test circle -131 trained in 1.62 seconds.\n",
      "Test circle -130 trained in 1.63 seconds.\n",
      "Test circle -129 trained in 1.65 seconds.\n",
      "Test circle -128 trained in 1.63 seconds.\n",
      "Test circle -127 trained in 1.63 seconds.\n",
      "Test circle -126 trained in 1.63 seconds.\n",
      "Test circle -125 trained in 1.61 seconds.\n",
      "Test circle -124 trained in 1.62 seconds.\n",
      "Test circle -123 trained in 1.65 seconds.\n",
      "Test circle -122 trained in 1.61 seconds.\n",
      "Test circle -121 trained in 1.61 seconds.\n",
      "Test circle -120 trained in 1.6 seconds.\n",
      "Test circle -119 trained in 1.63 seconds.\n",
      "Test circle -118 trained in 1.6 seconds.\n",
      "Test circle -117 trained in 1.69 seconds.\n",
      "Test circle -116 trained in 1.61 seconds.\n",
      "Test circle -115 trained in 1.61 seconds.\n",
      "Test circle -114 trained in 1.62 seconds.\n",
      "Test circle -113 trained in 1.62 seconds.\n",
      "Test circle -112 trained in 1.61 seconds.\n",
      "Test circle -111 trained in 1.6 seconds.\n",
      "Test circle -110 trained in 1.61 seconds.\n",
      "Test circle -109 trained in 1.62 seconds.\n",
      "Test circle -108 trained in 1.63 seconds.\n",
      "Test circle -107 trained in 1.61 seconds.\n",
      "Test circle -106 trained in 1.6 seconds.\n",
      "Test circle -105 trained in 1.59 seconds.\n",
      "Test circle -104 trained in 1.6 seconds.\n",
      "Test circle -103 trained in 1.61 seconds.\n",
      "Test circle -102 trained in 1.59 seconds.\n",
      "Test circle -101 trained in 1.61 seconds.\n",
      "Test circle -100 trained in 1.6 seconds.\n",
      "Test circle -99 trained in 1.6 seconds.\n",
      "Test circle -98 trained in 1.62 seconds.\n",
      "Test circle -97 trained in 1.71 seconds.\n",
      "Test circle -96 trained in 1.61 seconds.\n",
      "Test circle -95 trained in 1.6 seconds.\n",
      "Test circle -94 trained in 1.61 seconds.\n",
      "Test circle -93 trained in 1.65 seconds.\n",
      "Test circle -92 trained in 1.69 seconds.\n",
      "Test circle -91 trained in 1.61 seconds.\n",
      "Test circle -90 trained in 1.61 seconds.\n",
      "Test circle -89 trained in 1.62 seconds.\n",
      "Test circle -88 trained in 1.67 seconds.\n",
      "Test circle -87 trained in 1.62 seconds.\n",
      "Test circle -86 trained in 1.63 seconds.\n",
      "Test circle -85 trained in 1.63 seconds.\n",
      "Test circle -84 trained in 1.64 seconds.\n",
      "Test circle -83 trained in 1.66 seconds.\n",
      "Test circle -82 trained in 1.67 seconds.\n",
      "Test circle -81 trained in 1.67 seconds.\n",
      "Test circle -80 trained in 1.6 seconds.\n",
      "Test circle -79 trained in 1.6 seconds.\n",
      "Test circle -78 trained in 1.69 seconds.\n",
      "Test circle -77 trained in 1.66 seconds.\n",
      "Test circle -76 trained in 1.66 seconds.\n",
      "Test circle -75 trained in 1.63 seconds.\n",
      "Test circle -74 trained in 1.61 seconds.\n",
      "Test circle -73 trained in 1.67 seconds.\n",
      "Test circle -72 trained in 1.61 seconds.\n",
      "Test circle -71 trained in 1.61 seconds.\n",
      "Test circle -70 trained in 1.61 seconds.\n",
      "Test circle -69 trained in 1.69 seconds.\n",
      "Test circle -68 trained in 1.73 seconds.\n",
      "Test circle -67 trained in 1.64 seconds.\n",
      "Test circle -66 trained in 1.63 seconds.\n",
      "Test circle -65 trained in 1.6 seconds.\n",
      "Test circle -64 trained in 1.61 seconds.\n",
      "Test circle -63 trained in 1.61 seconds.\n",
      "Test circle -62 trained in 1.6 seconds.\n",
      "Test circle -61 trained in 1.66 seconds.\n",
      "Test circle -60 trained in 1.61 seconds.\n",
      "Test circle -59 trained in 1.69 seconds.\n",
      "Test circle -58 trained in 1.62 seconds.\n",
      "Test circle -57 trained in 1.62 seconds.\n",
      "Test circle -56 trained in 1.71 seconds.\n",
      "Test circle -55 trained in 1.63 seconds.\n",
      "Test circle -54 trained in 1.6 seconds.\n",
      "Test circle -53 trained in 1.61 seconds.\n",
      "Test circle -52 trained in 1.6 seconds.\n",
      "Test circle -51 trained in 1.62 seconds.\n",
      "Test circle -50 trained in 1.64 seconds.\n",
      "Test circle -49 trained in 1.61 seconds.\n",
      "Test circle -48 trained in 1.65 seconds.\n",
      "Test circle -47 trained in 1.69 seconds.\n",
      "Test circle -46 trained in 1.67 seconds.\n",
      "Test circle -45 trained in 1.75 seconds.\n",
      "Test circle -44 trained in 1.64 seconds.\n",
      "Test circle -43 trained in 1.61 seconds.\n",
      "Test circle -42 trained in 1.64 seconds.\n",
      "Test circle -41 trained in 1.61 seconds.\n",
      "Test circle -40 trained in 1.64 seconds.\n",
      "Test circle -39 trained in 1.61 seconds.\n",
      "Test circle -38 trained in 1.77 seconds.\n",
      "Test circle -37 trained in 1.61 seconds.\n",
      "Test circle -36 trained in 1.61 seconds.\n",
      "Test circle -35 trained in 1.65 seconds.\n",
      "Test circle -34 trained in 1.7 seconds.\n",
      "Test circle -33 trained in 1.77 seconds.\n",
      "Test circle -32 trained in 1.61 seconds.\n",
      "Test circle -31 trained in 1.59 seconds.\n",
      "Test circle -30 trained in 1.74 seconds.\n",
      "Test circle -29 trained in 1.69 seconds.\n",
      "Test circle -28 trained in 1.66 seconds.\n",
      "Test circle -27 trained in 1.64 seconds.\n",
      "Test circle -26 trained in 1.72 seconds.\n",
      "Test circle -25 trained in 1.84 seconds.\n",
      "Test circle -24 trained in 1.89 seconds.\n",
      "Test circle -23 trained in 1.77 seconds.\n",
      "Test circle -22 trained in 1.69 seconds.\n",
      "Test circle -21 trained in 1.62 seconds.\n",
      "Test circle -20 trained in 1.6 seconds.\n",
      "Test circle -19 trained in 1.7 seconds.\n",
      "Test circle -18 trained in 1.61 seconds.\n",
      "Test circle -17 trained in 1.61 seconds.\n",
      "Test circle -16 trained in 1.61 seconds.\n",
      "Test circle -15 trained in 1.61 seconds.\n",
      "Test circle -14 trained in 1.6 seconds.\n",
      "Test circle -13 trained in 1.59 seconds.\n",
      "Test circle -12 trained in 1.61 seconds.\n",
      "Test circle -11 trained in 1.61 seconds.\n",
      "Test circle -10 trained in 1.6 seconds.\n",
      "Test circle -9 trained in 1.74 seconds.\n",
      "Test circle -8 trained in 1.63 seconds.\n",
      "Test circle -7 trained in 1.61 seconds.\n",
      "Test circle -6 trained in 1.77 seconds.\n",
      "Test circle -5 trained in 1.64 seconds.\n",
      "Test circle -4 trained in 1.73 seconds.\n",
      "Test circle -3 trained in 1.65 seconds.\n",
      "Test circle -2 trained in 1.65 seconds.\n",
      "Test circle -1 trained in 1.6 seconds.\n",
      "Model trained in 1188.1 seconds.\n",
      "Hyperparams (25, 3, 3) trained in 1188.1 seconds.\n",
      "=========================================================================================\n",
      "Model (25, 3, 3) Ended\n",
      "=========================================================================================\n",
      "=========================================================================================\n",
      "Model (25, 3, 6) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 1.67 seconds.\n",
      "Test circle -724 trained in 1.73 seconds.\n",
      "Test circle -723 trained in 1.61 seconds.\n",
      "Test circle -722 trained in 1.65 seconds.\n",
      "Test circle -721 trained in 1.67 seconds.\n",
      "Test circle -720 trained in 1.64 seconds.\n",
      "Test circle -719 trained in 1.65 seconds.\n",
      "Test circle -718 trained in 1.62 seconds.\n",
      "Test circle -717 trained in 1.64 seconds.\n",
      "Test circle -716 trained in 1.73 seconds.\n",
      "Test circle -715 trained in 1.68 seconds.\n",
      "Test circle -714 trained in 1.8 seconds.\n",
      "Test circle -713 trained in 1.69 seconds.\n",
      "Test circle -712 trained in 1.61 seconds.\n",
      "Test circle -711 trained in 1.79 seconds.\n",
      "Test circle -710 trained in 1.77 seconds.\n",
      "Test circle -709 trained in 1.7 seconds.\n",
      "Test circle -708 trained in 1.69 seconds.\n",
      "Test circle -707 trained in 1.74 seconds.\n",
      "Test circle -706 trained in 1.75 seconds.\n",
      "Test circle -705 trained in 1.74 seconds.\n",
      "Test circle -704 trained in 1.83 seconds.\n",
      "Test circle -703 trained in 1.66 seconds.\n",
      "Test circle -702 trained in 1.84 seconds.\n",
      "Test circle -701 trained in 1.63 seconds.\n",
      "Test circle -700 trained in 1.6 seconds.\n",
      "Test circle -699 trained in 1.77 seconds.\n",
      "Test circle -698 trained in 1.74 seconds.\n",
      "Test circle -697 trained in 1.68 seconds.\n",
      "Test circle -696 trained in 1.84 seconds.\n",
      "Test circle -695 trained in 1.73 seconds.\n",
      "Test circle -694 trained in 1.76 seconds.\n",
      "Test circle -693 trained in 1.81 seconds.\n",
      "Test circle -692 trained in 1.64 seconds.\n",
      "Test circle -691 trained in 1.63 seconds.\n",
      "Test circle -690 trained in 1.63 seconds.\n",
      "Test circle -689 trained in 1.68 seconds.\n",
      "Test circle -688 trained in 1.84 seconds.\n",
      "Test circle -687 trained in 1.77 seconds.\n",
      "Test circle -686 trained in 1.67 seconds.\n",
      "Test circle -685 trained in 1.71 seconds.\n",
      "Test circle -684 trained in 1.64 seconds.\n",
      "Test circle -683 trained in 1.63 seconds.\n",
      "Test circle -682 trained in 1.7 seconds.\n",
      "Test circle -681 trained in 1.89 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -680 trained in 1.69 seconds.\n",
      "Test circle -679 trained in 1.72 seconds.\n",
      "Test circle -678 trained in 1.69 seconds.\n",
      "Test circle -677 trained in 1.65 seconds.\n",
      "Test circle -676 trained in 1.64 seconds.\n",
      "Test circle -675 trained in 1.63 seconds.\n",
      "Test circle -674 trained in 1.76 seconds.\n",
      "Test circle -673 trained in 1.68 seconds.\n",
      "Test circle -672 trained in 1.66 seconds.\n",
      "Test circle -671 trained in 1.73 seconds.\n",
      "Test circle -670 trained in 1.82 seconds.\n",
      "Test circle -669 trained in 1.66 seconds.\n",
      "Test circle -668 trained in 1.65 seconds.\n",
      "Test circle -667 trained in 1.65 seconds.\n",
      "Test circle -666 trained in 1.61 seconds.\n",
      "Test circle -665 trained in 1.69 seconds.\n",
      "Test circle -664 trained in 1.63 seconds.\n",
      "Test circle -663 trained in 1.62 seconds.\n",
      "Test circle -662 trained in 1.67 seconds.\n",
      "Test circle -661 trained in 1.64 seconds.\n",
      "Test circle -660 trained in 1.65 seconds.\n",
      "Test circle -659 trained in 1.73 seconds.\n",
      "Test circle -658 trained in 2.04 seconds.\n",
      "Test circle -657 trained in 1.83 seconds.\n",
      "Test circle -656 trained in 1.74 seconds.\n",
      "Test circle -655 trained in 1.74 seconds.\n",
      "Test circle -654 trained in 1.68 seconds.\n",
      "Test circle -653 trained in 1.66 seconds.\n",
      "Test circle -652 trained in 1.66 seconds.\n",
      "Test circle -651 trained in 1.62 seconds.\n",
      "Test circle -650 trained in 1.63 seconds.\n",
      "Test circle -649 trained in 1.68 seconds.\n",
      "Test circle -648 trained in 1.7 seconds.\n",
      "Test circle -647 trained in 1.64 seconds.\n",
      "Test circle -646 trained in 1.65 seconds.\n",
      "Test circle -645 trained in 1.7 seconds.\n",
      "Test circle -644 trained in 1.63 seconds.\n",
      "Test circle -643 trained in 1.66 seconds.\n",
      "Test circle -642 trained in 1.72 seconds.\n",
      "Test circle -641 trained in 1.71 seconds.\n",
      "Test circle -640 trained in 1.73 seconds.\n",
      "Test circle -639 trained in 1.69 seconds.\n",
      "Test circle -638 trained in 1.62 seconds.\n",
      "Test circle -637 trained in 1.65 seconds.\n",
      "Test circle -636 trained in 1.64 seconds.\n",
      "Test circle -635 trained in 1.63 seconds.\n",
      "Test circle -634 trained in 1.61 seconds.\n",
      "Test circle -633 trained in 1.61 seconds.\n",
      "Test circle -632 trained in 1.6 seconds.\n",
      "Test circle -631 trained in 1.82 seconds.\n",
      "Test circle -630 trained in 1.66 seconds.\n",
      "Test circle -629 trained in 1.73 seconds.\n",
      "Test circle -628 trained in 1.7 seconds.\n",
      "Test circle -627 trained in 1.56 seconds.\n",
      "Test circle -626 trained in 1.52 seconds.\n",
      "Test circle -625 trained in 1.6 seconds.\n",
      "Test circle -624 trained in 1.51 seconds.\n",
      "Test circle -623 trained in 1.52 seconds.\n",
      "Test circle -622 trained in 1.52 seconds.\n",
      "Test circle -621 trained in 1.54 seconds.\n",
      "Test circle -620 trained in 1.57 seconds.\n",
      "Test circle -619 trained in 1.55 seconds.\n",
      "Test circle -618 trained in 1.56 seconds.\n",
      "Test circle -617 trained in 1.59 seconds.\n",
      "Test circle -616 trained in 1.54 seconds.\n",
      "Test circle -615 trained in 1.65 seconds.\n",
      "Test circle -614 trained in 1.64 seconds.\n",
      "Test circle -613 trained in 1.7 seconds.\n",
      "Test circle -612 trained in 1.9 seconds.\n",
      "Test circle -611 trained in 1.73 seconds.\n",
      "Test circle -610 trained in 1.54 seconds.\n",
      "Test circle -609 trained in 1.62 seconds.\n",
      "Test circle -608 trained in 1.57 seconds.\n",
      "Test circle -607 trained in 1.59 seconds.\n",
      "Test circle -606 trained in 1.63 seconds.\n",
      "Test circle -605 trained in 1.52 seconds.\n",
      "Test circle -604 trained in 1.56 seconds.\n",
      "Test circle -603 trained in 1.62 seconds.\n",
      "Test circle -602 trained in 1.61 seconds.\n",
      "Test circle -601 trained in 1.61 seconds.\n",
      "Test circle -600 trained in 1.55 seconds.\n",
      "Test circle -599 trained in 1.57 seconds.\n",
      "Test circle -598 trained in 1.61 seconds.\n",
      "Test circle -597 trained in 1.62 seconds.\n",
      "Test circle -596 trained in 1.52 seconds.\n",
      "Test circle -595 trained in 1.57 seconds.\n",
      "Test circle -594 trained in 1.58 seconds.\n",
      "Test circle -593 trained in 1.62 seconds.\n",
      "Test circle -592 trained in 1.58 seconds.\n",
      "Test circle -591 trained in 1.58 seconds.\n",
      "Test circle -590 trained in 1.54 seconds.\n",
      "Test circle -589 trained in 1.54 seconds.\n",
      "Test circle -588 trained in 1.56 seconds.\n",
      "Test circle -587 trained in 1.66 seconds.\n",
      "Test circle -586 trained in 1.74 seconds.\n",
      "Test circle -585 trained in 1.6 seconds.\n",
      "Test circle -584 trained in 1.64 seconds.\n",
      "Test circle -583 trained in 1.62 seconds.\n",
      "Test circle -582 trained in 1.62 seconds.\n",
      "Test circle -581 trained in 1.6 seconds.\n",
      "Test circle -580 trained in 1.57 seconds.\n",
      "Test circle -579 trained in 1.55 seconds.\n",
      "Test circle -578 trained in 1.61 seconds.\n",
      "Test circle -577 trained in 1.61 seconds.\n",
      "Test circle -576 trained in 1.61 seconds.\n",
      "Test circle -575 trained in 1.61 seconds.\n",
      "Test circle -574 trained in 1.57 seconds.\n",
      "Test circle -573 trained in 1.59 seconds.\n",
      "Test circle -572 trained in 1.64 seconds.\n",
      "Test circle -571 trained in 1.58 seconds.\n",
      "Test circle -570 trained in 1.55 seconds.\n",
      "Test circle -569 trained in 1.6 seconds.\n",
      "Test circle -568 trained in 1.55 seconds.\n",
      "Test circle -567 trained in 1.63 seconds.\n",
      "Test circle -566 trained in 1.67 seconds.\n",
      "Test circle -565 trained in 1.74 seconds.\n",
      "Test circle -564 trained in 1.67 seconds.\n",
      "Test circle -563 trained in 2.29 seconds.\n",
      "Test circle -562 trained in 1.76 seconds.\n",
      "Test circle -561 trained in 1.7 seconds.\n",
      "Test circle -560 trained in 1.56 seconds.\n",
      "Test circle -559 trained in 1.59 seconds.\n",
      "Test circle -558 trained in 1.61 seconds.\n",
      "Test circle -557 trained in 1.69 seconds.\n",
      "Test circle -556 trained in 1.75 seconds.\n",
      "Test circle -555 trained in 1.73 seconds.\n",
      "Test circle -554 trained in 1.84 seconds.\n",
      "Test circle -553 trained in 1.64 seconds.\n",
      "Test circle -552 trained in 1.72 seconds.\n",
      "Test circle -551 trained in 1.67 seconds.\n",
      "Test circle -550 trained in 1.77 seconds.\n",
      "Test circle -549 trained in 1.7 seconds.\n",
      "Test circle -548 trained in 1.68 seconds.\n",
      "Test circle -547 trained in 1.77 seconds.\n",
      "Test circle -546 trained in 1.87 seconds.\n",
      "Test circle -545 trained in 1.76 seconds.\n",
      "Test circle -544 trained in 1.73 seconds.\n",
      "Test circle -543 trained in 1.75 seconds.\n",
      "Test circle -542 trained in 1.65 seconds.\n",
      "Test circle -541 trained in 1.61 seconds.\n",
      "Test circle -540 trained in 1.63 seconds.\n",
      "Test circle -539 trained in 1.65 seconds.\n",
      "Test circle -538 trained in 1.67 seconds.\n",
      "Test circle -537 trained in 1.63 seconds.\n",
      "Test circle -536 trained in 1.6 seconds.\n",
      "Test circle -535 trained in 1.58 seconds.\n",
      "Test circle -534 trained in 1.61 seconds.\n",
      "Test circle -533 trained in 1.66 seconds.\n",
      "Test circle -532 trained in 1.61 seconds.\n",
      "Test circle -531 trained in 1.59 seconds.\n",
      "Test circle -530 trained in 1.59 seconds.\n",
      "Test circle -529 trained in 1.58 seconds.\n",
      "Test circle -528 trained in 1.59 seconds.\n",
      "Test circle -527 trained in 1.71 seconds.\n",
      "Test circle -526 trained in 2.01 seconds.\n",
      "Test circle -525 trained in 1.76 seconds.\n",
      "Test circle -524 trained in 1.75 seconds.\n",
      "Test circle -523 trained in 1.66 seconds.\n",
      "Test circle -522 trained in 1.78 seconds.\n",
      "Test circle -521 trained in 1.82 seconds.\n",
      "Test circle -520 trained in 1.62 seconds.\n",
      "Test circle -519 trained in 1.62 seconds.\n",
      "Test circle -518 trained in 1.73 seconds.\n",
      "Test circle -517 trained in 1.83 seconds.\n",
      "Test circle -516 trained in 1.77 seconds.\n",
      "Test circle -515 trained in 1.66 seconds.\n",
      "Test circle -514 trained in 1.59 seconds.\n",
      "Test circle -513 trained in 1.73 seconds.\n",
      "Test circle -512 trained in 1.63 seconds.\n",
      "Test circle -511 trained in 1.65 seconds.\n",
      "Test circle -510 trained in 1.62 seconds.\n",
      "Test circle -509 trained in 1.64 seconds.\n",
      "Test circle -508 trained in 1.63 seconds.\n",
      "Test circle -507 trained in 1.8 seconds.\n",
      "Test circle -506 trained in 1.62 seconds.\n",
      "Test circle -505 trained in 1.74 seconds.\n",
      "Test circle -504 trained in 1.71 seconds.\n",
      "Test circle -503 trained in 1.67 seconds.\n",
      "Test circle -502 trained in 1.76 seconds.\n",
      "Test circle -501 trained in 1.69 seconds.\n",
      "Test circle -500 trained in 1.62 seconds.\n",
      "Test circle -499 trained in 1.62 seconds.\n",
      "Test circle -498 trained in 1.78 seconds.\n",
      "Test circle -497 trained in 1.69 seconds.\n",
      "Test circle -496 trained in 1.6 seconds.\n",
      "Test circle -495 trained in 1.62 seconds.\n",
      "Test circle -494 trained in 1.62 seconds.\n",
      "Test circle -493 trained in 1.69 seconds.\n",
      "Test circle -492 trained in 1.7 seconds.\n",
      "Test circle -491 trained in 1.68 seconds.\n",
      "Test circle -490 trained in 1.6 seconds.\n",
      "Test circle -489 trained in 1.73 seconds.\n",
      "Test circle -488 trained in 1.7 seconds.\n",
      "Test circle -487 trained in 1.89 seconds.\n",
      "Test circle -486 trained in 1.68 seconds.\n",
      "Test circle -485 trained in 1.71 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -484 trained in 1.66 seconds.\n",
      "Test circle -483 trained in 1.71 seconds.\n",
      "Test circle -482 trained in 1.72 seconds.\n",
      "Test circle -481 trained in 1.63 seconds.\n",
      "Test circle -480 trained in 1.61 seconds.\n",
      "Test circle -479 trained in 1.61 seconds.\n",
      "Test circle -478 trained in 1.61 seconds.\n",
      "Test circle -477 trained in 1.66 seconds.\n",
      "Test circle -476 trained in 1.87 seconds.\n",
      "Test circle -475 trained in 1.63 seconds.\n",
      "Test circle -474 trained in 1.62 seconds.\n",
      "Test circle -473 trained in 1.64 seconds.\n",
      "Test circle -472 trained in 1.65 seconds.\n",
      "Test circle -471 trained in 1.6 seconds.\n",
      "Test circle -470 trained in 1.69 seconds.\n",
      "Test circle -469 trained in 1.63 seconds.\n",
      "Test circle -468 trained in 1.62 seconds.\n",
      "Test circle -467 trained in 1.64 seconds.\n",
      "Test circle -466 trained in 1.67 seconds.\n",
      "Test circle -465 trained in 1.62 seconds.\n",
      "Test circle -464 trained in 1.62 seconds.\n",
      "Test circle -463 trained in 1.66 seconds.\n",
      "Test circle -462 trained in 1.68 seconds.\n",
      "Test circle -461 trained in 1.68 seconds.\n",
      "Test circle -460 trained in 1.68 seconds.\n",
      "Test circle -459 trained in 1.67 seconds.\n",
      "Test circle -458 trained in 1.68 seconds.\n",
      "Test circle -457 trained in 1.7 seconds.\n",
      "Test circle -456 trained in 1.63 seconds.\n",
      "Test circle -455 trained in 1.61 seconds.\n",
      "Test circle -454 trained in 1.64 seconds.\n",
      "Test circle -453 trained in 1.66 seconds.\n",
      "Test circle -452 trained in 1.63 seconds.\n",
      "Test circle -451 trained in 1.63 seconds.\n",
      "Test circle -450 trained in 1.61 seconds.\n",
      "Test circle -449 trained in 1.71 seconds.\n",
      "Test circle -448 trained in 1.7 seconds.\n",
      "Test circle -447 trained in 1.69 seconds.\n",
      "Test circle -446 trained in 1.63 seconds.\n",
      "Test circle -445 trained in 1.6 seconds.\n",
      "Test circle -444 trained in 1.75 seconds.\n",
      "Test circle -443 trained in 1.7 seconds.\n",
      "Test circle -442 trained in 1.68 seconds.\n",
      "Test circle -441 trained in 1.68 seconds.\n",
      "Test circle -440 trained in 1.64 seconds.\n",
      "Test circle -439 trained in 1.65 seconds.\n",
      "Test circle -438 trained in 1.69 seconds.\n",
      "Test circle -437 trained in 1.72 seconds.\n",
      "Test circle -436 trained in 1.64 seconds.\n",
      "Test circle -435 trained in 1.63 seconds.\n",
      "Test circle -434 trained in 1.64 seconds.\n",
      "Test circle -433 trained in 1.75 seconds.\n",
      "Test circle -432 trained in 1.75 seconds.\n",
      "Test circle -431 trained in 1.64 seconds.\n",
      "Test circle -430 trained in 1.87 seconds.\n",
      "Test circle -429 trained in 1.76 seconds.\n",
      "Test circle -428 trained in 1.86 seconds.\n",
      "Test circle -427 trained in 1.91 seconds.\n",
      "Test circle -426 trained in 1.89 seconds.\n",
      "Test circle -425 trained in 1.83 seconds.\n",
      "Test circle -424 trained in 1.77 seconds.\n",
      "Test circle -423 trained in 1.76 seconds.\n",
      "Test circle -422 trained in 1.77 seconds.\n",
      "Test circle -421 trained in 1.82 seconds.\n",
      "Test circle -420 trained in 1.69 seconds.\n",
      "Test circle -419 trained in 1.78 seconds.\n",
      "Test circle -418 trained in 1.94 seconds.\n",
      "Test circle -417 trained in 1.78 seconds.\n",
      "Test circle -416 trained in 1.75 seconds.\n",
      "Test circle -415 trained in 1.85 seconds.\n",
      "Test circle -414 trained in 1.84 seconds.\n",
      "Test circle -413 trained in 1.92 seconds.\n",
      "Test circle -412 trained in 1.8 seconds.\n",
      "Test circle -411 trained in 1.74 seconds.\n",
      "Test circle -410 trained in 1.81 seconds.\n",
      "Test circle -409 trained in 1.94 seconds.\n",
      "Test circle -408 trained in 1.86 seconds.\n",
      "Test circle -407 trained in 1.88 seconds.\n",
      "Test circle -406 trained in 1.97 seconds.\n",
      "Test circle -405 trained in 1.93 seconds.\n",
      "Test circle -404 trained in 1.92 seconds.\n",
      "Test circle -403 trained in 2.35 seconds.\n",
      "Test circle -402 trained in 2.21 seconds.\n",
      "Test circle -401 trained in 2.06 seconds.\n",
      "Test circle -400 trained in 1.93 seconds.\n",
      "Test circle -399 trained in 1.95 seconds.\n",
      "Test circle -398 trained in 1.74 seconds.\n",
      "Test circle -397 trained in 1.54 seconds.\n",
      "Test circle -396 trained in 1.58 seconds.\n",
      "Test circle -395 trained in 1.54 seconds.\n",
      "Test circle -394 trained in 1.55 seconds.\n",
      "Test circle -393 trained in 1.56 seconds.\n",
      "Test circle -392 trained in 1.62 seconds.\n",
      "Test circle -391 trained in 1.67 seconds.\n",
      "Test circle -390 trained in 1.65 seconds.\n",
      "Test circle -389 trained in 1.56 seconds.\n",
      "Test circle -388 trained in 1.76 seconds.\n",
      "Test circle -387 trained in 1.97 seconds.\n",
      "Test circle -386 trained in 1.67 seconds.\n",
      "Test circle -385 trained in 1.61 seconds.\n",
      "Test circle -384 trained in 1.94 seconds.\n",
      "Test circle -383 trained in 2.41 seconds.\n",
      "Test circle -382 trained in 1.81 seconds.\n",
      "Test circle -381 trained in 1.74 seconds.\n",
      "Test circle -380 trained in 1.85 seconds.\n",
      "Test circle -379 trained in 2.09 seconds.\n",
      "Test circle -378 trained in 1.66 seconds.\n",
      "Test circle -377 trained in 1.63 seconds.\n",
      "Test circle -376 trained in 1.58 seconds.\n",
      "Test circle -375 trained in 1.63 seconds.\n",
      "Test circle -374 trained in 1.69 seconds.\n",
      "Test circle -373 trained in 1.63 seconds.\n",
      "Test circle -372 trained in 1.61 seconds.\n",
      "Test circle -371 trained in 1.6 seconds.\n",
      "Test circle -370 trained in 1.58 seconds.\n",
      "Test circle -369 trained in 1.73 seconds.\n",
      "Test circle -368 trained in 1.64 seconds.\n",
      "Test circle -367 trained in 1.64 seconds.\n",
      "Test circle -366 trained in 1.61 seconds.\n",
      "Test circle -365 trained in 1.58 seconds.\n",
      "Test circle -364 trained in 1.62 seconds.\n",
      "Test circle -363 trained in 1.69 seconds.\n",
      "Test circle -362 trained in 1.66 seconds.\n",
      "Test circle -361 trained in 1.6 seconds.\n",
      "Test circle -360 trained in 1.63 seconds.\n",
      "Test circle -359 trained in 1.65 seconds.\n",
      "Test circle -358 trained in 1.67 seconds.\n",
      "Test circle -357 trained in 1.67 seconds.\n",
      "Test circle -356 trained in 1.59 seconds.\n",
      "Test circle -355 trained in 1.64 seconds.\n",
      "Test circle -354 trained in 1.61 seconds.\n",
      "Test circle -353 trained in 1.66 seconds.\n",
      "Test circle -352 trained in 1.66 seconds.\n",
      "Test circle -351 trained in 1.61 seconds.\n",
      "Test circle -350 trained in 1.73 seconds.\n",
      "Test circle -349 trained in 1.66 seconds.\n",
      "Test circle -348 trained in 1.68 seconds.\n",
      "Test circle -347 trained in 1.66 seconds.\n",
      "Test circle -346 trained in 1.63 seconds.\n",
      "Test circle -345 trained in 1.63 seconds.\n",
      "Test circle -344 trained in 1.64 seconds.\n",
      "Test circle -343 trained in 1.65 seconds.\n",
      "Test circle -342 trained in 1.66 seconds.\n",
      "Test circle -341 trained in 1.62 seconds.\n",
      "Test circle -340 trained in 1.59 seconds.\n",
      "Test circle -339 trained in 1.63 seconds.\n",
      "Test circle -338 trained in 1.83 seconds.\n",
      "Test circle -337 trained in 1.69 seconds.\n",
      "Test circle -336 trained in 1.6 seconds.\n",
      "Test circle -335 trained in 1.68 seconds.\n",
      "Test circle -334 trained in 1.61 seconds.\n",
      "Test circle -333 trained in 1.67 seconds.\n",
      "Test circle -332 trained in 1.63 seconds.\n",
      "Test circle -331 trained in 1.62 seconds.\n",
      "Test circle -330 trained in 1.69 seconds.\n",
      "Test circle -329 trained in 1.65 seconds.\n",
      "Test circle -328 trained in 1.83 seconds.\n",
      "Test circle -327 trained in 1.68 seconds.\n",
      "Test circle -326 trained in 1.62 seconds.\n",
      "Test circle -325 trained in 1.6 seconds.\n",
      "Test circle -324 trained in 1.65 seconds.\n",
      "Test circle -323 trained in 1.74 seconds.\n",
      "Test circle -322 trained in 1.65 seconds.\n",
      "Test circle -321 trained in 1.63 seconds.\n",
      "Test circle -320 trained in 1.62 seconds.\n",
      "Test circle -319 trained in 1.6 seconds.\n",
      "Test circle -318 trained in 1.66 seconds.\n",
      "Test circle -317 trained in 1.64 seconds.\n",
      "Test circle -316 trained in 1.62 seconds.\n",
      "Test circle -315 trained in 1.63 seconds.\n",
      "Test circle -314 trained in 1.63 seconds.\n",
      "Test circle -313 trained in 1.74 seconds.\n",
      "Test circle -312 trained in 1.65 seconds.\n",
      "Test circle -311 trained in 1.69 seconds.\n",
      "Test circle -310 trained in 1.62 seconds.\n",
      "Test circle -309 trained in 1.6 seconds.\n",
      "Test circle -308 trained in 1.69 seconds.\n",
      "Test circle -307 trained in 1.64 seconds.\n",
      "Test circle -306 trained in 1.62 seconds.\n",
      "Test circle -305 trained in 1.61 seconds.\n",
      "Test circle -304 trained in 1.64 seconds.\n",
      "Test circle -303 trained in 1.75 seconds.\n",
      "Test circle -302 trained in 1.73 seconds.\n",
      "Test circle -301 trained in 1.65 seconds.\n",
      "Test circle -300 trained in 1.65 seconds.\n",
      "Test circle -299 trained in 1.62 seconds.\n",
      "Test circle -298 trained in 1.71 seconds.\n",
      "Test circle -297 trained in 1.69 seconds.\n",
      "Test circle -296 trained in 1.6 seconds.\n",
      "Test circle -295 trained in 1.62 seconds.\n",
      "Test circle -294 trained in 1.65 seconds.\n",
      "Test circle -293 trained in 1.67 seconds.\n",
      "Test circle -292 trained in 1.7 seconds.\n",
      "Test circle -291 trained in 1.65 seconds.\n",
      "Test circle -290 trained in 1.7 seconds.\n",
      "Test circle -289 trained in 1.66 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -288 trained in 1.71 seconds.\n",
      "Test circle -287 trained in 1.73 seconds.\n",
      "Test circle -286 trained in 1.68 seconds.\n",
      "Test circle -285 trained in 1.71 seconds.\n",
      "Test circle -284 trained in 1.67 seconds.\n",
      "Test circle -283 trained in 1.74 seconds.\n",
      "Test circle -282 trained in 1.66 seconds.\n",
      "Test circle -281 trained in 1.69 seconds.\n",
      "Test circle -280 trained in 1.68 seconds.\n",
      "Test circle -279 trained in 1.65 seconds.\n",
      "Test circle -278 trained in 1.7 seconds.\n",
      "Test circle -277 trained in 1.67 seconds.\n",
      "Test circle -276 trained in 1.68 seconds.\n",
      "Test circle -275 trained in 1.65 seconds.\n",
      "Test circle -274 trained in 1.68 seconds.\n",
      "Test circle -273 trained in 1.66 seconds.\n",
      "Test circle -272 trained in 1.7 seconds.\n",
      "Test circle -271 trained in 1.68 seconds.\n",
      "Test circle -270 trained in 1.83 seconds.\n",
      "Test circle -269 trained in 1.69 seconds.\n",
      "Test circle -268 trained in 1.68 seconds.\n",
      "Test circle -267 trained in 1.71 seconds.\n",
      "Test circle -266 trained in 1.7 seconds.\n",
      "Test circle -265 trained in 2.01 seconds.\n",
      "Test circle -264 trained in 1.74 seconds.\n",
      "Test circle -263 trained in 1.67 seconds.\n",
      "Test circle -262 trained in 1.73 seconds.\n",
      "Test circle -261 trained in 1.67 seconds.\n",
      "Test circle -260 trained in 1.7 seconds.\n",
      "Test circle -259 trained in 1.67 seconds.\n",
      "Test circle -258 trained in 1.68 seconds.\n",
      "Test circle -257 trained in 1.68 seconds.\n",
      "Test circle -256 trained in 1.67 seconds.\n",
      "Test circle -255 trained in 1.72 seconds.\n",
      "Test circle -254 trained in 1.66 seconds.\n",
      "Test circle -253 trained in 1.69 seconds.\n",
      "Test circle -252 trained in 1.67 seconds.\n",
      "Test circle -251 trained in 1.72 seconds.\n",
      "Test circle -250 trained in 1.67 seconds.\n",
      "Test circle -249 trained in 1.77 seconds.\n",
      "Test circle -248 trained in 1.73 seconds.\n",
      "Test circle -247 trained in 1.78 seconds.\n",
      "Test circle -246 trained in 1.71 seconds.\n",
      "Test circle -245 trained in 1.69 seconds.\n",
      "Test circle -244 trained in 1.67 seconds.\n",
      "Test circle -243 trained in 1.7 seconds.\n",
      "Test circle -242 trained in 1.69 seconds.\n",
      "Test circle -241 trained in 1.77 seconds.\n",
      "Test circle -240 trained in 1.76 seconds.\n",
      "Test circle -239 trained in 1.72 seconds.\n",
      "Test circle -238 trained in 1.69 seconds.\n",
      "Test circle -237 trained in 1.69 seconds.\n",
      "Test circle -236 trained in 1.66 seconds.\n",
      "Test circle -235 trained in 1.7 seconds.\n",
      "Test circle -234 trained in 1.71 seconds.\n",
      "Test circle -233 trained in 1.69 seconds.\n",
      "Test circle -232 trained in 1.68 seconds.\n",
      "Test circle -231 trained in 1.72 seconds.\n",
      "Test circle -230 trained in 1.72 seconds.\n",
      "Test circle -229 trained in 1.77 seconds.\n",
      "Test circle -228 trained in 1.73 seconds.\n",
      "Test circle -227 trained in 1.68 seconds.\n",
      "Test circle -226 trained in 1.7 seconds.\n",
      "Test circle -225 trained in 1.68 seconds.\n",
      "Test circle -224 trained in 1.7 seconds.\n",
      "Test circle -223 trained in 1.7 seconds.\n",
      "Test circle -222 trained in 1.71 seconds.\n",
      "Test circle -221 trained in 1.67 seconds.\n",
      "Test circle -220 trained in 1.68 seconds.\n",
      "Test circle -219 trained in 1.67 seconds.\n",
      "Test circle -218 trained in 1.71 seconds.\n",
      "Test circle -217 trained in 1.68 seconds.\n",
      "Test circle -216 trained in 1.71 seconds.\n",
      "Test circle -215 trained in 1.68 seconds.\n",
      "Test circle -214 trained in 1.67 seconds.\n",
      "Test circle -213 trained in 1.7 seconds.\n",
      "Test circle -212 trained in 1.67 seconds.\n",
      "Test circle -211 trained in 1.7 seconds.\n",
      "Test circle -210 trained in 1.66 seconds.\n",
      "Test circle -209 trained in 1.72 seconds.\n",
      "Test circle -208 trained in 1.76 seconds.\n",
      "Test circle -207 trained in 1.68 seconds.\n",
      "Test circle -206 trained in 1.69 seconds.\n",
      "Test circle -205 trained in 1.67 seconds.\n",
      "Test circle -204 trained in 1.7 seconds.\n",
      "Test circle -203 trained in 1.66 seconds.\n",
      "Test circle -202 trained in 1.67 seconds.\n",
      "Test circle -201 trained in 1.67 seconds.\n",
      "Test circle -200 trained in 1.71 seconds.\n",
      "Test circle -199 trained in 1.68 seconds.\n",
      "Test circle -198 trained in 1.69 seconds.\n",
      "Test circle -197 trained in 1.68 seconds.\n",
      "Test circle -196 trained in 1.73 seconds.\n",
      "Test circle -195 trained in 1.71 seconds.\n",
      "Test circle -194 trained in 1.66 seconds.\n",
      "Test circle -193 trained in 1.69 seconds.\n",
      "Test circle -192 trained in 1.67 seconds.\n",
      "Test circle -191 trained in 1.68 seconds.\n",
      "Test circle -190 trained in 1.75 seconds.\n",
      "Test circle -189 trained in 1.69 seconds.\n",
      "Test circle -188 trained in 1.67 seconds.\n",
      "Test circle -187 trained in 1.77 seconds.\n",
      "Test circle -186 trained in 1.66 seconds.\n",
      "Test circle -185 trained in 1.72 seconds.\n",
      "Test circle -184 trained in 1.69 seconds.\n",
      "Test circle -183 trained in 1.69 seconds.\n",
      "Test circle -182 trained in 1.72 seconds.\n",
      "Test circle -181 trained in 1.71 seconds.\n",
      "Test circle -180 trained in 1.88 seconds.\n",
      "Test circle -179 trained in 1.68 seconds.\n",
      "Test circle -178 trained in 1.67 seconds.\n",
      "Test circle -177 trained in 1.66 seconds.\n",
      "Test circle -176 trained in 1.68 seconds.\n",
      "Test circle -175 trained in 1.67 seconds.\n",
      "Test circle -174 trained in 1.65 seconds.\n",
      "Test circle -173 trained in 1.7 seconds.\n",
      "Test circle -172 trained in 1.66 seconds.\n",
      "Test circle -171 trained in 1.68 seconds.\n",
      "Test circle -170 trained in 1.65 seconds.\n",
      "Test circle -169 trained in 1.69 seconds.\n",
      "Test circle -168 trained in 1.72 seconds.\n",
      "Test circle -167 trained in 1.68 seconds.\n",
      "Test circle -166 trained in 1.66 seconds.\n",
      "Test circle -165 trained in 1.68 seconds.\n",
      "Test circle -164 trained in 1.68 seconds.\n",
      "Test circle -163 trained in 1.68 seconds.\n",
      "Test circle -162 trained in 1.84 seconds.\n",
      "Test circle -161 trained in 1.73 seconds.\n",
      "Test circle -160 trained in 1.82 seconds.\n",
      "Test circle -159 trained in 1.68 seconds.\n",
      "Test circle -158 trained in 1.68 seconds.\n",
      "Test circle -157 trained in 1.67 seconds.\n",
      "Test circle -156 trained in 1.73 seconds.\n",
      "Test circle -155 trained in 1.7 seconds.\n",
      "Test circle -154 trained in 1.66 seconds.\n",
      "Test circle -153 trained in 1.77 seconds.\n",
      "Test circle -152 trained in 1.67 seconds.\n",
      "Test circle -151 trained in 1.67 seconds.\n",
      "Test circle -150 trained in 1.66 seconds.\n",
      "Test circle -149 trained in 1.67 seconds.\n",
      "Test circle -148 trained in 1.83 seconds.\n",
      "Test circle -147 trained in 1.67 seconds.\n",
      "Test circle -146 trained in 1.68 seconds.\n",
      "Test circle -145 trained in 1.74 seconds.\n",
      "Test circle -144 trained in 1.66 seconds.\n",
      "Test circle -143 trained in 1.67 seconds.\n",
      "Test circle -142 trained in 1.69 seconds.\n",
      "Test circle -141 trained in 1.66 seconds.\n",
      "Test circle -140 trained in 1.67 seconds.\n",
      "Test circle -139 trained in 1.66 seconds.\n",
      "Test circle -138 trained in 1.69 seconds.\n",
      "Test circle -137 trained in 1.65 seconds.\n",
      "Test circle -136 trained in 1.67 seconds.\n",
      "Test circle -135 trained in 1.66 seconds.\n",
      "Test circle -134 trained in 1.68 seconds.\n",
      "Test circle -133 trained in 1.65 seconds.\n",
      "Test circle -132 trained in 1.65 seconds.\n",
      "Test circle -131 trained in 1.66 seconds.\n",
      "Test circle -130 trained in 1.66 seconds.\n",
      "Test circle -129 trained in 1.68 seconds.\n",
      "Test circle -128 trained in 1.74 seconds.\n",
      "Test circle -127 trained in 1.67 seconds.\n",
      "Test circle -126 trained in 1.67 seconds.\n",
      "Test circle -125 trained in 1.69 seconds.\n",
      "Test circle -124 trained in 1.67 seconds.\n",
      "Test circle -123 trained in 1.68 seconds.\n",
      "Test circle -122 trained in 1.65 seconds.\n",
      "Test circle -121 trained in 1.67 seconds.\n",
      "Test circle -120 trained in 1.73 seconds.\n",
      "Test circle -119 trained in 1.67 seconds.\n",
      "Test circle -118 trained in 1.66 seconds.\n",
      "Test circle -117 trained in 1.67 seconds.\n",
      "Test circle -116 trained in 1.69 seconds.\n",
      "Test circle -115 trained in 1.67 seconds.\n",
      "Test circle -114 trained in 1.66 seconds.\n",
      "Test circle -113 trained in 1.68 seconds.\n",
      "Test circle -112 trained in 1.74 seconds.\n",
      "Test circle -111 trained in 1.67 seconds.\n",
      "Test circle -110 trained in 1.68 seconds.\n",
      "Test circle -109 trained in 1.67 seconds.\n",
      "Test circle -108 trained in 1.68 seconds.\n",
      "Test circle -107 trained in 1.65 seconds.\n",
      "Test circle -106 trained in 1.77 seconds.\n",
      "Test circle -105 trained in 1.65 seconds.\n",
      "Test circle -104 trained in 1.67 seconds.\n",
      "Test circle -103 trained in 1.65 seconds.\n",
      "Test circle -102 trained in 1.7 seconds.\n",
      "Test circle -101 trained in 1.65 seconds.\n",
      "Test circle -100 trained in 1.68 seconds.\n",
      "Test circle -99 trained in 1.67 seconds.\n",
      "Test circle -98 trained in 1.66 seconds.\n",
      "Test circle -97 trained in 1.66 seconds.\n",
      "Test circle -96 trained in 1.65 seconds.\n",
      "Test circle -95 trained in 1.67 seconds.\n",
      "Test circle -94 trained in 1.65 seconds.\n",
      "Test circle -93 trained in 1.73 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -92 trained in 1.67 seconds.\n",
      "Test circle -91 trained in 1.66 seconds.\n",
      "Test circle -90 trained in 1.65 seconds.\n",
      "Test circle -89 trained in 1.69 seconds.\n",
      "Test circle -88 trained in 1.7 seconds.\n",
      "Test circle -87 trained in 1.75 seconds.\n",
      "Test circle -86 trained in 1.65 seconds.\n",
      "Test circle -85 trained in 1.67 seconds.\n",
      "Test circle -84 trained in 1.65 seconds.\n",
      "Test circle -83 trained in 1.67 seconds.\n",
      "Test circle -82 trained in 1.68 seconds.\n",
      "Test circle -81 trained in 1.65 seconds.\n",
      "Test circle -80 trained in 1.68 seconds.\n",
      "Test circle -79 trained in 1.65 seconds.\n",
      "Test circle -78 trained in 1.67 seconds.\n",
      "Test circle -77 trained in 1.64 seconds.\n",
      "Test circle -76 trained in 1.67 seconds.\n",
      "Test circle -75 trained in 1.64 seconds.\n",
      "Test circle -74 trained in 1.69 seconds.\n",
      "Test circle -73 trained in 1.66 seconds.\n",
      "Test circle -72 trained in 1.67 seconds.\n",
      "Test circle -71 trained in 1.64 seconds.\n",
      "Test circle -70 trained in 1.68 seconds.\n",
      "Test circle -69 trained in 1.66 seconds.\n",
      "Test circle -68 trained in 1.74 seconds.\n",
      "Test circle -67 trained in 1.7 seconds.\n",
      "Test circle -66 trained in 1.66 seconds.\n",
      "Test circle -65 trained in 1.65 seconds.\n",
      "Test circle -64 trained in 1.66 seconds.\n",
      "Test circle -63 trained in 1.67 seconds.\n",
      "Test circle -62 trained in 1.65 seconds.\n",
      "Test circle -61 trained in 1.67 seconds.\n",
      "Test circle -60 trained in 1.65 seconds.\n",
      "Test circle -59 trained in 1.68 seconds.\n",
      "Test circle -58 trained in 1.65 seconds.\n",
      "Test circle -57 trained in 1.7 seconds.\n",
      "Test circle -56 trained in 1.68 seconds.\n",
      "Test circle -55 trained in 1.69 seconds.\n",
      "Test circle -54 trained in 1.66 seconds.\n",
      "Test circle -53 trained in 1.71 seconds.\n",
      "Test circle -52 trained in 1.66 seconds.\n",
      "Test circle -51 trained in 1.66 seconds.\n",
      "Test circle -50 trained in 1.66 seconds.\n",
      "Test circle -49 trained in 1.66 seconds.\n",
      "Test circle -48 trained in 1.75 seconds.\n",
      "Test circle -47 trained in 1.66 seconds.\n",
      "Test circle -46 trained in 1.67 seconds.\n",
      "Test circle -45 trained in 1.65 seconds.\n",
      "Test circle -44 trained in 1.71 seconds.\n",
      "Test circle -43 trained in 1.65 seconds.\n",
      "Test circle -42 trained in 1.67 seconds.\n",
      "Test circle -41 trained in 1.65 seconds.\n",
      "Test circle -40 trained in 1.75 seconds.\n",
      "Test circle -39 trained in 1.97 seconds.\n",
      "Test circle -38 trained in 1.69 seconds.\n",
      "Test circle -37 trained in 1.89 seconds.\n",
      "Test circle -36 trained in 1.73 seconds.\n",
      "Test circle -35 trained in 1.65 seconds.\n",
      "Test circle -34 trained in 1.68 seconds.\n",
      "Test circle -33 trained in 1.66 seconds.\n",
      "Test circle -32 trained in 1.69 seconds.\n",
      "Test circle -31 trained in 1.95 seconds.\n",
      "Test circle -30 trained in 1.65 seconds.\n",
      "Test circle -29 trained in 1.77 seconds.\n",
      "Test circle -28 trained in 1.65 seconds.\n",
      "Test circle -27 trained in 1.68 seconds.\n",
      "Test circle -26 trained in 1.64 seconds.\n",
      "Test circle -25 trained in 1.66 seconds.\n",
      "Test circle -24 trained in 1.82 seconds.\n",
      "Test circle -23 trained in 1.95 seconds.\n",
      "Test circle -22 trained in 1.68 seconds.\n",
      "Test circle -21 trained in 1.67 seconds.\n",
      "Test circle -20 trained in 1.69 seconds.\n",
      "Test circle -19 trained in 1.68 seconds.\n",
      "Test circle -18 trained in 1.75 seconds.\n",
      "Test circle -17 trained in 1.69 seconds.\n",
      "Test circle -16 trained in 1.68 seconds.\n",
      "Test circle -15 trained in 1.71 seconds.\n",
      "Test circle -14 trained in 1.71 seconds.\n",
      "Test circle -13 trained in 1.66 seconds.\n",
      "Test circle -12 trained in 1.65 seconds.\n",
      "Test circle -11 trained in 1.68 seconds.\n",
      "Test circle -10 trained in 1.66 seconds.\n",
      "Test circle -9 trained in 1.67 seconds.\n",
      "Test circle -8 trained in 1.74 seconds.\n",
      "Test circle -7 trained in 1.68 seconds.\n",
      "Test circle -6 trained in 1.67 seconds.\n",
      "Test circle -5 trained in 1.7 seconds.\n",
      "Test circle -4 trained in 1.67 seconds.\n",
      "Test circle -3 trained in 1.72 seconds.\n",
      "Test circle -2 trained in 1.72 seconds.\n",
      "Test circle -1 trained in 2.04 seconds.\n",
      "Model trained in 1224.08 seconds.\n",
      "Hyperparams (25, 3, 6) trained in 1224.08 seconds.\n",
      "=========================================================================================\n",
      "Model (25, 3, 6) Ended\n",
      "=========================================================================================\n",
      "=========================================================================================\n",
      "Model (25, 6, 3) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 2.01 seconds.\n",
      "Test circle -724 trained in 2.02 seconds.\n",
      "Test circle -723 trained in 2.05 seconds.\n",
      "Test circle -722 trained in 2.06 seconds.\n",
      "Test circle -721 trained in 2.02 seconds.\n",
      "Test circle -720 trained in 2.05 seconds.\n",
      "Test circle -719 trained in 2.05 seconds.\n",
      "Test circle -718 trained in 2.02 seconds.\n",
      "Test circle -717 trained in 2.01 seconds.\n",
      "Test circle -716 trained in 2.07 seconds.\n",
      "Test circle -715 trained in 2.05 seconds.\n",
      "Test circle -714 trained in 2.18 seconds.\n",
      "Test circle -713 trained in 2.08 seconds.\n",
      "Test circle -712 trained in 2.08 seconds.\n",
      "Test circle -711 trained in 2.09 seconds.\n",
      "Test circle -710 trained in 2.06 seconds.\n",
      "Test circle -709 trained in 2.09 seconds.\n",
      "Test circle -708 trained in 2.06 seconds.\n",
      "Test circle -707 trained in 2.07 seconds.\n",
      "Test circle -706 trained in 2.08 seconds.\n",
      "Test circle -705 trained in 2.08 seconds.\n",
      "Test circle -704 trained in 2.05 seconds.\n",
      "Test circle -703 trained in 2.08 seconds.\n",
      "Test circle -702 trained in 2.06 seconds.\n",
      "Test circle -701 trained in 2.08 seconds.\n",
      "Test circle -700 trained in 2.06 seconds.\n",
      "Test circle -699 trained in 2.09 seconds.\n",
      "Test circle -698 trained in 2.09 seconds.\n",
      "Test circle -697 trained in 2.08 seconds.\n",
      "Test circle -696 trained in 2.07 seconds.\n",
      "Test circle -695 trained in 2.06 seconds.\n",
      "Test circle -694 trained in 2.19 seconds.\n",
      "Test circle -693 trained in 2.08 seconds.\n",
      "Test circle -692 trained in 2.07 seconds.\n",
      "Test circle -691 trained in 2.08 seconds.\n",
      "Test circle -690 trained in 2.07 seconds.\n",
      "Test circle -689 trained in 2.09 seconds.\n",
      "Test circle -688 trained in 2.06 seconds.\n",
      "Test circle -687 trained in 2.09 seconds.\n",
      "Test circle -686 trained in 2.05 seconds.\n",
      "Test circle -685 trained in 2.03 seconds.\n",
      "Test circle -684 trained in 2.01 seconds.\n",
      "Test circle -683 trained in 2.01 seconds.\n",
      "Test circle -682 trained in 2.06 seconds.\n",
      "Test circle -681 trained in 2.04 seconds.\n",
      "Test circle -680 trained in 2.09 seconds.\n",
      "Test circle -679 trained in 2.03 seconds.\n",
      "Test circle -678 trained in 2.03 seconds.\n",
      "Test circle -677 trained in 2.04 seconds.\n",
      "Test circle -676 trained in 2.07 seconds.\n",
      "Test circle -675 trained in 2.15 seconds.\n",
      "Test circle -674 trained in 2.06 seconds.\n",
      "Test circle -673 trained in 2.06 seconds.\n",
      "Test circle -672 trained in 2.05 seconds.\n",
      "Test circle -671 trained in 2.07 seconds.\n",
      "Test circle -670 trained in 2.08 seconds.\n",
      "Test circle -669 trained in 2.14 seconds.\n",
      "Test circle -668 trained in 2.24 seconds.\n",
      "Test circle -667 trained in 2.04 seconds.\n",
      "Test circle -666 trained in 2.09 seconds.\n",
      "Test circle -665 trained in 2.1 seconds.\n",
      "Test circle -664 trained in 2.09 seconds.\n",
      "Test circle -663 trained in 2.06 seconds.\n",
      "Test circle -662 trained in 2.09 seconds.\n",
      "Test circle -661 trained in 2.26 seconds.\n",
      "Test circle -660 trained in 2.26 seconds.\n",
      "Test circle -659 trained in 2.22 seconds.\n",
      "Test circle -658 trained in 2.24 seconds.\n",
      "Test circle -657 trained in 2.24 seconds.\n",
      "Test circle -656 trained in 2.25 seconds.\n",
      "Test circle -655 trained in 2.2 seconds.\n",
      "Test circle -654 trained in 2.11 seconds.\n",
      "Test circle -653 trained in 2.15 seconds.\n",
      "Test circle -652 trained in 2.14 seconds.\n",
      "Test circle -651 trained in 2.11 seconds.\n",
      "Test circle -650 trained in 2.11 seconds.\n",
      "Test circle -649 trained in 2.12 seconds.\n",
      "Test circle -648 trained in 2.13 seconds.\n",
      "Test circle -647 trained in 2.13 seconds.\n",
      "Test circle -646 trained in 2.13 seconds.\n",
      "Test circle -645 trained in 2.16 seconds.\n",
      "Test circle -644 trained in 2.37 seconds.\n",
      "Test circle -643 trained in 2.47 seconds.\n",
      "Test circle -642 trained in 2.14 seconds.\n",
      "Test circle -641 trained in 2.19 seconds.\n",
      "Test circle -640 trained in 2.28 seconds.\n",
      "Test circle -639 trained in 1.85 seconds.\n",
      "Test circle -638 trained in 1.65 seconds.\n",
      "Test circle -637 trained in 1.61 seconds.\n",
      "Test circle -636 trained in 1.59 seconds.\n",
      "Test circle -635 trained in 1.69 seconds.\n",
      "Test circle -634 trained in 1.59 seconds.\n",
      "Test circle -633 trained in 1.61 seconds.\n",
      "Test circle -632 trained in 1.62 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -631 trained in 1.6 seconds.\n",
      "Test circle -630 trained in 1.6 seconds.\n",
      "Test circle -629 trained in 1.61 seconds.\n",
      "Test circle -628 trained in 1.6 seconds.\n",
      "Test circle -627 trained in 1.61 seconds.\n",
      "Test circle -626 trained in 1.61 seconds.\n",
      "Test circle -625 trained in 1.61 seconds.\n",
      "Test circle -624 trained in 1.6 seconds.\n",
      "Test circle -623 trained in 1.6 seconds.\n",
      "Test circle -622 trained in 1.63 seconds.\n",
      "Test circle -621 trained in 1.66 seconds.\n",
      "Test circle -620 trained in 1.61 seconds.\n",
      "Test circle -619 trained in 1.61 seconds.\n",
      "Test circle -618 trained in 1.61 seconds.\n",
      "Test circle -617 trained in 1.61 seconds.\n",
      "Test circle -616 trained in 1.69 seconds.\n",
      "Test circle -615 trained in 1.61 seconds.\n",
      "Test circle -614 trained in 1.61 seconds.\n",
      "Test circle -613 trained in 1.61 seconds.\n",
      "Test circle -612 trained in 1.6 seconds.\n",
      "Test circle -611 trained in 1.6 seconds.\n",
      "Test circle -610 trained in 1.6 seconds.\n",
      "Test circle -609 trained in 1.6 seconds.\n",
      "Test circle -608 trained in 1.61 seconds.\n",
      "Test circle -607 trained in 1.63 seconds.\n",
      "Test circle -606 trained in 1.65 seconds.\n",
      "Test circle -605 trained in 1.61 seconds.\n",
      "Test circle -604 trained in 1.6 seconds.\n",
      "Test circle -603 trained in 1.61 seconds.\n",
      "Test circle -602 trained in 1.61 seconds.\n",
      "Test circle -601 trained in 1.6 seconds.\n",
      "Test circle -600 trained in 1.61 seconds.\n",
      "Test circle -599 trained in 1.6 seconds.\n",
      "Test circle -598 trained in 1.62 seconds.\n",
      "Test circle -597 trained in 1.69 seconds.\n",
      "Test circle -596 trained in 1.62 seconds.\n",
      "Test circle -595 trained in 1.61 seconds.\n",
      "Test circle -594 trained in 1.6 seconds.\n",
      "Test circle -593 trained in 1.61 seconds.\n",
      "Test circle -592 trained in 1.61 seconds.\n",
      "Test circle -591 trained in 1.6 seconds.\n",
      "Test circle -590 trained in 1.61 seconds.\n",
      "Test circle -589 trained in 1.61 seconds.\n",
      "Test circle -588 trained in 1.6 seconds.\n",
      "Test circle -587 trained in 1.62 seconds.\n",
      "Test circle -586 trained in 1.6 seconds.\n",
      "Test circle -585 trained in 1.62 seconds.\n",
      "Test circle -584 trained in 1.62 seconds.\n",
      "Test circle -583 trained in 1.62 seconds.\n",
      "Test circle -582 trained in 1.61 seconds.\n",
      "Test circle -581 trained in 1.6 seconds.\n",
      "Test circle -580 trained in 1.6 seconds.\n",
      "Test circle -579 trained in 1.6 seconds.\n",
      "Test circle -578 trained in 1.7 seconds.\n",
      "Test circle -577 trained in 1.62 seconds.\n",
      "Test circle -576 trained in 1.6 seconds.\n",
      "Test circle -575 trained in 1.61 seconds.\n",
      "Test circle -574 trained in 1.61 seconds.\n",
      "Test circle -573 trained in 1.61 seconds.\n",
      "Test circle -572 trained in 1.66 seconds.\n",
      "Test circle -571 trained in 1.93 seconds.\n",
      "Test circle -570 trained in 1.97 seconds.\n",
      "Test circle -569 trained in 1.94 seconds.\n",
      "Test circle -568 trained in 1.78 seconds.\n",
      "Test circle -567 trained in 1.62 seconds.\n",
      "Test circle -566 trained in 1.62 seconds.\n",
      "Test circle -565 trained in 1.62 seconds.\n",
      "Test circle -564 trained in 1.61 seconds.\n",
      "Test circle -563 trained in 1.61 seconds.\n",
      "Test circle -562 trained in 1.6 seconds.\n",
      "Test circle -561 trained in 1.61 seconds.\n",
      "Test circle -560 trained in 1.63 seconds.\n",
      "Test circle -559 trained in 1.61 seconds.\n",
      "Test circle -558 trained in 1.6 seconds.\n",
      "Test circle -557 trained in 1.68 seconds.\n",
      "Test circle -556 trained in 1.6 seconds.\n",
      "Test circle -555 trained in 1.61 seconds.\n",
      "Test circle -554 trained in 1.62 seconds.\n",
      "Test circle -553 trained in 1.6 seconds.\n",
      "Test circle -552 trained in 1.62 seconds.\n",
      "Test circle -551 trained in 1.6 seconds.\n",
      "Test circle -550 trained in 1.61 seconds.\n",
      "Test circle -549 trained in 1.61 seconds.\n",
      "Test circle -548 trained in 1.65 seconds.\n",
      "Test circle -547 trained in 1.63 seconds.\n",
      "Test circle -546 trained in 1.61 seconds.\n",
      "Test circle -545 trained in 1.6 seconds.\n",
      "Test circle -544 trained in 1.6 seconds.\n",
      "Test circle -543 trained in 1.6 seconds.\n",
      "Test circle -542 trained in 1.62 seconds.\n",
      "Test circle -541 trained in 1.61 seconds.\n",
      "Test circle -540 trained in 1.6 seconds.\n",
      "Test circle -539 trained in 1.6 seconds.\n",
      "Test circle -538 trained in 1.69 seconds.\n",
      "Test circle -537 trained in 1.6 seconds.\n",
      "Test circle -536 trained in 1.6 seconds.\n",
      "Test circle -535 trained in 1.61 seconds.\n",
      "Test circle -534 trained in 1.6 seconds.\n",
      "Test circle -533 trained in 1.61 seconds.\n",
      "Test circle -532 trained in 1.65 seconds.\n",
      "Test circle -531 trained in 1.61 seconds.\n",
      "Test circle -530 trained in 1.6 seconds.\n",
      "Test circle -529 trained in 1.61 seconds.\n",
      "Test circle -528 trained in 1.61 seconds.\n",
      "Test circle -527 trained in 1.6 seconds.\n",
      "Test circle -526 trained in 1.61 seconds.\n",
      "Test circle -525 trained in 1.6 seconds.\n",
      "Test circle -524 trained in 1.6 seconds.\n",
      "Test circle -523 trained in 1.6 seconds.\n",
      "Test circle -522 trained in 1.6 seconds.\n",
      "Test circle -521 trained in 1.6 seconds.\n",
      "Test circle -520 trained in 1.6 seconds.\n",
      "Test circle -519 trained in 1.6 seconds.\n",
      "Test circle -518 trained in 1.6 seconds.\n",
      "Test circle -517 trained in 1.68 seconds.\n",
      "Test circle -516 trained in 1.62 seconds.\n",
      "Test circle -515 trained in 1.6 seconds.\n",
      "Test circle -514 trained in 1.6 seconds.\n",
      "Test circle -513 trained in 1.63 seconds.\n",
      "Test circle -512 trained in 1.6 seconds.\n",
      "Test circle -511 trained in 1.61 seconds.\n",
      "Test circle -510 trained in 1.62 seconds.\n",
      "Test circle -509 trained in 1.61 seconds.\n",
      "Test circle -508 trained in 1.6 seconds.\n",
      "Test circle -507 trained in 1.6 seconds.\n",
      "Test circle -506 trained in 1.6 seconds.\n",
      "Test circle -505 trained in 1.61 seconds.\n",
      "Test circle -504 trained in 1.6 seconds.\n",
      "Test circle -503 trained in 1.6 seconds.\n",
      "Test circle -502 trained in 1.6 seconds.\n",
      "Test circle -501 trained in 1.6 seconds.\n",
      "Test circle -500 trained in 1.59 seconds.\n",
      "Test circle -499 trained in 1.6 seconds.\n",
      "Test circle -498 trained in 1.68 seconds.\n",
      "Test circle -497 trained in 1.61 seconds.\n",
      "Test circle -496 trained in 1.6 seconds.\n",
      "Test circle -495 trained in 1.66 seconds.\n",
      "Test circle -494 trained in 1.61 seconds.\n",
      "Test circle -493 trained in 1.59 seconds.\n",
      "Test circle -492 trained in 1.61 seconds.\n",
      "Test circle -491 trained in 1.65 seconds.\n",
      "Test circle -490 trained in 1.61 seconds.\n",
      "Test circle -489 trained in 1.6 seconds.\n",
      "Test circle -488 trained in 1.61 seconds.\n",
      "Test circle -487 trained in 1.59 seconds.\n",
      "Test circle -486 trained in 1.61 seconds.\n",
      "Test circle -485 trained in 1.6 seconds.\n",
      "Test circle -484 trained in 1.61 seconds.\n",
      "Test circle -483 trained in 1.6 seconds.\n",
      "Test circle -482 trained in 1.6 seconds.\n",
      "Test circle -481 trained in 1.61 seconds.\n",
      "Test circle -480 trained in 1.6 seconds.\n",
      "Test circle -479 trained in 1.6 seconds.\n",
      "Test circle -478 trained in 1.69 seconds.\n",
      "Test circle -477 trained in 1.61 seconds.\n",
      "Test circle -476 trained in 1.67 seconds.\n",
      "Test circle -475 trained in 1.61 seconds.\n",
      "Test circle -474 trained in 1.62 seconds.\n",
      "Test circle -473 trained in 1.65 seconds.\n",
      "Test circle -472 trained in 1.61 seconds.\n",
      "Test circle -471 trained in 1.61 seconds.\n",
      "Test circle -470 trained in 1.6 seconds.\n",
      "Test circle -469 trained in 1.6 seconds.\n",
      "Test circle -468 trained in 1.61 seconds.\n",
      "Test circle -467 trained in 1.6 seconds.\n",
      "Test circle -466 trained in 1.6 seconds.\n",
      "Test circle -465 trained in 1.6 seconds.\n",
      "Test circle -464 trained in 1.59 seconds.\n",
      "Test circle -463 trained in 1.6 seconds.\n",
      "Test circle -462 trained in 1.59 seconds.\n",
      "Test circle -461 trained in 1.6 seconds.\n",
      "Test circle -460 trained in 1.6 seconds.\n",
      "Test circle -459 trained in 1.69 seconds.\n",
      "Test circle -458 trained in 1.75 seconds.\n",
      "Test circle -457 trained in 1.63 seconds.\n",
      "Test circle -456 trained in 1.6 seconds.\n",
      "Test circle -455 trained in 1.61 seconds.\n",
      "Test circle -454 trained in 1.61 seconds.\n",
      "Test circle -453 trained in 1.6 seconds.\n",
      "Test circle -452 trained in 1.6 seconds.\n",
      "Test circle -451 trained in 1.6 seconds.\n",
      "Test circle -450 trained in 1.61 seconds.\n",
      "Test circle -449 trained in 1.59 seconds.\n",
      "Test circle -448 trained in 1.61 seconds.\n",
      "Test circle -447 trained in 1.6 seconds.\n",
      "Test circle -446 trained in 1.6 seconds.\n",
      "Test circle -445 trained in 1.59 seconds.\n",
      "Test circle -444 trained in 1.61 seconds.\n",
      "Test circle -443 trained in 1.6 seconds.\n",
      "Test circle -442 trained in 1.6 seconds.\n",
      "Test circle -441 trained in 1.59 seconds.\n",
      "Test circle -440 trained in 1.6 seconds.\n",
      "Test circle -439 trained in 1.69 seconds.\n",
      "Test circle -438 trained in 1.63 seconds.\n",
      "Test circle -437 trained in 1.63 seconds.\n",
      "Test circle -436 trained in 1.61 seconds.\n",
      "Test circle -435 trained in 1.61 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -434 trained in 1.62 seconds.\n",
      "Test circle -433 trained in 1.61 seconds.\n",
      "Test circle -432 trained in 1.6 seconds.\n",
      "Test circle -431 trained in 1.61 seconds.\n",
      "Test circle -430 trained in 1.6 seconds.\n",
      "Test circle -429 trained in 1.61 seconds.\n",
      "Test circle -428 trained in 1.61 seconds.\n",
      "Test circle -427 trained in 1.6 seconds.\n",
      "Test circle -426 trained in 1.6 seconds.\n",
      "Test circle -425 trained in 1.6 seconds.\n",
      "Test circle -424 trained in 1.6 seconds.\n",
      "Test circle -423 trained in 1.6 seconds.\n",
      "Test circle -422 trained in 1.6 seconds.\n",
      "Test circle -421 trained in 1.62 seconds.\n",
      "Test circle -420 trained in 1.61 seconds.\n",
      "Test circle -419 trained in 1.68 seconds.\n",
      "Test circle -418 trained in 1.61 seconds.\n",
      "Test circle -417 trained in 1.61 seconds.\n",
      "Test circle -416 trained in 1.61 seconds.\n",
      "Test circle -415 trained in 1.6 seconds.\n",
      "Test circle -414 trained in 1.61 seconds.\n",
      "Test circle -413 trained in 1.6 seconds.\n",
      "Test circle -412 trained in 1.6 seconds.\n",
      "Test circle -411 trained in 1.61 seconds.\n",
      "Test circle -410 trained in 1.61 seconds.\n",
      "Test circle -409 trained in 1.6 seconds.\n",
      "Test circle -408 trained in 1.61 seconds.\n",
      "Test circle -407 trained in 1.6 seconds.\n",
      "Test circle -406 trained in 1.6 seconds.\n",
      "Test circle -405 trained in 1.6 seconds.\n",
      "Test circle -404 trained in 1.61 seconds.\n",
      "Test circle -403 trained in 1.6 seconds.\n",
      "Test circle -402 trained in 1.61 seconds.\n",
      "Test circle -401 trained in 1.6 seconds.\n",
      "Test circle -400 trained in 1.69 seconds.\n",
      "Test circle -399 trained in 1.66 seconds.\n",
      "Test circle -398 trained in 1.61 seconds.\n",
      "Test circle -397 trained in 1.61 seconds.\n",
      "Test circle -396 trained in 1.61 seconds.\n",
      "Test circle -395 trained in 1.61 seconds.\n",
      "Test circle -394 trained in 1.6 seconds.\n",
      "Test circle -393 trained in 1.6 seconds.\n",
      "Test circle -392 trained in 1.6 seconds.\n",
      "Test circle -391 trained in 1.6 seconds.\n",
      "Test circle -390 trained in 1.61 seconds.\n",
      "Test circle -389 trained in 1.61 seconds.\n",
      "Test circle -388 trained in 1.6 seconds.\n",
      "Test circle -387 trained in 1.6 seconds.\n",
      "Test circle -386 trained in 1.6 seconds.\n",
      "Test circle -385 trained in 1.61 seconds.\n",
      "Test circle -384 trained in 1.62 seconds.\n",
      "Test circle -383 trained in 1.64 seconds.\n",
      "Test circle -382 trained in 1.61 seconds.\n",
      "Test circle -381 trained in 1.6 seconds.\n",
      "Test circle -380 trained in 1.69 seconds.\n",
      "Test circle -379 trained in 1.61 seconds.\n",
      "Test circle -378 trained in 1.61 seconds.\n",
      "Test circle -377 trained in 1.6 seconds.\n",
      "Test circle -376 trained in 1.6 seconds.\n",
      "Test circle -375 trained in 1.6 seconds.\n",
      "Test circle -374 trained in 1.61 seconds.\n",
      "Test circle -373 trained in 1.61 seconds.\n",
      "Test circle -372 trained in 1.61 seconds.\n",
      "Test circle -371 trained in 1.65 seconds.\n",
      "Test circle -370 trained in 1.6 seconds.\n",
      "Test circle -369 trained in 1.59 seconds.\n",
      "Test circle -368 trained in 1.61 seconds.\n",
      "Test circle -367 trained in 1.6 seconds.\n",
      "Test circle -366 trained in 1.61 seconds.\n",
      "Test circle -365 trained in 1.6 seconds.\n",
      "Test circle -364 trained in 1.61 seconds.\n",
      "Test circle -363 trained in 1.6 seconds.\n",
      "Test circle -362 trained in 1.6 seconds.\n",
      "Test circle -361 trained in 1.62 seconds.\n",
      "Test circle -360 trained in 1.68 seconds.\n",
      "Test circle -359 trained in 1.62 seconds.\n",
      "Test circle -358 trained in 1.61 seconds.\n",
      "Test circle -357 trained in 1.6 seconds.\n",
      "Test circle -356 trained in 1.62 seconds.\n",
      "Test circle -355 trained in 1.6 seconds.\n",
      "Test circle -354 trained in 1.6 seconds.\n",
      "Test circle -353 trained in 1.6 seconds.\n",
      "Test circle -352 trained in 1.6 seconds.\n",
      "Test circle -351 trained in 1.6 seconds.\n",
      "Test circle -350 trained in 1.6 seconds.\n",
      "Test circle -349 trained in 1.6 seconds.\n",
      "Test circle -348 trained in 1.61 seconds.\n",
      "Test circle -347 trained in 1.6 seconds.\n",
      "Test circle -346 trained in 1.65 seconds.\n",
      "Test circle -345 trained in 1.6 seconds.\n",
      "Test circle -344 trained in 1.6 seconds.\n",
      "Test circle -343 trained in 1.61 seconds.\n",
      "Test circle -342 trained in 1.6 seconds.\n",
      "Test circle -341 trained in 1.6 seconds.\n",
      "Test circle -340 trained in 1.68 seconds.\n",
      "Test circle -339 trained in 1.6 seconds.\n",
      "Test circle -338 trained in 1.6 seconds.\n",
      "Test circle -337 trained in 1.6 seconds.\n",
      "Test circle -336 trained in 1.6 seconds.\n",
      "Test circle -335 trained in 1.6 seconds.\n",
      "Test circle -334 trained in 1.6 seconds.\n",
      "Test circle -333 trained in 1.6 seconds.\n",
      "Test circle -332 trained in 1.61 seconds.\n",
      "Test circle -331 trained in 1.6 seconds.\n",
      "Test circle -330 trained in 1.6 seconds.\n",
      "Test circle -329 trained in 1.6 seconds.\n",
      "Test circle -328 trained in 1.61 seconds.\n",
      "Test circle -327 trained in 1.75 seconds.\n",
      "Test circle -326 trained in 1.64 seconds.\n",
      "Test circle -325 trained in 1.62 seconds.\n",
      "Test circle -324 trained in 1.65 seconds.\n",
      "Test circle -323 trained in 1.6 seconds.\n",
      "Test circle -322 trained in 1.61 seconds.\n",
      "Test circle -321 trained in 1.6 seconds.\n",
      "Test circle -320 trained in 1.6 seconds.\n",
      "Test circle -319 trained in 1.68 seconds.\n",
      "Test circle -318 trained in 1.6 seconds.\n",
      "Test circle -317 trained in 1.6 seconds.\n",
      "Test circle -316 trained in 1.6 seconds.\n",
      "Test circle -315 trained in 1.6 seconds.\n",
      "Test circle -314 trained in 1.59 seconds.\n",
      "Test circle -313 trained in 1.6 seconds.\n",
      "Test circle -312 trained in 1.6 seconds.\n",
      "Test circle -311 trained in 1.6 seconds.\n",
      "Test circle -310 trained in 1.6 seconds.\n",
      "Test circle -309 trained in 1.65 seconds.\n",
      "Test circle -308 trained in 1.6 seconds.\n",
      "Test circle -307 trained in 1.6 seconds.\n",
      "Test circle -306 trained in 1.61 seconds.\n",
      "Test circle -305 trained in 1.61 seconds.\n",
      "Test circle -304 trained in 1.59 seconds.\n",
      "Test circle -303 trained in 1.6 seconds.\n",
      "Test circle -302 trained in 1.61 seconds.\n",
      "Test circle -301 trained in 1.59 seconds.\n",
      "Test circle -300 trained in 1.6 seconds.\n",
      "Test circle -299 trained in 1.68 seconds.\n",
      "Test circle -298 trained in 1.6 seconds.\n",
      "Test circle -297 trained in 1.6 seconds.\n",
      "Test circle -296 trained in 1.6 seconds.\n",
      "Test circle -295 trained in 1.61 seconds.\n",
      "Test circle -294 trained in 1.6 seconds.\n",
      "Test circle -293 trained in 1.61 seconds.\n",
      "Test circle -292 trained in 1.6 seconds.\n",
      "Test circle -291 trained in 1.6 seconds.\n",
      "Test circle -290 trained in 1.6 seconds.\n",
      "Test circle -289 trained in 1.62 seconds.\n",
      "Test circle -288 trained in 1.62 seconds.\n",
      "Test circle -287 trained in 1.62 seconds.\n",
      "Test circle -286 trained in 1.6 seconds.\n",
      "Test circle -285 trained in 1.61 seconds.\n",
      "Test circle -284 trained in 1.6 seconds.\n",
      "Test circle -283 trained in 1.6 seconds.\n",
      "Test circle -282 trained in 1.6 seconds.\n",
      "Test circle -281 trained in 1.6 seconds.\n",
      "Test circle -280 trained in 1.69 seconds.\n",
      "Test circle -279 trained in 1.6 seconds.\n",
      "Test circle -278 trained in 1.6 seconds.\n",
      "Test circle -277 trained in 1.6 seconds.\n",
      "Test circle -276 trained in 1.6 seconds.\n",
      "Test circle -275 trained in 1.6 seconds.\n",
      "Test circle -274 trained in 1.6 seconds.\n",
      "Test circle -273 trained in 1.6 seconds.\n",
      "Test circle -272 trained in 1.64 seconds.\n",
      "Test circle -271 trained in 1.61 seconds.\n",
      "Test circle -270 trained in 1.6 seconds.\n",
      "Test circle -269 trained in 1.6 seconds.\n",
      "Test circle -268 trained in 1.6 seconds.\n",
      "Test circle -267 trained in 1.61 seconds.\n",
      "Test circle -266 trained in 1.6 seconds.\n",
      "Test circle -265 trained in 1.6 seconds.\n",
      "Test circle -264 trained in 1.6 seconds.\n",
      "Test circle -263 trained in 1.6 seconds.\n",
      "Test circle -262 trained in 1.61 seconds.\n",
      "Test circle -261 trained in 1.6 seconds.\n",
      "Test circle -260 trained in 1.68 seconds.\n",
      "Test circle -259 trained in 1.6 seconds.\n",
      "Test circle -258 trained in 1.6 seconds.\n",
      "Test circle -257 trained in 1.61 seconds.\n",
      "Test circle -256 trained in 1.6 seconds.\n",
      "Test circle -255 trained in 1.61 seconds.\n",
      "Test circle -254 trained in 1.6 seconds.\n",
      "Test circle -253 trained in 1.6 seconds.\n",
      "Test circle -252 trained in 1.61 seconds.\n",
      "Test circle -251 trained in 1.61 seconds.\n",
      "Test circle -250 trained in 1.67 seconds.\n",
      "Test circle -249 trained in 1.62 seconds.\n",
      "Test circle -248 trained in 1.6 seconds.\n",
      "Test circle -247 trained in 1.61 seconds.\n",
      "Test circle -246 trained in 1.6 seconds.\n",
      "Test circle -245 trained in 1.6 seconds.\n",
      "Test circle -244 trained in 1.6 seconds.\n",
      "Test circle -243 trained in 1.61 seconds.\n",
      "Test circle -242 trained in 1.61 seconds.\n",
      "Test circle -241 trained in 1.61 seconds.\n",
      "Test circle -240 trained in 1.6 seconds.\n",
      "Test circle -239 trained in 1.68 seconds.\n",
      "Test circle -238 trained in 1.61 seconds.\n",
      "Test circle -237 trained in 1.6 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -236 trained in 1.59 seconds.\n",
      "Test circle -235 trained in 1.62 seconds.\n",
      "Test circle -234 trained in 1.62 seconds.\n",
      "Test circle -233 trained in 1.61 seconds.\n",
      "Test circle -232 trained in 1.6 seconds.\n",
      "Test circle -231 trained in 1.59 seconds.\n",
      "Test circle -230 trained in 1.61 seconds.\n",
      "Test circle -229 trained in 1.61 seconds.\n",
      "Test circle -228 trained in 1.6 seconds.\n",
      "Test circle -227 trained in 1.61 seconds.\n",
      "Test circle -226 trained in 1.6 seconds.\n",
      "Test circle -225 trained in 1.6 seconds.\n",
      "Test circle -224 trained in 1.61 seconds.\n",
      "Test circle -223 trained in 1.6 seconds.\n",
      "Test circle -222 trained in 1.6 seconds.\n",
      "Test circle -221 trained in 1.6 seconds.\n",
      "Test circle -220 trained in 1.69 seconds.\n",
      "Test circle -219 trained in 1.6 seconds.\n",
      "Test circle -218 trained in 1.61 seconds.\n",
      "Test circle -217 trained in 1.61 seconds.\n",
      "Test circle -216 trained in 1.61 seconds.\n",
      "Test circle -215 trained in 1.6 seconds.\n",
      "Test circle -214 trained in 1.6 seconds.\n",
      "Test circle -213 trained in 1.62 seconds.\n",
      "Test circle -212 trained in 1.62 seconds.\n",
      "Test circle -211 trained in 1.61 seconds.\n",
      "Test circle -210 trained in 1.62 seconds.\n",
      "Test circle -209 trained in 1.61 seconds.\n",
      "Test circle -208 trained in 1.59 seconds.\n",
      "Test circle -207 trained in 1.61 seconds.\n",
      "Test circle -206 trained in 1.6 seconds.\n",
      "Test circle -205 trained in 1.6 seconds.\n",
      "Test circle -204 trained in 1.6 seconds.\n",
      "Test circle -203 trained in 1.6 seconds.\n",
      "Test circle -202 trained in 1.6 seconds.\n",
      "Test circle -201 trained in 1.6 seconds.\n",
      "Test circle -200 trained in 1.7 seconds.\n",
      "Test circle -199 trained in 1.6 seconds.\n",
      "Test circle -198 trained in 1.61 seconds.\n",
      "Test circle -197 trained in 1.65 seconds.\n",
      "Test circle -196 trained in 1.6 seconds.\n",
      "Test circle -195 trained in 1.62 seconds.\n",
      "Test circle -194 trained in 1.61 seconds.\n",
      "Test circle -193 trained in 1.61 seconds.\n",
      "Test circle -192 trained in 1.6 seconds.\n",
      "Test circle -191 trained in 1.61 seconds.\n",
      "Test circle -190 trained in 1.6 seconds.\n",
      "Test circle -189 trained in 1.6 seconds.\n",
      "Test circle -188 trained in 1.62 seconds.\n",
      "Test circle -187 trained in 1.61 seconds.\n",
      "Test circle -186 trained in 1.6 seconds.\n",
      "Test circle -185 trained in 1.6 seconds.\n",
      "Test circle -184 trained in 1.6 seconds.\n",
      "Test circle -183 trained in 1.59 seconds.\n",
      "Test circle -182 trained in 1.6 seconds.\n",
      "Test circle -181 trained in 1.68 seconds.\n",
      "Test circle -180 trained in 1.61 seconds.\n",
      "Test circle -179 trained in 1.6 seconds.\n",
      "Test circle -178 trained in 1.62 seconds.\n",
      "Test circle -177 trained in 1.61 seconds.\n",
      "Test circle -176 trained in 1.63 seconds.\n",
      "Test circle -175 trained in 1.66 seconds.\n",
      "Test circle -174 trained in 1.6 seconds.\n",
      "Test circle -173 trained in 1.6 seconds.\n",
      "Test circle -172 trained in 1.64 seconds.\n",
      "Test circle -171 trained in 1.62 seconds.\n",
      "Test circle -170 trained in 1.62 seconds.\n",
      "Test circle -169 trained in 1.61 seconds.\n",
      "Test circle -168 trained in 1.62 seconds.\n",
      "Test circle -167 trained in 1.6 seconds.\n",
      "Test circle -166 trained in 1.6 seconds.\n",
      "Test circle -165 trained in 1.6 seconds.\n",
      "Test circle -164 trained in 1.6 seconds.\n",
      "Test circle -163 trained in 1.59 seconds.\n",
      "Test circle -162 trained in 1.6 seconds.\n",
      "Test circle -161 trained in 1.68 seconds.\n",
      "Test circle -160 trained in 1.65 seconds.\n",
      "Test circle -159 trained in 1.6 seconds.\n",
      "Test circle -158 trained in 1.61 seconds.\n",
      "Test circle -157 trained in 1.61 seconds.\n",
      "Test circle -156 trained in 1.61 seconds.\n",
      "Test circle -155 trained in 1.6 seconds.\n",
      "Test circle -154 trained in 1.6 seconds.\n",
      "Test circle -153 trained in 1.6 seconds.\n",
      "Test circle -152 trained in 1.6 seconds.\n",
      "Test circle -151 trained in 1.6 seconds.\n",
      "Test circle -150 trained in 1.61 seconds.\n",
      "Test circle -149 trained in 1.6 seconds.\n",
      "Test circle -148 trained in 1.61 seconds.\n",
      "Test circle -147 trained in 1.6 seconds.\n",
      "Test circle -146 trained in 1.6 seconds.\n",
      "Test circle -145 trained in 1.6 seconds.\n",
      "Test circle -144 trained in 1.6 seconds.\n",
      "Test circle -143 trained in 1.6 seconds.\n",
      "Test circle -142 trained in 1.6 seconds.\n",
      "Test circle -141 trained in 1.69 seconds.\n",
      "Test circle -140 trained in 1.6 seconds.\n",
      "Test circle -139 trained in 1.63 seconds.\n",
      "Test circle -138 trained in 1.62 seconds.\n",
      "Test circle -137 trained in 1.6 seconds.\n",
      "Test circle -136 trained in 1.61 seconds.\n",
      "Test circle -135 trained in 1.61 seconds.\n",
      "Test circle -134 trained in 1.6 seconds.\n",
      "Test circle -133 trained in 1.61 seconds.\n",
      "Test circle -132 trained in 1.6 seconds.\n",
      "Test circle -131 trained in 1.61 seconds.\n",
      "Test circle -130 trained in 1.6 seconds.\n",
      "Test circle -129 trained in 1.6 seconds.\n",
      "Test circle -128 trained in 1.6 seconds.\n",
      "Test circle -127 trained in 1.6 seconds.\n",
      "Test circle -126 trained in 1.6 seconds.\n",
      "Test circle -125 trained in 1.6 seconds.\n",
      "Test circle -124 trained in 1.59 seconds.\n",
      "Test circle -123 trained in 1.63 seconds.\n",
      "Test circle -122 trained in 1.68 seconds.\n",
      "Test circle -121 trained in 1.6 seconds.\n",
      "Test circle -120 trained in 1.6 seconds.\n",
      "Test circle -119 trained in 1.61 seconds.\n",
      "Test circle -118 trained in 1.6 seconds.\n",
      "Test circle -117 trained in 1.6 seconds.\n",
      "Test circle -116 trained in 1.6 seconds.\n",
      "Test circle -115 trained in 1.6 seconds.\n",
      "Test circle -114 trained in 1.59 seconds.\n",
      "Test circle -113 trained in 1.6 seconds.\n",
      "Test circle -112 trained in 1.6 seconds.\n",
      "Test circle -111 trained in 1.61 seconds.\n",
      "Test circle -110 trained in 1.59 seconds.\n",
      "Test circle -109 trained in 1.6 seconds.\n",
      "Test circle -108 trained in 1.63 seconds.\n",
      "Test circle -107 trained in 1.59 seconds.\n",
      "Test circle -106 trained in 1.6 seconds.\n",
      "Test circle -105 trained in 1.6 seconds.\n",
      "Test circle -104 trained in 1.6 seconds.\n",
      "Test circle -103 trained in 1.6 seconds.\n",
      "Test circle -102 trained in 1.69 seconds.\n",
      "Test circle -101 trained in 1.66 seconds.\n",
      "Test circle -100 trained in 1.62 seconds.\n",
      "Test circle -99 trained in 1.6 seconds.\n",
      "Test circle -98 trained in 1.61 seconds.\n",
      "Test circle -97 trained in 1.61 seconds.\n",
      "Test circle -96 trained in 1.59 seconds.\n",
      "Test circle -95 trained in 1.6 seconds.\n",
      "Test circle -94 trained in 1.61 seconds.\n",
      "Test circle -93 trained in 1.6 seconds.\n",
      "Test circle -92 trained in 1.6 seconds.\n",
      "Test circle -91 trained in 1.61 seconds.\n",
      "Test circle -90 trained in 1.6 seconds.\n",
      "Test circle -89 trained in 1.6 seconds.\n",
      "Test circle -88 trained in 1.61 seconds.\n",
      "Test circle -87 trained in 1.6 seconds.\n",
      "Test circle -86 trained in 1.62 seconds.\n",
      "Test circle -85 trained in 1.64 seconds.\n",
      "Test circle -84 trained in 1.6 seconds.\n",
      "Test circle -83 trained in 1.6 seconds.\n",
      "Test circle -82 trained in 1.68 seconds.\n",
      "Test circle -81 trained in 1.61 seconds.\n",
      "Test circle -80 trained in 1.6 seconds.\n",
      "Test circle -79 trained in 1.59 seconds.\n",
      "Test circle -78 trained in 1.61 seconds.\n",
      "Test circle -77 trained in 1.59 seconds.\n",
      "Test circle -76 trained in 1.6 seconds.\n",
      "Test circle -75 trained in 1.61 seconds.\n",
      "Test circle -74 trained in 1.6 seconds.\n",
      "Test circle -73 trained in 1.6 seconds.\n",
      "Test circle -72 trained in 1.6 seconds.\n",
      "Test circle -71 trained in 1.6 seconds.\n",
      "Test circle -70 trained in 1.6 seconds.\n",
      "Test circle -69 trained in 1.6 seconds.\n",
      "Test circle -68 trained in 1.6 seconds.\n",
      "Test circle -67 trained in 1.6 seconds.\n",
      "Test circle -66 trained in 1.59 seconds.\n",
      "Test circle -65 trained in 1.6 seconds.\n",
      "Test circle -64 trained in 1.62 seconds.\n",
      "Test circle -63 trained in 1.62 seconds.\n",
      "Test circle -62 trained in 1.69 seconds.\n",
      "Test circle -61 trained in 1.62 seconds.\n",
      "Test circle -60 trained in 1.6 seconds.\n",
      "Test circle -59 trained in 1.6 seconds.\n",
      "Test circle -58 trained in 1.61 seconds.\n",
      "Test circle -57 trained in 1.6 seconds.\n",
      "Test circle -56 trained in 1.61 seconds.\n",
      "Test circle -55 trained in 1.59 seconds.\n",
      "Test circle -54 trained in 1.59 seconds.\n",
      "Test circle -53 trained in 1.6 seconds.\n",
      "Test circle -52 trained in 1.6 seconds.\n",
      "Test circle -51 trained in 1.6 seconds.\n",
      "Test circle -50 trained in 1.6 seconds.\n",
      "Test circle -49 trained in 1.64 seconds.\n",
      "Test circle -48 trained in 1.68 seconds.\n",
      "Test circle -47 trained in 1.7 seconds.\n",
      "Test circle -46 trained in 1.6 seconds.\n",
      "Test circle -45 trained in 1.6 seconds.\n",
      "Test circle -44 trained in 1.6 seconds.\n",
      "Test circle -43 trained in 1.68 seconds.\n",
      "Test circle -42 trained in 1.6 seconds.\n",
      "Test circle -41 trained in 1.6 seconds.\n",
      "Test circle -40 trained in 1.6 seconds.\n",
      "Test circle -39 trained in 1.6 seconds.\n",
      "Test circle -38 trained in 1.62 seconds.\n",
      "Test circle -37 trained in 1.61 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -36 trained in 1.6 seconds.\n",
      "Test circle -35 trained in 1.59 seconds.\n",
      "Test circle -34 trained in 1.6 seconds.\n",
      "Test circle -33 trained in 1.6 seconds.\n",
      "Test circle -32 trained in 1.6 seconds.\n",
      "Test circle -31 trained in 1.62 seconds.\n",
      "Test circle -30 trained in 1.6 seconds.\n",
      "Test circle -29 trained in 1.61 seconds.\n",
      "Test circle -28 trained in 1.6 seconds.\n",
      "Test circle -27 trained in 1.63 seconds.\n",
      "Test circle -26 trained in 1.64 seconds.\n",
      "Test circle -25 trained in 1.6 seconds.\n",
      "Test circle -24 trained in 1.68 seconds.\n",
      "Test circle -23 trained in 1.59 seconds.\n",
      "Test circle -22 trained in 1.6 seconds.\n",
      "Test circle -21 trained in 1.61 seconds.\n",
      "Test circle -20 trained in 1.61 seconds.\n",
      "Test circle -19 trained in 1.6 seconds.\n",
      "Test circle -18 trained in 1.6 seconds.\n",
      "Test circle -17 trained in 1.6 seconds.\n",
      "Test circle -16 trained in 1.6 seconds.\n",
      "Test circle -15 trained in 1.6 seconds.\n",
      "Test circle -14 trained in 1.6 seconds.\n",
      "Test circle -13 trained in 1.61 seconds.\n",
      "Test circle -12 trained in 1.61 seconds.\n",
      "Test circle -11 trained in 1.64 seconds.\n",
      "Test circle -10 trained in 1.6 seconds.\n",
      "Test circle -9 trained in 1.6 seconds.\n",
      "Test circle -8 trained in 1.6 seconds.\n",
      "Test circle -7 trained in 1.74 seconds.\n",
      "Test circle -6 trained in 1.68 seconds.\n",
      "Test circle -5 trained in 1.62 seconds.\n",
      "Test circle -4 trained in 1.7 seconds.\n",
      "Test circle -3 trained in 1.61 seconds.\n",
      "Test circle -2 trained in 1.6 seconds.\n",
      "Test circle -1 trained in 1.64 seconds.\n",
      "Model trained in 1212.92 seconds.\n",
      "Hyperparams (25, 6, 3) trained in 1212.92 seconds.\n",
      "=========================================================================================\n",
      "Model (25, 6, 3) Ended\n",
      "=========================================================================================\n",
      "=========================================================================================\n",
      "Model (25, 6, 6) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 1.64 seconds.\n",
      "Test circle -724 trained in 1.62 seconds.\n",
      "Test circle -723 trained in 1.61 seconds.\n",
      "Test circle -722 trained in 1.65 seconds.\n",
      "Test circle -721 trained in 1.63 seconds.\n",
      "Test circle -720 trained in 1.63 seconds.\n",
      "Test circle -719 trained in 1.66 seconds.\n",
      "Test circle -718 trained in 1.62 seconds.\n",
      "Test circle -717 trained in 1.63 seconds.\n",
      "Test circle -716 trained in 1.62 seconds.\n",
      "Test circle -715 trained in 1.63 seconds.\n",
      "Test circle -714 trained in 1.63 seconds.\n",
      "Test circle -713 trained in 1.68 seconds.\n",
      "Test circle -712 trained in 1.63 seconds.\n",
      "Test circle -711 trained in 1.62 seconds.\n",
      "Test circle -710 trained in 1.61 seconds.\n",
      "Test circle -709 trained in 1.72 seconds.\n",
      "Test circle -708 trained in 1.63 seconds.\n",
      "Test circle -707 trained in 1.62 seconds.\n",
      "Test circle -706 trained in 1.62 seconds.\n",
      "Test circle -705 trained in 1.56 seconds.\n",
      "Test circle -704 trained in 1.53 seconds.\n",
      "Test circle -703 trained in 1.53 seconds.\n",
      "Test circle -702 trained in 1.52 seconds.\n",
      "Test circle -701 trained in 1.51 seconds.\n",
      "Test circle -700 trained in 1.52 seconds.\n",
      "Test circle -699 trained in 1.55 seconds.\n",
      "Test circle -698 trained in 1.53 seconds.\n",
      "Test circle -697 trained in 1.51 seconds.\n",
      "Test circle -696 trained in 1.51 seconds.\n",
      "Test circle -695 trained in 1.53 seconds.\n",
      "Test circle -694 trained in 1.51 seconds.\n",
      "Test circle -693 trained in 1.52 seconds.\n",
      "Test circle -692 trained in 1.52 seconds.\n",
      "Test circle -691 trained in 1.53 seconds.\n",
      "Test circle -690 trained in 1.53 seconds.\n",
      "Test circle -689 trained in 1.53 seconds.\n",
      "Test circle -688 trained in 1.61 seconds.\n",
      "Test circle -687 trained in 1.53 seconds.\n",
      "Test circle -686 trained in 1.58 seconds.\n",
      "Test circle -685 trained in 1.54 seconds.\n",
      "Test circle -684 trained in 1.56 seconds.\n",
      "Test circle -683 trained in 1.53 seconds.\n",
      "Test circle -682 trained in 1.54 seconds.\n",
      "Test circle -681 trained in 1.55 seconds.\n",
      "Test circle -680 trained in 1.57 seconds.\n",
      "Test circle -679 trained in 1.57 seconds.\n",
      "Test circle -678 trained in 1.52 seconds.\n",
      "Test circle -677 trained in 1.53 seconds.\n",
      "Test circle -676 trained in 1.55 seconds.\n",
      "Test circle -675 trained in 1.62 seconds.\n",
      "Test circle -674 trained in 1.58 seconds.\n",
      "Test circle -673 trained in 1.53 seconds.\n",
      "Test circle -672 trained in 1.53 seconds.\n",
      "Test circle -671 trained in 1.54 seconds.\n",
      "Test circle -670 trained in 1.57 seconds.\n",
      "Test circle -669 trained in 1.59 seconds.\n",
      "Test circle -668 trained in 1.67 seconds.\n",
      "Test circle -667 trained in 1.54 seconds.\n",
      "Test circle -666 trained in 1.55 seconds.\n",
      "Test circle -665 trained in 1.55 seconds.\n",
      "Test circle -664 trained in 1.59 seconds.\n",
      "Test circle -663 trained in 1.6 seconds.\n",
      "Test circle -662 trained in 1.56 seconds.\n",
      "Test circle -661 trained in 1.54 seconds.\n",
      "Test circle -660 trained in 1.58 seconds.\n",
      "Test circle -659 trained in 1.6 seconds.\n",
      "Test circle -658 trained in 1.58 seconds.\n",
      "Test circle -657 trained in 1.55 seconds.\n",
      "Test circle -656 trained in 1.57 seconds.\n",
      "Test circle -655 trained in 1.55 seconds.\n",
      "Test circle -654 trained in 1.62 seconds.\n",
      "Test circle -653 trained in 1.6 seconds.\n",
      "Test circle -652 trained in 1.55 seconds.\n",
      "Test circle -651 trained in 1.55 seconds.\n",
      "Test circle -650 trained in 1.57 seconds.\n",
      "Test circle -649 trained in 1.57 seconds.\n",
      "Test circle -648 trained in 1.6 seconds.\n",
      "Test circle -647 trained in 1.68 seconds.\n",
      "Test circle -646 trained in 1.57 seconds.\n",
      "Test circle -645 trained in 1.55 seconds.\n",
      "Test circle -644 trained in 1.85 seconds.\n",
      "Test circle -643 trained in 1.65 seconds.\n",
      "Test circle -642 trained in 1.61 seconds.\n",
      "Test circle -641 trained in 1.57 seconds.\n",
      "Test circle -640 trained in 1.57 seconds.\n",
      "Test circle -639 trained in 1.6 seconds.\n",
      "Test circle -638 trained in 1.66 seconds.\n",
      "Test circle -637 trained in 1.62 seconds.\n",
      "Test circle -636 trained in 1.71 seconds.\n",
      "Test circle -635 trained in 1.68 seconds.\n",
      "Test circle -634 trained in 1.59 seconds.\n",
      "Test circle -633 trained in 1.65 seconds.\n",
      "Test circle -632 trained in 1.6 seconds.\n",
      "Test circle -631 trained in 1.58 seconds.\n",
      "Test circle -630 trained in 1.57 seconds.\n",
      "Test circle -629 trained in 1.58 seconds.\n",
      "Test circle -628 trained in 1.64 seconds.\n",
      "Test circle -627 trained in 1.62 seconds.\n",
      "Test circle -626 trained in 1.65 seconds.\n",
      "Test circle -625 trained in 1.58 seconds.\n",
      "Test circle -624 trained in 1.57 seconds.\n",
      "Test circle -623 trained in 1.71 seconds.\n",
      "Test circle -622 trained in 1.61 seconds.\n",
      "Test circle -621 trained in 1.58 seconds.\n",
      "Test circle -620 trained in 1.58 seconds.\n",
      "Test circle -619 trained in 1.58 seconds.\n",
      "Test circle -618 trained in 1.64 seconds.\n",
      "Test circle -617 trained in 1.6 seconds.\n",
      "Test circle -616 trained in 1.58 seconds.\n",
      "Test circle -615 trained in 1.57 seconds.\n",
      "Test circle -614 trained in 1.57 seconds.\n",
      "Test circle -613 trained in 1.6 seconds.\n",
      "Test circle -612 trained in 1.62 seconds.\n",
      "Test circle -611 trained in 1.61 seconds.\n",
      "Test circle -610 trained in 1.58 seconds.\n",
      "Test circle -609 trained in 1.58 seconds.\n",
      "Test circle -608 trained in 1.59 seconds.\n",
      "Test circle -607 trained in 1.7 seconds.\n",
      "Test circle -606 trained in 1.64 seconds.\n",
      "Test circle -605 trained in 1.58 seconds.\n",
      "Test circle -604 trained in 1.64 seconds.\n",
      "Test circle -603 trained in 1.6 seconds.\n",
      "Test circle -602 trained in 1.62 seconds.\n",
      "Test circle -601 trained in 1.65 seconds.\n",
      "Test circle -600 trained in 1.57 seconds.\n",
      "Test circle -599 trained in 1.58 seconds.\n",
      "Test circle -598 trained in 1.58 seconds.\n",
      "Test circle -597 trained in 1.61 seconds.\n",
      "Test circle -596 trained in 1.63 seconds.\n",
      "Test circle -595 trained in 1.59 seconds.\n",
      "Test circle -594 trained in 1.58 seconds.\n",
      "Test circle -593 trained in 1.58 seconds.\n",
      "Test circle -592 trained in 1.63 seconds.\n",
      "Test circle -591 trained in 1.59 seconds.\n",
      "Test circle -590 trained in 1.57 seconds.\n",
      "Test circle -589 trained in 1.57 seconds.\n",
      "Test circle -588 trained in 1.58 seconds.\n",
      "Test circle -587 trained in 1.64 seconds.\n",
      "Test circle -586 trained in 1.75 seconds.\n",
      "Test circle -585 trained in 1.59 seconds.\n",
      "Test circle -584 trained in 1.58 seconds.\n",
      "Test circle -583 trained in 1.58 seconds.\n",
      "Test circle -582 trained in 1.63 seconds.\n",
      "Test circle -581 trained in 1.61 seconds.\n",
      "Test circle -580 trained in 1.61 seconds.\n",
      "Test circle -579 trained in 1.59 seconds.\n",
      "Test circle -578 trained in 1.58 seconds.\n",
      "Test circle -577 trained in 1.59 seconds.\n",
      "Test circle -576 trained in 1.63 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -575 trained in 1.64 seconds.\n",
      "Test circle -574 trained in 1.58 seconds.\n",
      "Test circle -573 trained in 1.57 seconds.\n",
      "Test circle -572 trained in 1.58 seconds.\n",
      "Test circle -571 trained in 1.61 seconds.\n",
      "Test circle -570 trained in 1.61 seconds.\n",
      "Test circle -569 trained in 1.57 seconds.\n",
      "Test circle -568 trained in 1.58 seconds.\n",
      "Test circle -567 trained in 1.57 seconds.\n",
      "Test circle -566 trained in 1.7 seconds.\n",
      "Test circle -565 trained in 1.61 seconds.\n",
      "Test circle -564 trained in 1.59 seconds.\n",
      "Test circle -563 trained in 1.58 seconds.\n",
      "Test circle -562 trained in 1.57 seconds.\n",
      "Test circle -561 trained in 1.62 seconds.\n",
      "Test circle -560 trained in 1.61 seconds.\n",
      "Test circle -559 trained in 1.58 seconds.\n",
      "Test circle -558 trained in 1.58 seconds.\n",
      "Test circle -557 trained in 1.58 seconds.\n",
      "Test circle -556 trained in 1.62 seconds.\n",
      "Test circle -555 trained in 1.61 seconds.\n",
      "Test circle -554 trained in 1.61 seconds.\n",
      "Test circle -553 trained in 1.59 seconds.\n",
      "Test circle -552 trained in 1.57 seconds.\n",
      "Test circle -551 trained in 1.59 seconds.\n",
      "Test circle -550 trained in 1.63 seconds.\n",
      "Test circle -549 trained in 1.59 seconds.\n",
      "Test circle -548 trained in 1.62 seconds.\n",
      "Test circle -547 trained in 1.67 seconds.\n",
      "Test circle -546 trained in 1.58 seconds.\n",
      "Test circle -545 trained in 1.63 seconds.\n",
      "Test circle -544 trained in 1.63 seconds.\n",
      "Test circle -543 trained in 1.59 seconds.\n",
      "Test circle -542 trained in 1.58 seconds.\n",
      "Test circle -541 trained in 1.58 seconds.\n",
      "Test circle -540 trained in 1.63 seconds.\n",
      "Test circle -539 trained in 1.63 seconds.\n",
      "Test circle -538 trained in 1.59 seconds.\n",
      "Test circle -537 trained in 1.58 seconds.\n",
      "Test circle -536 trained in 1.56 seconds.\n",
      "Test circle -535 trained in 1.65 seconds.\n",
      "Test circle -534 trained in 1.64 seconds.\n",
      "Test circle -533 trained in 1.57 seconds.\n",
      "Test circle -532 trained in 1.57 seconds.\n",
      "Test circle -531 trained in 1.58 seconds.\n",
      "Test circle -530 trained in 1.65 seconds.\n",
      "Test circle -529 trained in 1.65 seconds.\n",
      "Test circle -528 trained in 1.7 seconds.\n",
      "Test circle -527 trained in 1.59 seconds.\n",
      "Test circle -526 trained in 1.59 seconds.\n",
      "Test circle -525 trained in 1.6 seconds.\n",
      "Test circle -524 trained in 1.6 seconds.\n",
      "Test circle -523 trained in 1.62 seconds.\n",
      "Test circle -522 trained in 1.61 seconds.\n",
      "Test circle -521 trained in 1.61 seconds.\n",
      "Test circle -520 trained in 1.6 seconds.\n",
      "Test circle -519 trained in 1.62 seconds.\n",
      "Test circle -518 trained in 1.65 seconds.\n",
      "Test circle -517 trained in 1.57 seconds.\n",
      "Test circle -516 trained in 1.59 seconds.\n",
      "Test circle -515 trained in 1.61 seconds.\n",
      "Test circle -514 trained in 1.62 seconds.\n",
      "Test circle -513 trained in 1.63 seconds.\n",
      "Test circle -512 trained in 1.58 seconds.\n",
      "Test circle -511 trained in 1.62 seconds.\n",
      "Test circle -510 trained in 1.58 seconds.\n",
      "Test circle -509 trained in 1.64 seconds.\n",
      "Test circle -508 trained in 1.63 seconds.\n",
      "Test circle -507 trained in 1.67 seconds.\n",
      "Test circle -506 trained in 1.59 seconds.\n",
      "Test circle -505 trained in 1.58 seconds.\n",
      "Test circle -504 trained in 1.64 seconds.\n",
      "Test circle -503 trained in 1.63 seconds.\n",
      "Test circle -502 trained in 1.59 seconds.\n",
      "Test circle -501 trained in 1.59 seconds.\n",
      "Test circle -500 trained in 1.58 seconds.\n",
      "Test circle -499 trained in 1.65 seconds.\n",
      "Test circle -498 trained in 1.63 seconds.\n",
      "Test circle -497 trained in 1.65 seconds.\n",
      "Test circle -496 trained in 1.75 seconds.\n",
      "Test circle -495 trained in 1.59 seconds.\n",
      "Test circle -494 trained in 1.65 seconds.\n",
      "Test circle -493 trained in 1.63 seconds.\n",
      "Test circle -492 trained in 1.58 seconds.\n",
      "Test circle -491 trained in 1.65 seconds.\n",
      "Test circle -490 trained in 1.6 seconds.\n",
      "Test circle -489 trained in 1.66 seconds.\n",
      "Test circle -488 trained in 1.62 seconds.\n",
      "Test circle -487 trained in 1.67 seconds.\n",
      "Test circle -486 trained in 1.59 seconds.\n",
      "Test circle -485 trained in 1.59 seconds.\n",
      "Test circle -484 trained in 1.66 seconds.\n",
      "Test circle -483 trained in 1.61 seconds.\n",
      "Test circle -482 trained in 1.61 seconds.\n",
      "Test circle -481 trained in 1.59 seconds.\n",
      "Test circle -480 trained in 1.57 seconds.\n",
      "Test circle -479 trained in 1.6 seconds.\n",
      "Test circle -478 trained in 1.83 seconds.\n",
      "Test circle -477 trained in 1.65 seconds.\n",
      "Test circle -476 trained in 1.59 seconds.\n",
      "Test circle -475 trained in 1.58 seconds.\n",
      "Test circle -474 trained in 1.65 seconds.\n",
      "Test circle -473 trained in 1.62 seconds.\n",
      "Test circle -472 trained in 1.65 seconds.\n",
      "Test circle -471 trained in 1.58 seconds.\n",
      "Test circle -470 trained in 1.59 seconds.\n",
      "Test circle -469 trained in 1.6 seconds.\n",
      "Test circle -468 trained in 1.61 seconds.\n",
      "Test circle -467 trained in 1.72 seconds.\n",
      "Test circle -466 trained in 1.59 seconds.\n",
      "Test circle -465 trained in 1.57 seconds.\n",
      "Test circle -464 trained in 1.59 seconds.\n",
      "Test circle -463 trained in 1.66 seconds.\n",
      "Test circle -462 trained in 1.64 seconds.\n",
      "Test circle -461 trained in 1.58 seconds.\n",
      "Test circle -460 trained in 1.59 seconds.\n",
      "Test circle -459 trained in 1.6 seconds.\n",
      "Test circle -458 trained in 1.68 seconds.\n",
      "Test circle -457 trained in 1.64 seconds.\n",
      "Test circle -456 trained in 1.59 seconds.\n",
      "Test circle -455 trained in 1.66 seconds.\n",
      "Test circle -454 trained in 1.64 seconds.\n",
      "Test circle -453 trained in 1.67 seconds.\n",
      "Test circle -452 trained in 1.63 seconds.\n",
      "Test circle -451 trained in 1.58 seconds.\n",
      "Test circle -450 trained in 1.59 seconds.\n",
      "Test circle -449 trained in 1.58 seconds.\n",
      "Test circle -448 trained in 1.71 seconds.\n",
      "Test circle -447 trained in 1.68 seconds.\n",
      "Test circle -446 trained in 1.61 seconds.\n",
      "Test circle -445 trained in 1.59 seconds.\n",
      "Test circle -444 trained in 1.61 seconds.\n",
      "Test circle -443 trained in 1.65 seconds.\n",
      "Test circle -442 trained in 1.66 seconds.\n",
      "Test circle -441 trained in 1.64 seconds.\n",
      "Test circle -440 trained in 1.61 seconds.\n",
      "Test circle -439 trained in 1.63 seconds.\n",
      "Test circle -438 trained in 1.66 seconds.\n",
      "Test circle -437 trained in 1.72 seconds.\n",
      "Test circle -436 trained in 1.66 seconds.\n",
      "Test circle -435 trained in 1.64 seconds.\n",
      "Test circle -434 trained in 1.64 seconds.\n",
      "Test circle -433 trained in 1.66 seconds.\n",
      "Test circle -432 trained in 1.71 seconds.\n",
      "Test circle -431 trained in 1.68 seconds.\n",
      "Test circle -430 trained in 1.66 seconds.\n",
      "Test circle -429 trained in 1.65 seconds.\n",
      "Test circle -428 trained in 1.79 seconds.\n",
      "Test circle -427 trained in 1.76 seconds.\n",
      "Test circle -426 trained in 1.69 seconds.\n",
      "Test circle -425 trained in 1.69 seconds.\n",
      "Test circle -424 trained in 1.69 seconds.\n",
      "Test circle -423 trained in 1.75 seconds.\n",
      "Test circle -422 trained in 1.77 seconds.\n",
      "Test circle -421 trained in 1.71 seconds.\n",
      "Test circle -420 trained in 1.71 seconds.\n",
      "Test circle -419 trained in 1.71 seconds.\n",
      "Test circle -418 trained in 1.82 seconds.\n",
      "Test circle -417 trained in 1.81 seconds.\n",
      "Test circle -416 trained in 1.75 seconds.\n",
      "Test circle -415 trained in 1.73 seconds.\n",
      "Test circle -414 trained in 1.74 seconds.\n",
      "Test circle -413 trained in 1.84 seconds.\n",
      "Test circle -412 trained in 1.81 seconds.\n",
      "Test circle -411 trained in 1.76 seconds.\n",
      "Test circle -410 trained in 1.76 seconds.\n",
      "Test circle -409 trained in 1.81 seconds.\n",
      "Test circle -408 trained in 1.87 seconds.\n",
      "Test circle -407 trained in 1.88 seconds.\n",
      "Test circle -406 trained in 1.81 seconds.\n",
      "Test circle -405 trained in 1.78 seconds.\n",
      "Test circle -404 trained in 1.86 seconds.\n",
      "Test circle -403 trained in 1.94 seconds.\n",
      "Test circle -402 trained in 1.84 seconds.\n",
      "Test circle -401 trained in 1.83 seconds.\n",
      "Test circle -400 trained in 1.84 seconds.\n",
      "Test circle -399 trained in 1.93 seconds.\n",
      "Test circle -398 trained in 1.86 seconds.\n",
      "Test circle -397 trained in 1.87 seconds.\n",
      "Test circle -396 trained in 1.81 seconds.\n",
      "Test circle -395 trained in 1.95 seconds.\n",
      "Test circle -394 trained in 1.97 seconds.\n",
      "Test circle -393 trained in 1.9 seconds.\n",
      "Test circle -392 trained in 1.9 seconds.\n",
      "Test circle -391 trained in 1.92 seconds.\n",
      "Test circle -390 trained in 1.61 seconds.\n",
      "Test circle -389 trained in 1.52 seconds.\n",
      "Test circle -388 trained in 1.53 seconds.\n",
      "Test circle -387 trained in 1.61 seconds.\n",
      "Test circle -386 trained in 1.58 seconds.\n",
      "Test circle -385 trained in 1.59 seconds.\n",
      "Test circle -384 trained in 1.61 seconds.\n",
      "Test circle -383 trained in 1.54 seconds.\n",
      "Test circle -382 trained in 1.54 seconds.\n",
      "Test circle -381 trained in 1.58 seconds.\n",
      "Test circle -380 trained in 1.56 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -379 trained in 1.57 seconds.\n",
      "Test circle -378 trained in 1.55 seconds.\n",
      "Test circle -377 trained in 1.55 seconds.\n",
      "Test circle -376 trained in 1.56 seconds.\n",
      "Test circle -375 trained in 1.61 seconds.\n",
      "Test circle -374 trained in 1.61 seconds.\n",
      "Test circle -373 trained in 1.56 seconds.\n",
      "Test circle -372 trained in 1.55 seconds.\n",
      "Test circle -371 trained in 1.57 seconds.\n",
      "Test circle -370 trained in 1.64 seconds.\n",
      "Test circle -369 trained in 1.6 seconds.\n",
      "Test circle -368 trained in 1.65 seconds.\n",
      "Test circle -367 trained in 1.61 seconds.\n",
      "Test circle -366 trained in 1.57 seconds.\n",
      "Test circle -365 trained in 1.65 seconds.\n",
      "Test circle -364 trained in 1.63 seconds.\n",
      "Test circle -363 trained in 1.58 seconds.\n",
      "Test circle -362 trained in 1.58 seconds.\n",
      "Test circle -361 trained in 1.57 seconds.\n",
      "Test circle -360 trained in 1.62 seconds.\n",
      "Test circle -359 trained in 1.61 seconds.\n",
      "Test circle -358 trained in 1.61 seconds.\n",
      "Test circle -357 trained in 1.57 seconds.\n",
      "Test circle -356 trained in 1.58 seconds.\n",
      "Test circle -355 trained in 1.61 seconds.\n",
      "Test circle -354 trained in 1.6 seconds.\n",
      "Test circle -353 trained in 1.61 seconds.\n",
      "Test circle -352 trained in 1.58 seconds.\n",
      "Test circle -351 trained in 1.57 seconds.\n",
      "Test circle -350 trained in 1.6 seconds.\n",
      "Test circle -349 trained in 1.71 seconds.\n",
      "Test circle -348 trained in 1.62 seconds.\n",
      "Test circle -347 trained in 1.59 seconds.\n",
      "Test circle -346 trained in 1.58 seconds.\n",
      "Test circle -345 trained in 1.6 seconds.\n",
      "Test circle -344 trained in 1.65 seconds.\n",
      "Test circle -343 trained in 1.63 seconds.\n",
      "Test circle -342 trained in 1.58 seconds.\n",
      "Test circle -341 trained in 1.59 seconds.\n",
      "Test circle -340 trained in 1.59 seconds.\n",
      "Test circle -339 trained in 1.63 seconds.\n",
      "Test circle -338 trained in 1.63 seconds.\n",
      "Test circle -337 trained in 1.59 seconds.\n",
      "Test circle -336 trained in 1.59 seconds.\n",
      "Test circle -335 trained in 1.6 seconds.\n",
      "Test circle -334 trained in 1.61 seconds.\n",
      "Test circle -333 trained in 1.64 seconds.\n",
      "Test circle -332 trained in 1.57 seconds.\n",
      "Test circle -331 trained in 1.58 seconds.\n",
      "Test circle -330 trained in 1.61 seconds.\n",
      "Test circle -329 trained in 1.65 seconds.\n",
      "Test circle -328 trained in 1.7 seconds.\n",
      "Test circle -327 trained in 1.58 seconds.\n",
      "Test circle -326 trained in 1.59 seconds.\n",
      "Test circle -325 trained in 1.6 seconds.\n",
      "Test circle -324 trained in 1.62 seconds.\n",
      "Test circle -323 trained in 1.62 seconds.\n",
      "Test circle -322 trained in 1.6 seconds.\n",
      "Test circle -321 trained in 1.59 seconds.\n",
      "Test circle -320 trained in 1.59 seconds.\n",
      "Test circle -319 trained in 1.61 seconds.\n",
      "Test circle -318 trained in 1.62 seconds.\n",
      "Test circle -317 trained in 1.64 seconds.\n",
      "Test circle -316 trained in 1.59 seconds.\n",
      "Test circle -315 trained in 1.59 seconds.\n",
      "Test circle -314 trained in 1.58 seconds.\n",
      "Test circle -313 trained in 1.64 seconds.\n",
      "Test circle -312 trained in 1.63 seconds.\n",
      "Test circle -311 trained in 1.6 seconds.\n",
      "Test circle -310 trained in 1.72 seconds.\n",
      "Test circle -309 trained in 1.58 seconds.\n",
      "Test circle -308 trained in 1.66 seconds.\n",
      "Test circle -307 trained in 1.66 seconds.\n",
      "Test circle -306 trained in 1.59 seconds.\n",
      "Test circle -305 trained in 1.59 seconds.\n",
      "Test circle -304 trained in 1.6 seconds.\n",
      "Test circle -303 trained in 1.67 seconds.\n",
      "Test circle -302 trained in 1.69 seconds.\n",
      "Test circle -301 trained in 1.61 seconds.\n",
      "Test circle -300 trained in 1.61 seconds.\n",
      "Test circle -299 trained in 1.62 seconds.\n",
      "Test circle -298 trained in 1.67 seconds.\n",
      "Test circle -297 trained in 1.68 seconds.\n",
      "Test circle -296 trained in 1.64 seconds.\n",
      "Test circle -295 trained in 1.63 seconds.\n",
      "Test circle -294 trained in 1.64 seconds.\n",
      "Test circle -293 trained in 1.75 seconds.\n",
      "Test circle -292 trained in 1.72 seconds.\n",
      "Test circle -291 trained in 1.84 seconds.\n",
      "Test circle -290 trained in 1.79 seconds.\n",
      "Test circle -289 trained in 1.68 seconds.\n",
      "Test circle -288 trained in 1.73 seconds.\n",
      "Test circle -287 trained in 1.74 seconds.\n",
      "Test circle -286 trained in 1.68 seconds.\n",
      "Test circle -285 trained in 1.69 seconds.\n",
      "Test circle -284 trained in 1.69 seconds.\n",
      "Test circle -283 trained in 1.78 seconds.\n",
      "Test circle -282 trained in 1.77 seconds.\n",
      "Test circle -281 trained in 1.7 seconds.\n",
      "Test circle -280 trained in 1.71 seconds.\n",
      "Test circle -279 trained in 1.74 seconds.\n",
      "Test circle -278 trained in 1.8 seconds.\n",
      "Test circle -277 trained in 1.75 seconds.\n",
      "Test circle -276 trained in 1.73 seconds.\n",
      "Test circle -275 trained in 1.75 seconds.\n",
      "Test circle -274 trained in 1.79 seconds.\n",
      "Test circle -273 trained in 1.84 seconds.\n",
      "Test circle -272 trained in 1.78 seconds.\n",
      "Test circle -271 trained in 1.86 seconds.\n",
      "Test circle -270 trained in 1.74 seconds.\n",
      "Test circle -269 trained in 1.84 seconds.\n",
      "Test circle -268 trained in 1.86 seconds.\n",
      "Test circle -267 trained in 1.79 seconds.\n",
      "Test circle -266 trained in 1.8 seconds.\n",
      "Test circle -265 trained in 1.82 seconds.\n",
      "Test circle -264 trained in 1.92 seconds.\n",
      "Test circle -263 trained in 2.0 seconds.\n",
      "Test circle -262 trained in 1.53 seconds.\n",
      "Test circle -261 trained in 1.54 seconds.\n",
      "Test circle -260 trained in 1.6 seconds.\n",
      "Test circle -259 trained in 1.61 seconds.\n",
      "Test circle -258 trained in 1.54 seconds.\n",
      "Test circle -257 trained in 1.55 seconds.\n",
      "Test circle -256 trained in 1.54 seconds.\n",
      "Test circle -255 trained in 1.61 seconds.\n",
      "Test circle -254 trained in 1.61 seconds.\n",
      "Test circle -253 trained in 1.55 seconds.\n",
      "Test circle -252 trained in 1.55 seconds.\n",
      "Test circle -251 trained in 1.64 seconds.\n",
      "Test circle -250 trained in 1.61 seconds.\n",
      "Test circle -249 trained in 1.6 seconds.\n",
      "Test circle -248 trained in 1.6 seconds.\n",
      "Test circle -247 trained in 1.56 seconds.\n",
      "Test circle -246 trained in 1.57 seconds.\n",
      "Test circle -245 trained in 1.59 seconds.\n",
      "Test circle -244 trained in 1.6 seconds.\n",
      "Test circle -243 trained in 1.6 seconds.\n",
      "Test circle -242 trained in 1.58 seconds.\n",
      "Test circle -241 trained in 1.6 seconds.\n",
      "Test circle -240 trained in 1.57 seconds.\n",
      "Test circle -239 trained in 1.64 seconds.\n",
      "Test circle -238 trained in 1.64 seconds.\n",
      "Test circle -237 trained in 1.59 seconds.\n",
      "Test circle -236 trained in 1.58 seconds.\n",
      "Test circle -235 trained in 1.58 seconds.\n",
      "Test circle -234 trained in 1.64 seconds.\n",
      "Test circle -233 trained in 1.63 seconds.\n",
      "Test circle -232 trained in 1.58 seconds.\n",
      "Test circle -231 trained in 1.67 seconds.\n",
      "Test circle -230 trained in 1.59 seconds.\n",
      "Test circle -229 trained in 1.64 seconds.\n",
      "Test circle -228 trained in 1.64 seconds.\n",
      "Test circle -227 trained in 1.59 seconds.\n",
      "Test circle -226 trained in 1.59 seconds.\n",
      "Test circle -225 trained in 1.59 seconds.\n",
      "Test circle -224 trained in 1.6 seconds.\n",
      "Test circle -223 trained in 1.61 seconds.\n",
      "Test circle -222 trained in 1.6 seconds.\n",
      "Test circle -221 trained in 1.6 seconds.\n",
      "Test circle -220 trained in 1.57 seconds.\n",
      "Test circle -219 trained in 1.57 seconds.\n",
      "Test circle -218 trained in 1.63 seconds.\n",
      "Test circle -217 trained in 1.62 seconds.\n",
      "Test circle -216 trained in 1.57 seconds.\n",
      "Test circle -215 trained in 1.57 seconds.\n",
      "Test circle -214 trained in 1.58 seconds.\n",
      "Test circle -213 trained in 1.64 seconds.\n",
      "Test circle -212 trained in 1.75 seconds.\n",
      "Test circle -211 trained in 1.6 seconds.\n",
      "Test circle -210 trained in 1.59 seconds.\n",
      "Test circle -209 trained in 1.59 seconds.\n",
      "Test circle -208 trained in 1.67 seconds.\n",
      "Test circle -207 trained in 1.65 seconds.\n",
      "Test circle -206 trained in 1.62 seconds.\n",
      "Test circle -205 trained in 1.62 seconds.\n",
      "Test circle -204 trained in 1.62 seconds.\n",
      "Test circle -203 trained in 1.69 seconds.\n",
      "Test circle -202 trained in 1.71 seconds.\n",
      "Test circle -201 trained in 1.63 seconds.\n",
      "Test circle -200 trained in 1.66 seconds.\n",
      "Test circle -199 trained in 1.64 seconds.\n",
      "Test circle -198 trained in 1.74 seconds.\n",
      "Test circle -197 trained in 1.72 seconds.\n",
      "Test circle -196 trained in 1.66 seconds.\n",
      "Test circle -195 trained in 1.65 seconds.\n",
      "Test circle -194 trained in 1.66 seconds.\n",
      "Test circle -193 trained in 1.74 seconds.\n",
      "Test circle -192 trained in 1.82 seconds.\n",
      "Test circle -191 trained in 1.69 seconds.\n",
      "Test circle -190 trained in 1.68 seconds.\n",
      "Test circle -189 trained in 1.7 seconds.\n",
      "Test circle -188 trained in 1.74 seconds.\n",
      "Test circle -187 trained in 1.77 seconds.\n",
      "Test circle -186 trained in 1.7 seconds.\n",
      "Test circle -185 trained in 1.77 seconds.\n",
      "Test circle -184 trained in 1.74 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -183 trained in 1.78 seconds.\n",
      "Test circle -182 trained in 1.79 seconds.\n",
      "Test circle -181 trained in 1.73 seconds.\n",
      "Test circle -180 trained in 1.76 seconds.\n",
      "Test circle -179 trained in 1.79 seconds.\n",
      "Test circle -178 trained in 1.83 seconds.\n",
      "Test circle -177 trained in 1.76 seconds.\n",
      "Test circle -176 trained in 1.79 seconds.\n",
      "Test circle -175 trained in 1.76 seconds.\n",
      "Test circle -174 trained in 1.83 seconds.\n",
      "Test circle -173 trained in 1.94 seconds.\n",
      "Test circle -172 trained in 1.8 seconds.\n",
      "Test circle -171 trained in 1.8 seconds.\n",
      "Test circle -170 trained in 1.81 seconds.\n",
      "Test circle -169 trained in 1.9 seconds.\n",
      "Test circle -168 trained in 1.89 seconds.\n",
      "Test circle -167 trained in 1.91 seconds.\n",
      "Test circle -166 trained in 1.81 seconds.\n",
      "Test circle -165 trained in 1.91 seconds.\n",
      "Test circle -164 trained in 1.91 seconds.\n",
      "Test circle -163 trained in 1.87 seconds.\n",
      "Test circle -162 trained in 1.87 seconds.\n",
      "Test circle -161 trained in 1.88 seconds.\n",
      "Test circle -160 trained in 1.98 seconds.\n",
      "Test circle -159 trained in 1.9 seconds.\n",
      "Test circle -158 trained in 1.62 seconds.\n",
      "Test circle -157 trained in 1.54 seconds.\n",
      "Test circle -156 trained in 1.56 seconds.\n",
      "Test circle -155 trained in 1.58 seconds.\n",
      "Test circle -154 trained in 1.53 seconds.\n",
      "Test circle -153 trained in 1.62 seconds.\n",
      "Test circle -152 trained in 1.54 seconds.\n",
      "Test circle -151 trained in 1.6 seconds.\n",
      "Test circle -150 trained in 1.59 seconds.\n",
      "Test circle -149 trained in 1.55 seconds.\n",
      "Test circle -148 trained in 1.62 seconds.\n",
      "Test circle -147 trained in 1.56 seconds.\n",
      "Test circle -146 trained in 1.6 seconds.\n",
      "Test circle -145 trained in 1.6 seconds.\n",
      "Test circle -144 trained in 1.58 seconds.\n",
      "Test circle -143 trained in 1.57 seconds.\n",
      "Test circle -142 trained in 1.57 seconds.\n",
      "Test circle -141 trained in 1.59 seconds.\n",
      "Test circle -140 trained in 1.6 seconds.\n",
      "Test circle -139 trained in 1.6 seconds.\n",
      "Test circle -138 trained in 1.57 seconds.\n",
      "Test circle -137 trained in 1.56 seconds.\n",
      "Test circle -136 trained in 1.58 seconds.\n",
      "Test circle -135 trained in 1.59 seconds.\n",
      "Test circle -134 trained in 1.72 seconds.\n",
      "Test circle -133 trained in 1.57 seconds.\n",
      "Test circle -132 trained in 1.58 seconds.\n",
      "Test circle -131 trained in 1.58 seconds.\n",
      "Test circle -130 trained in 1.66 seconds.\n",
      "Test circle -129 trained in 1.64 seconds.\n",
      "Test circle -128 trained in 1.58 seconds.\n",
      "Test circle -127 trained in 1.59 seconds.\n",
      "Test circle -126 trained in 1.57 seconds.\n",
      "Test circle -125 trained in 1.63 seconds.\n",
      "Test circle -124 trained in 1.59 seconds.\n",
      "Test circle -123 trained in 1.56 seconds.\n",
      "Test circle -122 trained in 1.55 seconds.\n",
      "Test circle -121 trained in 1.54 seconds.\n",
      "Test circle -120 trained in 1.63 seconds.\n",
      "Test circle -119 trained in 1.61 seconds.\n",
      "Test circle -118 trained in 1.57 seconds.\n",
      "Test circle -117 trained in 1.56 seconds.\n",
      "Test circle -116 trained in 1.56 seconds.\n",
      "Test circle -115 trained in 1.57 seconds.\n",
      "Test circle -114 trained in 1.64 seconds.\n",
      "Test circle -113 trained in 1.71 seconds.\n",
      "Test circle -112 trained in 1.57 seconds.\n",
      "Test circle -111 trained in 1.59 seconds.\n",
      "Test circle -110 trained in 1.59 seconds.\n",
      "Test circle -109 trained in 1.64 seconds.\n",
      "Test circle -108 trained in 1.65 seconds.\n",
      "Test circle -107 trained in 1.59 seconds.\n",
      "Test circle -106 trained in 1.59 seconds.\n",
      "Test circle -105 trained in 1.6 seconds.\n",
      "Test circle -104 trained in 1.66 seconds.\n",
      "Test circle -103 trained in 1.67 seconds.\n",
      "Test circle -102 trained in 1.61 seconds.\n",
      "Test circle -101 trained in 1.61 seconds.\n",
      "Test circle -100 trained in 1.62 seconds.\n",
      "Test circle -99 trained in 1.68 seconds.\n",
      "Test circle -98 trained in 1.69 seconds.\n",
      "Test circle -97 trained in 1.63 seconds.\n",
      "Test circle -96 trained in 1.64 seconds.\n",
      "Test circle -95 trained in 1.66 seconds.\n",
      "Test circle -94 trained in 1.77 seconds.\n",
      "Test circle -93 trained in 1.72 seconds.\n",
      "Test circle -92 trained in 1.75 seconds.\n",
      "Test circle -91 trained in 1.67 seconds.\n",
      "Test circle -90 trained in 1.67 seconds.\n",
      "Test circle -89 trained in 1.73 seconds.\n",
      "Test circle -88 trained in 1.73 seconds.\n",
      "Test circle -87 trained in 1.68 seconds.\n",
      "Test circle -86 trained in 1.69 seconds.\n",
      "Test circle -85 trained in 1.69 seconds.\n",
      "Test circle -84 trained in 1.75 seconds.\n",
      "Test circle -83 trained in 1.76 seconds.\n",
      "Test circle -82 trained in 1.71 seconds.\n",
      "Test circle -81 trained in 1.71 seconds.\n",
      "Test circle -80 trained in 1.73 seconds.\n",
      "Test circle -79 trained in 1.79 seconds.\n",
      "Test circle -78 trained in 1.8 seconds.\n",
      "Test circle -77 trained in 1.74 seconds.\n",
      "Test circle -76 trained in 1.74 seconds.\n",
      "Test circle -75 trained in 1.77 seconds.\n",
      "Test circle -74 trained in 1.84 seconds.\n",
      "Test circle -73 trained in 1.76 seconds.\n",
      "Test circle -72 trained in 1.85 seconds.\n",
      "Test circle -71 trained in 1.76 seconds.\n",
      "Test circle -70 trained in 1.82 seconds.\n",
      "Test circle -69 trained in 1.86 seconds.\n",
      "Test circle -68 trained in 1.8 seconds.\n",
      "Test circle -67 trained in 1.8 seconds.\n",
      "Test circle -66 trained in 1.82 seconds.\n",
      "Test circle -65 trained in 1.88 seconds.\n",
      "Test circle -64 trained in 1.84 seconds.\n",
      "Test circle -63 trained in 1.84 seconds.\n",
      "Test circle -62 trained in 1.82 seconds.\n",
      "Test circle -61 trained in 1.87 seconds.\n",
      "Test circle -60 trained in 1.93 seconds.\n",
      "Test circle -59 trained in 1.86 seconds.\n",
      "Test circle -58 trained in 1.89 seconds.\n",
      "Test circle -57 trained in 1.87 seconds.\n",
      "Test circle -56 trained in 1.96 seconds.\n",
      "Test circle -55 trained in 1.89 seconds.\n",
      "Test circle -54 trained in 1.87 seconds.\n",
      "Test circle -53 trained in 1.53 seconds.\n",
      "Test circle -52 trained in 1.55 seconds.\n",
      "Test circle -51 trained in 1.64 seconds.\n",
      "Test circle -50 trained in 1.54 seconds.\n",
      "Test circle -49 trained in 1.54 seconds.\n",
      "Test circle -48 trained in 1.53 seconds.\n",
      "Test circle -47 trained in 1.6 seconds.\n",
      "Test circle -46 trained in 1.58 seconds.\n",
      "Test circle -45 trained in 1.61 seconds.\n",
      "Test circle -44 trained in 1.58 seconds.\n",
      "Test circle -43 trained in 1.54 seconds.\n",
      "Test circle -42 trained in 1.58 seconds.\n",
      "Test circle -41 trained in 1.6 seconds.\n",
      "Test circle -40 trained in 1.63 seconds.\n",
      "Test circle -39 trained in 1.82 seconds.\n",
      "Test circle -38 trained in 1.57 seconds.\n",
      "Test circle -37 trained in 1.56 seconds.\n",
      "Test circle -36 trained in 1.57 seconds.\n",
      "Test circle -35 trained in 1.56 seconds.\n",
      "Test circle -34 trained in 1.63 seconds.\n",
      "Test circle -33 trained in 1.62 seconds.\n",
      "Test circle -32 trained in 1.57 seconds.\n",
      "Test circle -31 trained in 1.57 seconds.\n",
      "Test circle -30 trained in 1.66 seconds.\n",
      "Test circle -29 trained in 1.63 seconds.\n",
      "Test circle -28 trained in 1.63 seconds.\n",
      "Test circle -27 trained in 1.56 seconds.\n",
      "Test circle -26 trained in 1.57 seconds.\n",
      "Test circle -25 trained in 1.64 seconds.\n",
      "Test circle -24 trained in 1.68 seconds.\n",
      "Test circle -23 trained in 1.64 seconds.\n",
      "Test circle -22 trained in 1.59 seconds.\n",
      "Test circle -21 trained in 1.58 seconds.\n",
      "Test circle -20 trained in 1.59 seconds.\n",
      "Test circle -19 trained in 1.59 seconds.\n",
      "Test circle -18 trained in 1.59 seconds.\n",
      "Test circle -17 trained in 1.61 seconds.\n",
      "Test circle -16 trained in 1.58 seconds.\n",
      "Test circle -15 trained in 1.57 seconds.\n",
      "Test circle -14 trained in 1.58 seconds.\n",
      "Test circle -13 trained in 1.62 seconds.\n",
      "Test circle -12 trained in 1.63 seconds.\n",
      "Test circle -11 trained in 1.57 seconds.\n",
      "Test circle -10 trained in 1.67 seconds.\n",
      "Test circle -9 trained in 1.59 seconds.\n",
      "Test circle -8 trained in 1.64 seconds.\n",
      "Test circle -7 trained in 1.67 seconds.\n",
      "Test circle -6 trained in 1.59 seconds.\n",
      "Test circle -5 trained in 1.58 seconds.\n",
      "Test circle -4 trained in 1.59 seconds.\n",
      "Test circle -3 trained in 1.63 seconds.\n",
      "Test circle -2 trained in 1.62 seconds.\n",
      "Test circle -1 trained in 1.58 seconds.\n",
      "Model trained in 1192.16 seconds.\n",
      "Hyperparams (25, 6, 6) trained in 1192.16 seconds.\n",
      "=========================================================================================\n",
      "Model (25, 6, 6) Ended\n",
      "=========================================================================================\n",
      "=========================================================================================\n",
      "Model (50, 3, 3) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 2.7 seconds.\n",
      "Test circle -724 trained in 2.74 seconds.\n",
      "Test circle -723 trained in 2.74 seconds.\n",
      "Test circle -722 trained in 2.71 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -721 trained in 2.76 seconds.\n",
      "Test circle -720 trained in 2.74 seconds.\n",
      "Test circle -719 trained in 2.69 seconds.\n",
      "Test circle -718 trained in 2.75 seconds.\n",
      "Test circle -717 trained in 2.74 seconds.\n",
      "Test circle -716 trained in 2.8 seconds.\n",
      "Test circle -715 trained in 2.78 seconds.\n",
      "Test circle -714 trained in 2.75 seconds.\n",
      "Test circle -713 trained in 2.7 seconds.\n",
      "Test circle -712 trained in 2.77 seconds.\n",
      "Test circle -711 trained in 2.74 seconds.\n",
      "Test circle -710 trained in 2.7 seconds.\n",
      "Test circle -709 trained in 2.76 seconds.\n",
      "Test circle -708 trained in 2.74 seconds.\n",
      "Test circle -707 trained in 2.72 seconds.\n",
      "Test circle -706 trained in 2.76 seconds.\n",
      "Test circle -705 trained in 2.75 seconds.\n",
      "Test circle -704 trained in 2.71 seconds.\n",
      "Test circle -703 trained in 2.77 seconds.\n",
      "Test circle -702 trained in 2.74 seconds.\n",
      "Test circle -701 trained in 2.67 seconds.\n",
      "Test circle -700 trained in 2.76 seconds.\n",
      "Test circle -699 trained in 2.75 seconds.\n",
      "Test circle -698 trained in 3.15 seconds.\n",
      "Test circle -697 trained in 2.93 seconds.\n",
      "Test circle -696 trained in 3.03 seconds.\n",
      "Test circle -695 trained in 2.95 seconds.\n",
      "Test circle -694 trained in 3.04 seconds.\n",
      "Test circle -693 trained in 2.81 seconds.\n",
      "Test circle -692 trained in 2.78 seconds.\n",
      "Test circle -691 trained in 3.11 seconds.\n",
      "Test circle -690 trained in 2.91 seconds.\n",
      "Test circle -689 trained in 3.12 seconds.\n",
      "Test circle -688 trained in 3.16 seconds.\n",
      "Test circle -687 trained in 3.13 seconds.\n",
      "Test circle -686 trained in 2.95 seconds.\n",
      "Test circle -685 trained in 2.86 seconds.\n",
      "Test circle -684 trained in 2.93 seconds.\n",
      "Test circle -683 trained in 2.98 seconds.\n",
      "Test circle -682 trained in 3.07 seconds.\n",
      "Test circle -681 trained in 3.11 seconds.\n",
      "Test circle -680 trained in 3.03 seconds.\n",
      "Test circle -679 trained in 2.89 seconds.\n",
      "Test circle -678 trained in 2.93 seconds.\n",
      "Test circle -677 trained in 2.86 seconds.\n",
      "Test circle -676 trained in 2.81 seconds.\n",
      "Test circle -675 trained in 2.86 seconds.\n",
      "Test circle -674 trained in 2.79 seconds.\n",
      "Test circle -673 trained in 2.83 seconds.\n",
      "Test circle -672 trained in 2.86 seconds.\n",
      "Test circle -671 trained in 2.79 seconds.\n",
      "Test circle -670 trained in 2.79 seconds.\n",
      "Test circle -669 trained in 2.84 seconds.\n",
      "Test circle -668 trained in 2.84 seconds.\n",
      "Test circle -667 trained in 2.85 seconds.\n",
      "Test circle -666 trained in 2.87 seconds.\n",
      "Test circle -665 trained in 3.07 seconds.\n",
      "Test circle -664 trained in 3.39 seconds.\n",
      "Test circle -663 trained in 3.03 seconds.\n",
      "Test circle -662 trained in 3.04 seconds.\n",
      "Test circle -661 trained in 2.99 seconds.\n",
      "Test circle -660 trained in 3.24 seconds.\n",
      "Test circle -659 trained in 3.46 seconds.\n",
      "Test circle -658 trained in 3.0 seconds.\n",
      "Test circle -657 trained in 2.83 seconds.\n",
      "Test circle -656 trained in 3.11 seconds.\n",
      "Test circle -655 trained in 3.18 seconds.\n",
      "Test circle -654 trained in 2.88 seconds.\n",
      "Test circle -653 trained in 3.48 seconds.\n",
      "Test circle -652 trained in 2.88 seconds.\n",
      "Test circle -651 trained in 2.82 seconds.\n",
      "Test circle -650 trained in 2.82 seconds.\n",
      "Test circle -649 trained in 2.8 seconds.\n",
      "Test circle -648 trained in 2.88 seconds.\n",
      "Test circle -647 trained in 2.85 seconds.\n",
      "Test circle -646 trained in 2.9 seconds.\n",
      "Test circle -645 trained in 2.86 seconds.\n",
      "Test circle -644 trained in 2.88 seconds.\n",
      "Test circle -643 trained in 2.94 seconds.\n",
      "Test circle -642 trained in 2.86 seconds.\n",
      "Test circle -641 trained in 2.8 seconds.\n",
      "Test circle -640 trained in 2.81 seconds.\n",
      "Test circle -639 trained in 2.76 seconds.\n",
      "Test circle -638 trained in 2.75 seconds.\n",
      "Test circle -637 trained in 2.77 seconds.\n",
      "Test circle -636 trained in 2.77 seconds.\n",
      "Test circle -635 trained in 2.77 seconds.\n",
      "Test circle -634 trained in 2.79 seconds.\n",
      "Test circle -633 trained in 2.83 seconds.\n",
      "Test circle -632 trained in 2.82 seconds.\n",
      "Test circle -631 trained in 2.76 seconds.\n",
      "Test circle -630 trained in 2.75 seconds.\n",
      "Test circle -629 trained in 2.77 seconds.\n",
      "Test circle -628 trained in 2.76 seconds.\n",
      "Test circle -627 trained in 2.75 seconds.\n",
      "Test circle -626 trained in 2.77 seconds.\n",
      "Test circle -625 trained in 2.86 seconds.\n",
      "Test circle -624 trained in 2.8 seconds.\n",
      "Test circle -623 trained in 2.81 seconds.\n",
      "Test circle -622 trained in 2.79 seconds.\n",
      "Test circle -621 trained in 2.77 seconds.\n",
      "Test circle -620 trained in 2.75 seconds.\n",
      "Test circle -619 trained in 2.91 seconds.\n",
      "Test circle -618 trained in 2.83 seconds.\n",
      "Test circle -617 trained in 2.88 seconds.\n",
      "Test circle -616 trained in 2.85 seconds.\n",
      "Test circle -615 trained in 2.78 seconds.\n",
      "Test circle -614 trained in 2.79 seconds.\n",
      "Test circle -613 trained in 2.78 seconds.\n",
      "Test circle -612 trained in 2.84 seconds.\n",
      "Test circle -611 trained in 2.76 seconds.\n",
      "Test circle -610 trained in 2.94 seconds.\n",
      "Test circle -609 trained in 2.82 seconds.\n",
      "Test circle -608 trained in 2.75 seconds.\n",
      "Test circle -607 trained in 2.79 seconds.\n",
      "Test circle -606 trained in 2.86 seconds.\n",
      "Test circle -605 trained in 2.78 seconds.\n",
      "Test circle -604 trained in 2.8 seconds.\n",
      "Test circle -603 trained in 2.85 seconds.\n",
      "Test circle -602 trained in 2.84 seconds.\n",
      "Test circle -601 trained in 2.77 seconds.\n",
      "Test circle -600 trained in 2.77 seconds.\n",
      "Test circle -599 trained in 2.85 seconds.\n",
      "Test circle -598 trained in 2.89 seconds.\n",
      "Test circle -597 trained in 2.77 seconds.\n",
      "Test circle -596 trained in 2.81 seconds.\n",
      "Test circle -595 trained in 2.89 seconds.\n",
      "Test circle -594 trained in 2.84 seconds.\n",
      "Test circle -593 trained in 2.84 seconds.\n",
      "Test circle -592 trained in 2.8 seconds.\n",
      "Test circle -591 trained in 2.84 seconds.\n",
      "Test circle -590 trained in 2.78 seconds.\n",
      "Test circle -589 trained in 2.78 seconds.\n",
      "Test circle -588 trained in 2.79 seconds.\n",
      "Test circle -587 trained in 2.85 seconds.\n",
      "Test circle -586 trained in 2.81 seconds.\n",
      "Test circle -585 trained in 2.76 seconds.\n",
      "Test circle -584 trained in 2.78 seconds.\n",
      "Test circle -583 trained in 2.81 seconds.\n",
      "Test circle -582 trained in 2.81 seconds.\n",
      "Test circle -581 trained in 2.77 seconds.\n",
      "Test circle -580 trained in 2.8 seconds.\n",
      "Test circle -579 trained in 2.76 seconds.\n",
      "Test circle -578 trained in 2.74 seconds.\n",
      "Test circle -577 trained in 2.76 seconds.\n",
      "Test circle -576 trained in 2.83 seconds.\n",
      "Test circle -575 trained in 2.79 seconds.\n",
      "Test circle -574 trained in 2.78 seconds.\n",
      "Test circle -573 trained in 2.75 seconds.\n",
      "Test circle -572 trained in 2.77 seconds.\n",
      "Test circle -571 trained in 2.88 seconds.\n",
      "Test circle -570 trained in 2.79 seconds.\n",
      "Test circle -569 trained in 2.91 seconds.\n",
      "Test circle -568 trained in 2.86 seconds.\n",
      "Test circle -567 trained in 2.8 seconds.\n",
      "Test circle -566 trained in 2.76 seconds.\n",
      "Test circle -565 trained in 2.77 seconds.\n",
      "Test circle -564 trained in 2.78 seconds.\n",
      "Test circle -563 trained in 2.77 seconds.\n",
      "Test circle -562 trained in 2.75 seconds.\n",
      "Test circle -561 trained in 2.89 seconds.\n",
      "Test circle -560 trained in 2.84 seconds.\n",
      "Test circle -559 trained in 2.76 seconds.\n",
      "Test circle -558 trained in 2.79 seconds.\n",
      "Test circle -557 trained in 2.81 seconds.\n",
      "Test circle -556 trained in 2.75 seconds.\n",
      "Test circle -555 trained in 2.74 seconds.\n",
      "Test circle -554 trained in 2.75 seconds.\n",
      "Test circle -553 trained in 2.74 seconds.\n",
      "Test circle -552 trained in 2.75 seconds.\n",
      "Test circle -551 trained in 2.94 seconds.\n",
      "Test circle -550 trained in 2.96 seconds.\n",
      "Test circle -549 trained in 2.82 seconds.\n",
      "Test circle -548 trained in 2.85 seconds.\n",
      "Test circle -547 trained in 2.82 seconds.\n",
      "Test circle -546 trained in 2.81 seconds.\n",
      "Test circle -545 trained in 2.93 seconds.\n",
      "Test circle -544 trained in 2.82 seconds.\n",
      "Test circle -543 trained in 2.82 seconds.\n",
      "Test circle -542 trained in 2.91 seconds.\n",
      "Test circle -541 trained in 2.89 seconds.\n",
      "Test circle -540 trained in 2.88 seconds.\n",
      "Test circle -539 trained in 2.96 seconds.\n",
      "Test circle -538 trained in 2.83 seconds.\n",
      "Test circle -537 trained in 2.93 seconds.\n",
      "Test circle -536 trained in 2.82 seconds.\n",
      "Test circle -535 trained in 2.77 seconds.\n",
      "Test circle -534 trained in 2.92 seconds.\n",
      "Test circle -533 trained in 2.75 seconds.\n",
      "Test circle -532 trained in 2.9 seconds.\n",
      "Test circle -531 trained in 2.85 seconds.\n",
      "Test circle -530 trained in 2.84 seconds.\n",
      "Test circle -529 trained in 2.8 seconds.\n",
      "Test circle -528 trained in 2.78 seconds.\n",
      "Test circle -527 trained in 2.87 seconds.\n",
      "Test circle -526 trained in 2.81 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -525 trained in 2.77 seconds.\n",
      "Test circle -524 trained in 2.76 seconds.\n",
      "Test circle -523 trained in 2.75 seconds.\n",
      "Test circle -522 trained in 2.75 seconds.\n",
      "Test circle -521 trained in 2.73 seconds.\n",
      "Test circle -520 trained in 2.75 seconds.\n",
      "Test circle -519 trained in 2.75 seconds.\n",
      "Test circle -518 trained in 2.79 seconds.\n",
      "Test circle -517 trained in 2.81 seconds.\n",
      "Test circle -516 trained in 2.77 seconds.\n",
      "Test circle -515 trained in 2.75 seconds.\n",
      "Test circle -514 trained in 2.78 seconds.\n",
      "Test circle -513 trained in 2.85 seconds.\n",
      "Test circle -512 trained in 2.92 seconds.\n",
      "Test circle -511 trained in 2.78 seconds.\n",
      "Test circle -510 trained in 2.78 seconds.\n",
      "Test circle -509 trained in 2.84 seconds.\n",
      "Test circle -508 trained in 2.79 seconds.\n",
      "Test circle -507 trained in 2.79 seconds.\n",
      "Test circle -506 trained in 2.86 seconds.\n",
      "Test circle -505 trained in 2.81 seconds.\n",
      "Test circle -504 trained in 2.79 seconds.\n",
      "Test circle -503 trained in 2.76 seconds.\n",
      "Test circle -502 trained in 2.91 seconds.\n",
      "Test circle -501 trained in 2.78 seconds.\n",
      "Test circle -500 trained in 2.84 seconds.\n",
      "Test circle -499 trained in 2.73 seconds.\n",
      "Test circle -498 trained in 2.91 seconds.\n",
      "Test circle -497 trained in 2.97 seconds.\n",
      "Test circle -496 trained in 2.76 seconds.\n",
      "Test circle -495 trained in 2.83 seconds.\n",
      "Test circle -494 trained in 2.76 seconds.\n",
      "Test circle -493 trained in 2.85 seconds.\n",
      "Test circle -492 trained in 2.9 seconds.\n",
      "Test circle -491 trained in 2.82 seconds.\n",
      "Test circle -490 trained in 2.89 seconds.\n",
      "Test circle -489 trained in 2.87 seconds.\n",
      "Test circle -488 trained in 2.91 seconds.\n",
      "Test circle -487 trained in 2.83 seconds.\n",
      "Test circle -486 trained in 2.79 seconds.\n",
      "Test circle -485 trained in 2.87 seconds.\n",
      "Test circle -484 trained in 2.88 seconds.\n",
      "Test circle -483 trained in 2.81 seconds.\n",
      "Test circle -482 trained in 2.74 seconds.\n",
      "Test circle -481 trained in 2.85 seconds.\n",
      "Test circle -480 trained in 2.8 seconds.\n",
      "Test circle -479 trained in 2.79 seconds.\n",
      "Test circle -478 trained in 2.89 seconds.\n",
      "Test circle -477 trained in 2.79 seconds.\n",
      "Test circle -476 trained in 2.86 seconds.\n",
      "Test circle -475 trained in 2.81 seconds.\n",
      "Test circle -474 trained in 2.87 seconds.\n",
      "Test circle -473 trained in 2.93 seconds.\n",
      "Test circle -472 trained in 2.75 seconds.\n",
      "Test circle -471 trained in 2.75 seconds.\n",
      "Test circle -470 trained in 2.74 seconds.\n",
      "Test circle -469 trained in 2.83 seconds.\n",
      "Test circle -468 trained in 3.81 seconds.\n",
      "Test circle -467 trained in 2.93 seconds.\n",
      "Test circle -466 trained in 2.8 seconds.\n",
      "Test circle -465 trained in 2.95 seconds.\n",
      "Test circle -464 trained in 2.9 seconds.\n",
      "Test circle -463 trained in 2.82 seconds.\n",
      "Test circle -462 trained in 2.87 seconds.\n",
      "Test circle -461 trained in 2.85 seconds.\n",
      "Test circle -460 trained in 2.98 seconds.\n",
      "Test circle -459 trained in 2.87 seconds.\n",
      "Test circle -458 trained in 2.92 seconds.\n",
      "Test circle -457 trained in 2.73 seconds.\n",
      "Test circle -456 trained in 2.76 seconds.\n",
      "Test circle -455 trained in 2.8 seconds.\n",
      "Test circle -454 trained in 2.82 seconds.\n",
      "Test circle -453 trained in 2.82 seconds.\n",
      "Test circle -452 trained in 2.76 seconds.\n",
      "Test circle -451 trained in 2.75 seconds.\n",
      "Test circle -450 trained in 2.73 seconds.\n",
      "Test circle -449 trained in 2.75 seconds.\n",
      "Test circle -448 trained in 2.75 seconds.\n",
      "Test circle -447 trained in 2.74 seconds.\n",
      "Test circle -446 trained in 2.74 seconds.\n",
      "Test circle -445 trained in 2.75 seconds.\n",
      "Test circle -444 trained in 2.76 seconds.\n",
      "Test circle -443 trained in 2.74 seconds.\n",
      "Test circle -442 trained in 2.83 seconds.\n",
      "Test circle -441 trained in 2.74 seconds.\n",
      "Test circle -440 trained in 2.81 seconds.\n",
      "Test circle -439 trained in 2.79 seconds.\n",
      "Test circle -438 trained in 3.23 seconds.\n",
      "Test circle -437 trained in 3.13 seconds.\n",
      "Test circle -436 trained in 2.92 seconds.\n",
      "Test circle -435 trained in 2.77 seconds.\n",
      "Test circle -434 trained in 2.81 seconds.\n",
      "Test circle -433 trained in 2.83 seconds.\n",
      "Test circle -432 trained in 2.81 seconds.\n",
      "Test circle -431 trained in 2.79 seconds.\n",
      "Test circle -430 trained in 2.8 seconds.\n",
      "Test circle -429 trained in 2.86 seconds.\n",
      "Test circle -428 trained in 2.95 seconds.\n",
      "Test circle -427 trained in 2.82 seconds.\n",
      "Test circle -426 trained in 2.76 seconds.\n",
      "Test circle -425 trained in 2.82 seconds.\n",
      "Test circle -424 trained in 3.29 seconds.\n",
      "Test circle -423 trained in 2.78 seconds.\n",
      "Test circle -422 trained in 2.9 seconds.\n",
      "Test circle -421 trained in 2.86 seconds.\n",
      "Test circle -420 trained in 2.76 seconds.\n",
      "Test circle -419 trained in 2.74 seconds.\n",
      "Test circle -418 trained in 2.83 seconds.\n",
      "Test circle -417 trained in 3.08 seconds.\n",
      "Test circle -416 trained in 2.94 seconds.\n",
      "Test circle -415 trained in 2.94 seconds.\n",
      "Test circle -414 trained in 2.8 seconds.\n",
      "Test circle -413 trained in 2.73 seconds.\n",
      "Test circle -412 trained in 2.77 seconds.\n",
      "Test circle -411 trained in 2.91 seconds.\n",
      "Test circle -410 trained in 2.79 seconds.\n",
      "Test circle -409 trained in 2.83 seconds.\n",
      "Test circle -408 trained in 2.8 seconds.\n",
      "Test circle -407 trained in 2.8 seconds.\n",
      "Test circle -406 trained in 2.81 seconds.\n",
      "Test circle -405 trained in 2.83 seconds.\n",
      "Test circle -404 trained in 2.83 seconds.\n",
      "Test circle -403 trained in 2.9 seconds.\n",
      "Test circle -402 trained in 2.85 seconds.\n",
      "Test circle -401 trained in 2.97 seconds.\n",
      "Test circle -400 trained in 2.94 seconds.\n",
      "Test circle -399 trained in 2.84 seconds.\n",
      "Test circle -398 trained in 2.86 seconds.\n",
      "Test circle -397 trained in 2.84 seconds.\n",
      "Test circle -396 trained in 2.76 seconds.\n",
      "Test circle -395 trained in 2.77 seconds.\n",
      "Test circle -394 trained in 2.74 seconds.\n",
      "Test circle -393 trained in 2.74 seconds.\n",
      "Test circle -392 trained in 2.75 seconds.\n",
      "Test circle -391 trained in 2.96 seconds.\n",
      "Test circle -390 trained in 3.05 seconds.\n",
      "Test circle -389 trained in 2.94 seconds.\n",
      "Test circle -388 trained in 2.89 seconds.\n",
      "Test circle -387 trained in 2.82 seconds.\n",
      "Test circle -386 trained in 2.97 seconds.\n",
      "Test circle -385 trained in 2.92 seconds.\n",
      "Test circle -384 trained in 2.89 seconds.\n",
      "Test circle -383 trained in 2.88 seconds.\n",
      "Test circle -382 trained in 2.84 seconds.\n",
      "Test circle -381 trained in 2.82 seconds.\n",
      "Test circle -380 trained in 2.75 seconds.\n",
      "Test circle -379 trained in 2.82 seconds.\n",
      "Test circle -378 trained in 2.83 seconds.\n",
      "Test circle -377 trained in 2.75 seconds.\n",
      "Test circle -376 trained in 2.81 seconds.\n",
      "Test circle -375 trained in 2.8 seconds.\n",
      "Test circle -374 trained in 2.87 seconds.\n",
      "Test circle -373 trained in 2.8 seconds.\n",
      "Test circle -372 trained in 2.77 seconds.\n",
      "Test circle -371 trained in 2.74 seconds.\n",
      "Test circle -370 trained in 2.82 seconds.\n",
      "Test circle -369 trained in 2.84 seconds.\n",
      "Test circle -368 trained in 2.84 seconds.\n",
      "Test circle -367 trained in 2.76 seconds.\n",
      "Test circle -366 trained in 2.8 seconds.\n",
      "Test circle -365 trained in 2.9 seconds.\n",
      "Test circle -364 trained in 2.76 seconds.\n",
      "Test circle -363 trained in 2.82 seconds.\n",
      "Test circle -362 trained in 2.97 seconds.\n",
      "Test circle -361 trained in 3.02 seconds.\n",
      "Test circle -360 trained in 2.91 seconds.\n",
      "Test circle -359 trained in 3.0 seconds.\n",
      "Test circle -358 trained in 2.84 seconds.\n",
      "Test circle -357 trained in 2.8 seconds.\n",
      "Test circle -356 trained in 2.74 seconds.\n",
      "Test circle -355 trained in 2.86 seconds.\n",
      "Test circle -354 trained in 2.84 seconds.\n",
      "Test circle -353 trained in 2.79 seconds.\n",
      "Test circle -352 trained in 2.75 seconds.\n",
      "Test circle -351 trained in 2.85 seconds.\n",
      "Test circle -350 trained in 2.83 seconds.\n",
      "Test circle -349 trained in 2.96 seconds.\n",
      "Test circle -348 trained in 2.85 seconds.\n",
      "Test circle -347 trained in 2.98 seconds.\n",
      "Test circle -346 trained in 2.87 seconds.\n",
      "Test circle -345 trained in 2.85 seconds.\n",
      "Test circle -344 trained in 2.86 seconds.\n",
      "Test circle -343 trained in 2.85 seconds.\n",
      "Test circle -342 trained in 2.9 seconds.\n",
      "Test circle -341 trained in 2.86 seconds.\n",
      "Test circle -340 trained in 2.89 seconds.\n",
      "Test circle -339 trained in 3.07 seconds.\n",
      "Test circle -338 trained in 2.89 seconds.\n",
      "Test circle -337 trained in 3.23 seconds.\n",
      "Test circle -336 trained in 2.95 seconds.\n",
      "Test circle -335 trained in 2.85 seconds.\n",
      "Test circle -334 trained in 2.79 seconds.\n",
      "Test circle -333 trained in 2.89 seconds.\n",
      "Test circle -332 trained in 2.88 seconds.\n",
      "Test circle -331 trained in 2.85 seconds.\n",
      "Test circle -330 trained in 2.96 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -329 trained in 3.41 seconds.\n",
      "Test circle -328 trained in 3.12 seconds.\n",
      "Test circle -327 trained in 3.37 seconds.\n",
      "Test circle -326 trained in 3.15 seconds.\n",
      "Test circle -325 trained in 2.85 seconds.\n",
      "Test circle -324 trained in 2.82 seconds.\n",
      "Test circle -323 trained in 2.76 seconds.\n",
      "Test circle -322 trained in 2.74 seconds.\n",
      "Test circle -321 trained in 2.75 seconds.\n",
      "Test circle -320 trained in 3.06 seconds.\n",
      "Test circle -319 trained in 2.76 seconds.\n",
      "Test circle -318 trained in 2.74 seconds.\n",
      "Test circle -317 trained in 2.97 seconds.\n",
      "Test circle -316 trained in 3.1 seconds.\n",
      "Test circle -315 trained in 2.83 seconds.\n",
      "Test circle -314 trained in 2.78 seconds.\n",
      "Test circle -313 trained in 2.8 seconds.\n",
      "Test circle -312 trained in 2.64 seconds.\n",
      "Test circle -311 trained in 2.59 seconds.\n",
      "Test circle -310 trained in 2.6 seconds.\n",
      "Test circle -309 trained in 2.66 seconds.\n",
      "Test circle -308 trained in 2.65 seconds.\n",
      "Test circle -307 trained in 2.59 seconds.\n",
      "Test circle -306 trained in 2.68 seconds.\n",
      "Test circle -305 trained in 2.65 seconds.\n",
      "Test circle -304 trained in 2.66 seconds.\n",
      "Test circle -303 trained in 2.58 seconds.\n",
      "Test circle -302 trained in 2.61 seconds.\n",
      "Test circle -301 trained in 2.66 seconds.\n",
      "Test circle -300 trained in 2.62 seconds.\n",
      "Test circle -299 trained in 2.62 seconds.\n",
      "Test circle -298 trained in 2.77 seconds.\n",
      "Test circle -297 trained in 2.71 seconds.\n",
      "Test circle -296 trained in 2.68 seconds.\n",
      "Test circle -295 trained in 2.73 seconds.\n",
      "Test circle -294 trained in 2.75 seconds.\n",
      "Test circle -293 trained in 2.73 seconds.\n",
      "Test circle -292 trained in 2.88 seconds.\n",
      "Test circle -291 trained in 2.78 seconds.\n",
      "Test circle -290 trained in 2.79 seconds.\n",
      "Test circle -289 trained in 2.77 seconds.\n",
      "Test circle -288 trained in 2.71 seconds.\n",
      "Test circle -287 trained in 2.69 seconds.\n",
      "Test circle -286 trained in 2.67 seconds.\n",
      "Test circle -285 trained in 2.72 seconds.\n",
      "Test circle -284 trained in 2.66 seconds.\n",
      "Test circle -283 trained in 2.82 seconds.\n",
      "Test circle -282 trained in 2.73 seconds.\n",
      "Test circle -281 trained in 2.71 seconds.\n",
      "Test circle -280 trained in 2.74 seconds.\n",
      "Test circle -279 trained in 2.93 seconds.\n",
      "Test circle -278 trained in 2.71 seconds.\n",
      "Test circle -277 trained in 2.73 seconds.\n",
      "Test circle -276 trained in 2.77 seconds.\n",
      "Test circle -275 trained in 2.74 seconds.\n",
      "Test circle -274 trained in 2.79 seconds.\n",
      "Test circle -273 trained in 2.74 seconds.\n",
      "Test circle -272 trained in 2.7 seconds.\n",
      "Test circle -271 trained in 2.81 seconds.\n",
      "Test circle -270 trained in 2.76 seconds.\n",
      "Test circle -269 trained in 2.69 seconds.\n",
      "Test circle -268 trained in 2.82 seconds.\n",
      "Test circle -267 trained in 2.79 seconds.\n",
      "Test circle -266 trained in 2.7 seconds.\n",
      "Test circle -265 trained in 2.78 seconds.\n",
      "Test circle -264 trained in 2.86 seconds.\n",
      "Test circle -263 trained in 2.75 seconds.\n",
      "Test circle -262 trained in 2.77 seconds.\n",
      "Test circle -261 trained in 2.77 seconds.\n",
      "Test circle -260 trained in 2.73 seconds.\n",
      "Test circle -259 trained in 2.81 seconds.\n",
      "Test circle -258 trained in 3.15 seconds.\n",
      "Test circle -257 trained in 2.69 seconds.\n",
      "Test circle -256 trained in 2.77 seconds.\n",
      "Test circle -255 trained in 2.75 seconds.\n",
      "Test circle -254 trained in 2.68 seconds.\n",
      "Test circle -253 trained in 2.81 seconds.\n",
      "Test circle -252 trained in 2.77 seconds.\n",
      "Test circle -251 trained in 2.73 seconds.\n",
      "Test circle -250 trained in 2.8 seconds.\n",
      "Test circle -249 trained in 2.76 seconds.\n",
      "Test circle -248 trained in 2.72 seconds.\n",
      "Test circle -247 trained in 2.77 seconds.\n",
      "Test circle -246 trained in 2.74 seconds.\n",
      "Test circle -245 trained in 2.75 seconds.\n",
      "Test circle -244 trained in 2.79 seconds.\n",
      "Test circle -243 trained in 2.77 seconds.\n",
      "Test circle -242 trained in 2.74 seconds.\n",
      "Test circle -241 trained in 2.79 seconds.\n",
      "Test circle -240 trained in 2.76 seconds.\n",
      "Test circle -239 trained in 2.72 seconds.\n",
      "Test circle -238 trained in 2.77 seconds.\n",
      "Test circle -237 trained in 2.77 seconds.\n",
      "Test circle -236 trained in 2.72 seconds.\n",
      "Test circle -235 trained in 2.78 seconds.\n",
      "Test circle -234 trained in 2.87 seconds.\n",
      "Test circle -233 trained in 2.75 seconds.\n",
      "Test circle -232 trained in 2.82 seconds.\n",
      "Test circle -231 trained in 2.76 seconds.\n",
      "Test circle -230 trained in 2.72 seconds.\n",
      "Test circle -229 trained in 2.84 seconds.\n",
      "Test circle -228 trained in 2.79 seconds.\n",
      "Test circle -227 trained in 2.73 seconds.\n",
      "Test circle -226 trained in 2.73 seconds.\n",
      "Test circle -225 trained in 2.79 seconds.\n",
      "Test circle -224 trained in 2.71 seconds.\n",
      "Test circle -223 trained in 2.97 seconds.\n",
      "Test circle -222 trained in 2.75 seconds.\n",
      "Test circle -221 trained in 2.75 seconds.\n",
      "Test circle -220 trained in 2.83 seconds.\n",
      "Test circle -219 trained in 2.76 seconds.\n",
      "Test circle -218 trained in 2.74 seconds.\n",
      "Test circle -217 trained in 2.78 seconds.\n",
      "Test circle -216 trained in 2.77 seconds.\n",
      "Test circle -215 trained in 2.8 seconds.\n",
      "Test circle -214 trained in 2.76 seconds.\n",
      "Test circle -213 trained in 2.75 seconds.\n",
      "Test circle -212 trained in 2.69 seconds.\n",
      "Test circle -211 trained in 2.75 seconds.\n",
      "Test circle -210 trained in 2.71 seconds.\n",
      "Test circle -209 trained in 2.7 seconds.\n",
      "Test circle -208 trained in 2.79 seconds.\n",
      "Test circle -207 trained in 2.76 seconds.\n",
      "Test circle -206 trained in 2.7 seconds.\n",
      "Test circle -205 trained in 2.76 seconds.\n",
      "Test circle -204 trained in 2.76 seconds.\n",
      "Test circle -203 trained in 2.7 seconds.\n",
      "Test circle -202 trained in 2.75 seconds.\n",
      "Test circle -201 trained in 2.74 seconds.\n",
      "Test circle -200 trained in 2.69 seconds.\n",
      "Test circle -199 trained in 2.78 seconds.\n",
      "Test circle -198 trained in 2.79 seconds.\n",
      "Test circle -197 trained in 2.78 seconds.\n",
      "Test circle -196 trained in 2.74 seconds.\n",
      "Test circle -195 trained in 2.76 seconds.\n",
      "Test circle -194 trained in 2.7 seconds.\n",
      "Test circle -193 trained in 2.75 seconds.\n",
      "Test circle -192 trained in 2.75 seconds.\n",
      "Test circle -191 trained in 2.68 seconds.\n",
      "Test circle -190 trained in 2.71 seconds.\n",
      "Test circle -189 trained in 2.79 seconds.\n",
      "Test circle -188 trained in 2.69 seconds.\n",
      "Test circle -187 trained in 2.73 seconds.\n",
      "Test circle -186 trained in 2.87 seconds.\n",
      "Test circle -185 trained in 2.71 seconds.\n",
      "Test circle -184 trained in 2.73 seconds.\n",
      "Test circle -183 trained in 2.81 seconds.\n",
      "Test circle -182 trained in 2.74 seconds.\n",
      "Test circle -181 trained in 2.73 seconds.\n",
      "Test circle -180 trained in 2.85 seconds.\n",
      "Test circle -179 trained in 2.83 seconds.\n",
      "Test circle -178 trained in 2.75 seconds.\n",
      "Test circle -177 trained in 2.93 seconds.\n",
      "Test circle -176 trained in 2.78 seconds.\n",
      "Test circle -175 trained in 2.78 seconds.\n",
      "Test circle -174 trained in 2.91 seconds.\n",
      "Test circle -173 trained in 2.76 seconds.\n",
      "Test circle -172 trained in 2.83 seconds.\n",
      "Test circle -171 trained in 2.86 seconds.\n",
      "Test circle -170 trained in 2.78 seconds.\n",
      "Test circle -169 trained in 2.83 seconds.\n",
      "Test circle -168 trained in 2.88 seconds.\n",
      "Test circle -167 trained in 2.82 seconds.\n",
      "Test circle -166 trained in 2.91 seconds.\n",
      "Test circle -165 trained in 2.93 seconds.\n",
      "Test circle -164 trained in 2.94 seconds.\n",
      "Test circle -163 trained in 2.97 seconds.\n",
      "Test circle -162 trained in 2.94 seconds.\n",
      "Test circle -161 trained in 3.03 seconds.\n",
      "Test circle -160 trained in 3.1 seconds.\n",
      "Test circle -159 trained in 2.94 seconds.\n",
      "Test circle -158 trained in 2.92 seconds.\n",
      "Test circle -157 trained in 3.13 seconds.\n",
      "Test circle -156 trained in 3.08 seconds.\n",
      "Test circle -155 trained in 3.05 seconds.\n",
      "Test circle -154 trained in 3.1 seconds.\n",
      "Test circle -153 trained in 3.03 seconds.\n",
      "Test circle -152 trained in 2.85 seconds.\n",
      "Test circle -151 trained in 2.67 seconds.\n",
      "Test circle -150 trained in 2.67 seconds.\n",
      "Test circle -149 trained in 4.3 seconds.\n",
      "Test circle -148 trained in 3.0 seconds.\n",
      "Test circle -147 trained in 3.29 seconds.\n",
      "Test circle -146 trained in 3.0 seconds.\n",
      "Test circle -145 trained in 2.95 seconds.\n",
      "Test circle -144 trained in 2.91 seconds.\n",
      "Test circle -143 trained in 2.97 seconds.\n",
      "Test circle -142 trained in 3.26 seconds.\n",
      "Test circle -141 trained in 2.97 seconds.\n",
      "Test circle -140 trained in 2.94 seconds.\n",
      "Test circle -139 trained in 3.18 seconds.\n",
      "Test circle -138 trained in 3.69 seconds.\n",
      "Test circle -137 trained in 2.99 seconds.\n",
      "Test circle -136 trained in 3.07 seconds.\n",
      "Test circle -135 trained in 3.02 seconds.\n",
      "Test circle -134 trained in 3.19 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -133 trained in 2.83 seconds.\n",
      "Test circle -132 trained in 2.77 seconds.\n",
      "Test circle -131 trained in 2.73 seconds.\n",
      "Test circle -130 trained in 3.55 seconds.\n",
      "Test circle -129 trained in 3.51 seconds.\n",
      "Test circle -128 trained in 2.75 seconds.\n",
      "Test circle -127 trained in 2.82 seconds.\n",
      "Test circle -126 trained in 2.8 seconds.\n",
      "Test circle -125 trained in 3.21 seconds.\n",
      "Test circle -124 trained in 3.05 seconds.\n",
      "Test circle -123 trained in 3.04 seconds.\n",
      "Test circle -122 trained in 3.07 seconds.\n",
      "Test circle -121 trained in 2.99 seconds.\n",
      "Test circle -120 trained in 2.9 seconds.\n",
      "Test circle -119 trained in 2.86 seconds.\n",
      "Test circle -118 trained in 2.78 seconds.\n",
      "Test circle -117 trained in 2.88 seconds.\n",
      "Test circle -116 trained in 2.83 seconds.\n",
      "Test circle -115 trained in 2.91 seconds.\n",
      "Test circle -114 trained in 2.98 seconds.\n",
      "Test circle -113 trained in 2.98 seconds.\n",
      "Test circle -112 trained in 2.71 seconds.\n",
      "Test circle -111 trained in 2.81 seconds.\n",
      "Test circle -110 trained in 3.0 seconds.\n",
      "Test circle -109 trained in 2.86 seconds.\n",
      "Test circle -108 trained in 2.8 seconds.\n",
      "Test circle -107 trained in 3.08 seconds.\n",
      "Test circle -106 trained in 2.75 seconds.\n",
      "Test circle -105 trained in 2.81 seconds.\n",
      "Test circle -104 trained in 2.81 seconds.\n",
      "Test circle -103 trained in 2.75 seconds.\n",
      "Test circle -102 trained in 2.83 seconds.\n",
      "Test circle -101 trained in 2.81 seconds.\n",
      "Test circle -100 trained in 2.79 seconds.\n",
      "Test circle -99 trained in 2.94 seconds.\n",
      "Test circle -98 trained in 2.81 seconds.\n",
      "Test circle -97 trained in 2.75 seconds.\n",
      "Test circle -96 trained in 2.83 seconds.\n",
      "Test circle -95 trained in 2.76 seconds.\n",
      "Test circle -94 trained in 2.74 seconds.\n",
      "Test circle -93 trained in 2.8 seconds.\n",
      "Test circle -92 trained in 2.75 seconds.\n",
      "Test circle -91 trained in 2.72 seconds.\n",
      "Test circle -90 trained in 2.85 seconds.\n",
      "Test circle -89 trained in 2.89 seconds.\n",
      "Test circle -88 trained in 2.82 seconds.\n",
      "Test circle -87 trained in 2.97 seconds.\n",
      "Test circle -86 trained in 2.89 seconds.\n",
      "Test circle -85 trained in 3.1 seconds.\n",
      "Test circle -84 trained in 3.0 seconds.\n",
      "Test circle -83 trained in 2.82 seconds.\n",
      "Test circle -82 trained in 2.92 seconds.\n",
      "Test circle -81 trained in 3.05 seconds.\n",
      "Test circle -80 trained in 2.94 seconds.\n",
      "Test circle -79 trained in 3.03 seconds.\n",
      "Test circle -78 trained in 3.01 seconds.\n",
      "Test circle -77 trained in 3.0 seconds.\n",
      "Test circle -76 trained in 3.02 seconds.\n",
      "Test circle -75 trained in 2.99 seconds.\n",
      "Test circle -74 trained in 2.92 seconds.\n",
      "Test circle -73 trained in 3.11 seconds.\n",
      "Test circle -72 trained in 3.1 seconds.\n",
      "Test circle -71 trained in 3.27 seconds.\n",
      "Test circle -70 trained in 3.19 seconds.\n",
      "Test circle -69 trained in 3.17 seconds.\n",
      "Test circle -68 trained in 3.13 seconds.\n",
      "Test circle -67 trained in 3.25 seconds.\n",
      "Test circle -66 trained in 3.09 seconds.\n",
      "Test circle -65 trained in 3.47 seconds.\n",
      "Test circle -64 trained in 3.23 seconds.\n",
      "Test circle -63 trained in 3.18 seconds.\n",
      "Test circle -62 trained in 3.3 seconds.\n",
      "Test circle -61 trained in 3.26 seconds.\n",
      "Test circle -60 trained in 3.11 seconds.\n",
      "Test circle -59 trained in 2.59 seconds.\n",
      "Test circle -58 trained in 2.7 seconds.\n",
      "Test circle -57 trained in 2.71 seconds.\n",
      "Test circle -56 trained in 2.93 seconds.\n",
      "Test circle -55 trained in 2.74 seconds.\n",
      "Test circle -54 trained in 2.83 seconds.\n",
      "Test circle -53 trained in 2.74 seconds.\n",
      "Test circle -52 trained in 2.8 seconds.\n",
      "Test circle -51 trained in 2.76 seconds.\n",
      "Test circle -50 trained in 2.69 seconds.\n",
      "Test circle -49 trained in 2.71 seconds.\n",
      "Test circle -48 trained in 2.76 seconds.\n",
      "Test circle -47 trained in 2.73 seconds.\n",
      "Test circle -46 trained in 2.77 seconds.\n",
      "Test circle -45 trained in 2.8 seconds.\n",
      "Test circle -44 trained in 2.74 seconds.\n",
      "Test circle -43 trained in 2.74 seconds.\n",
      "Test circle -42 trained in 2.83 seconds.\n",
      "Test circle -41 trained in 2.74 seconds.\n",
      "Test circle -40 trained in 2.72 seconds.\n",
      "Test circle -39 trained in 2.84 seconds.\n",
      "Test circle -38 trained in 2.73 seconds.\n",
      "Test circle -37 trained in 2.76 seconds.\n",
      "Test circle -36 trained in 2.83 seconds.\n",
      "Test circle -35 trained in 2.82 seconds.\n",
      "Test circle -34 trained in 2.84 seconds.\n",
      "Test circle -33 trained in 2.92 seconds.\n",
      "Test circle -32 trained in 2.67 seconds.\n",
      "Test circle -31 trained in 2.67 seconds.\n",
      "Test circle -30 trained in 2.82 seconds.\n",
      "Test circle -29 trained in 2.79 seconds.\n",
      "Test circle -28 trained in 2.77 seconds.\n",
      "Test circle -27 trained in 2.83 seconds.\n",
      "Test circle -26 trained in 2.77 seconds.\n",
      "Test circle -25 trained in 2.78 seconds.\n",
      "Test circle -24 trained in 2.89 seconds.\n",
      "Test circle -23 trained in 2.85 seconds.\n",
      "Test circle -22 trained in 2.95 seconds.\n",
      "Test circle -21 trained in 2.94 seconds.\n",
      "Test circle -20 trained in 2.82 seconds.\n",
      "Test circle -19 trained in 2.9 seconds.\n",
      "Test circle -18 trained in 2.88 seconds.\n",
      "Test circle -17 trained in 2.87 seconds.\n",
      "Test circle -16 trained in 2.98 seconds.\n",
      "Test circle -15 trained in 3.24 seconds.\n",
      "Test circle -14 trained in 2.9 seconds.\n",
      "Test circle -13 trained in 3.07 seconds.\n",
      "Test circle -12 trained in 2.93 seconds.\n",
      "Test circle -11 trained in 2.9 seconds.\n",
      "Test circle -10 trained in 3.09 seconds.\n",
      "Test circle -9 trained in 2.94 seconds.\n",
      "Test circle -8 trained in 2.98 seconds.\n",
      "Test circle -7 trained in 3.04 seconds.\n",
      "Test circle -6 trained in 3.0 seconds.\n",
      "Test circle -5 trained in 3.02 seconds.\n",
      "Test circle -4 trained in 3.14 seconds.\n",
      "Test circle -3 trained in 3.14 seconds.\n",
      "Test circle -2 trained in 3.16 seconds.\n",
      "Test circle -1 trained in 3.2 seconds.\n",
      "Model trained in 2068.36 seconds.\n",
      "Hyperparams (50, 3, 3) trained in 2068.36 seconds.\n",
      "=========================================================================================\n",
      "Model (50, 3, 3) Ended\n",
      "=========================================================================================\n",
      "=========================================================================================\n",
      "Model (50, 3, 6) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 3.17 seconds.\n",
      "Test circle -724 trained in 3.27 seconds.\n",
      "Test circle -723 trained in 3.22 seconds.\n",
      "Test circle -722 trained in 3.42 seconds.\n",
      "Test circle -721 trained in 3.27 seconds.\n",
      "Test circle -720 trained in 3.18 seconds.\n",
      "Test circle -719 trained in 2.86 seconds.\n",
      "Test circle -718 trained in 2.61 seconds.\n",
      "Test circle -717 trained in 2.67 seconds.\n",
      "Test circle -716 trained in 2.7 seconds.\n",
      "Test circle -715 trained in 2.63 seconds.\n",
      "Test circle -714 trained in 2.65 seconds.\n",
      "Test circle -713 trained in 2.76 seconds.\n",
      "Test circle -712 trained in 2.67 seconds.\n",
      "Test circle -711 trained in 2.66 seconds.\n",
      "Test circle -710 trained in 2.77 seconds.\n",
      "Test circle -709 trained in 2.68 seconds.\n",
      "Test circle -708 trained in 2.69 seconds.\n",
      "Test circle -707 trained in 2.74 seconds.\n",
      "Test circle -706 trained in 2.73 seconds.\n",
      "Test circle -705 trained in 2.71 seconds.\n",
      "Test circle -704 trained in 2.93 seconds.\n",
      "Test circle -703 trained in 2.71 seconds.\n",
      "Test circle -702 trained in 2.72 seconds.\n",
      "Test circle -701 trained in 2.83 seconds.\n",
      "Test circle -700 trained in 2.72 seconds.\n",
      "Test circle -699 trained in 2.71 seconds.\n",
      "Test circle -698 trained in 2.83 seconds.\n",
      "Test circle -697 trained in 2.76 seconds.\n",
      "Test circle -696 trained in 2.73 seconds.\n",
      "Test circle -695 trained in 2.81 seconds.\n",
      "Test circle -694 trained in 2.72 seconds.\n",
      "Test circle -693 trained in 2.71 seconds.\n",
      "Test circle -692 trained in 2.77 seconds.\n",
      "Test circle -691 trained in 2.72 seconds.\n",
      "Test circle -690 trained in 2.7 seconds.\n",
      "Test circle -689 trained in 2.86 seconds.\n",
      "Test circle -688 trained in 2.84 seconds.\n",
      "Test circle -687 trained in 2.72 seconds.\n",
      "Test circle -686 trained in 3.01 seconds.\n",
      "Test circle -685 trained in 2.89 seconds.\n",
      "Test circle -684 trained in 2.88 seconds.\n",
      "Test circle -683 trained in 2.94 seconds.\n",
      "Test circle -682 trained in 2.86 seconds.\n",
      "Test circle -681 trained in 2.82 seconds.\n",
      "Test circle -680 trained in 2.93 seconds.\n",
      "Test circle -679 trained in 2.9 seconds.\n",
      "Test circle -678 trained in 2.93 seconds.\n",
      "Test circle -677 trained in 2.94 seconds.\n",
      "Test circle -676 trained in 2.87 seconds.\n",
      "Test circle -675 trained in 3.03 seconds.\n",
      "Test circle -674 trained in 3.05 seconds.\n",
      "Test circle -673 trained in 2.95 seconds.\n",
      "Test circle -672 trained in 2.99 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -671 trained in 3.09 seconds.\n",
      "Test circle -670 trained in 3.27 seconds.\n",
      "Test circle -669 trained in 3.17 seconds.\n",
      "Test circle -668 trained in 3.09 seconds.\n",
      "Test circle -667 trained in 3.2 seconds.\n",
      "Test circle -666 trained in 3.12 seconds.\n",
      "Test circle -665 trained in 3.14 seconds.\n",
      "Test circle -664 trained in 3.26 seconds.\n",
      "Test circle -663 trained in 3.14 seconds.\n",
      "Test circle -662 trained in 3.3 seconds.\n",
      "Test circle -661 trained in 3.24 seconds.\n",
      "Test circle -660 trained in 3.21 seconds.\n",
      "Test circle -659 trained in 3.39 seconds.\n",
      "Test circle -658 trained in 3.31 seconds.\n",
      "Test circle -657 trained in 3.34 seconds.\n",
      "Test circle -656 trained in 3.35 seconds.\n",
      "Test circle -655 trained in 3.36 seconds.\n",
      "Test circle -654 trained in 3.47 seconds.\n",
      "Test circle -653 trained in 3.37 seconds.\n",
      "Test circle -652 trained in 2.69 seconds.\n",
      "Test circle -651 trained in 2.7 seconds.\n",
      "Test circle -650 trained in 2.63 seconds.\n",
      "Test circle -649 trained in 2.69 seconds.\n",
      "Test circle -648 trained in 2.66 seconds.\n",
      "Test circle -647 trained in 2.64 seconds.\n",
      "Test circle -646 trained in 2.7 seconds.\n",
      "Test circle -645 trained in 2.73 seconds.\n",
      "Test circle -644 trained in 2.68 seconds.\n",
      "Test circle -643 trained in 2.73 seconds.\n",
      "Test circle -642 trained in 2.73 seconds.\n",
      "Test circle -641 trained in 2.67 seconds.\n",
      "Test circle -640 trained in 2.74 seconds.\n",
      "Test circle -639 trained in 2.71 seconds.\n",
      "Test circle -638 trained in 2.73 seconds.\n",
      "Test circle -637 trained in 2.79 seconds.\n",
      "Test circle -636 trained in 2.75 seconds.\n",
      "Test circle -635 trained in 2.73 seconds.\n",
      "Test circle -634 trained in 2.76 seconds.\n",
      "Test circle -633 trained in 2.84 seconds.\n",
      "Test circle -632 trained in 2.72 seconds.\n",
      "Test circle -631 trained in 2.75 seconds.\n",
      "Test circle -630 trained in 2.76 seconds.\n",
      "Test circle -629 trained in 2.7 seconds.\n",
      "Test circle -628 trained in 2.77 seconds.\n",
      "Test circle -627 trained in 2.77 seconds.\n",
      "Test circle -626 trained in 2.71 seconds.\n",
      "Test circle -625 trained in 2.79 seconds.\n",
      "Test circle -624 trained in 2.76 seconds.\n",
      "Test circle -623 trained in 2.79 seconds.\n",
      "Test circle -622 trained in 2.82 seconds.\n",
      "Test circle -621 trained in 2.78 seconds.\n",
      "Test circle -620 trained in 2.77 seconds.\n",
      "Test circle -619 trained in 2.82 seconds.\n",
      "Test circle -618 trained in 2.77 seconds.\n",
      "Test circle -617 trained in 2.71 seconds.\n",
      "Test circle -616 trained in 2.77 seconds.\n",
      "Test circle -615 trained in 3.19 seconds.\n",
      "Test circle -614 trained in 2.79 seconds.\n",
      "Test circle -613 trained in 2.97 seconds.\n",
      "Test circle -612 trained in 2.98 seconds.\n",
      "Test circle -611 trained in 2.81 seconds.\n",
      "Test circle -610 trained in 3.12 seconds.\n",
      "Test circle -609 trained in 2.83 seconds.\n",
      "Test circle -608 trained in 2.95 seconds.\n",
      "Test circle -607 trained in 2.86 seconds.\n",
      "Test circle -606 trained in 3.26 seconds.\n",
      "Test circle -605 trained in 2.88 seconds.\n",
      "Test circle -604 trained in 2.79 seconds.\n",
      "Test circle -603 trained in 2.74 seconds.\n",
      "Test circle -602 trained in 3.8 seconds.\n",
      "Test circle -601 trained in 2.84 seconds.\n",
      "Test circle -600 trained in 2.74 seconds.\n",
      "Test circle -599 trained in 2.8 seconds.\n",
      "Test circle -598 trained in 2.95 seconds.\n",
      "Test circle -597 trained in 3.52 seconds.\n",
      "Test circle -596 trained in 3.66 seconds.\n",
      "Test circle -595 trained in 3.66 seconds.\n",
      "Test circle -594 trained in 3.34 seconds.\n",
      "Test circle -593 trained in 4.57 seconds.\n",
      "Test circle -592 trained in 3.03 seconds.\n",
      "Test circle -591 trained in 2.94 seconds.\n",
      "Test circle -590 trained in 2.88 seconds.\n",
      "Test circle -589 trained in 2.98 seconds.\n",
      "Test circle -588 trained in 3.11 seconds.\n",
      "Test circle -587 trained in 2.94 seconds.\n",
      "Test circle -586 trained in 3.04 seconds.\n",
      "Test circle -585 trained in 2.94 seconds.\n",
      "Test circle -584 trained in 2.97 seconds.\n",
      "Test circle -583 trained in 3.26 seconds.\n",
      "Test circle -582 trained in 3.0 seconds.\n",
      "Test circle -581 trained in 3.04 seconds.\n",
      "Test circle -580 trained in 3.1 seconds.\n",
      "Test circle -579 trained in 3.15 seconds.\n",
      "Test circle -578 trained in 3.14 seconds.\n",
      "Test circle -577 trained in 3.16 seconds.\n",
      "Test circle -576 trained in 3.2 seconds.\n",
      "Test circle -575 trained in 3.36 seconds.\n",
      "Test circle -574 trained in 3.22 seconds.\n",
      "Test circle -573 trained in 3.3 seconds.\n",
      "Test circle -572 trained in 3.28 seconds.\n",
      "Test circle -571 trained in 3.28 seconds.\n",
      "Test circle -570 trained in 3.32 seconds.\n",
      "Test circle -569 trained in 3.36 seconds.\n",
      "Test circle -568 trained in 3.64 seconds.\n",
      "Test circle -567 trained in 3.55 seconds.\n",
      "Test circle -566 trained in 3.43 seconds.\n",
      "Test circle -565 trained in 3.54 seconds.\n",
      "Test circle -564 trained in 3.47 seconds.\n",
      "Test circle -563 trained in 3.55 seconds.\n",
      "Test circle -562 trained in 2.94 seconds.\n",
      "Test circle -561 trained in 2.66 seconds.\n",
      "Test circle -560 trained in 2.67 seconds.\n",
      "Test circle -559 trained in 2.63 seconds.\n",
      "Test circle -558 trained in 2.68 seconds.\n",
      "Test circle -557 trained in 2.72 seconds.\n",
      "Test circle -556 trained in 2.71 seconds.\n",
      "Test circle -555 trained in 2.72 seconds.\n",
      "Test circle -554 trained in 2.74 seconds.\n",
      "Test circle -553 trained in 2.68 seconds.\n",
      "Test circle -552 trained in 2.69 seconds.\n",
      "Test circle -551 trained in 2.76 seconds.\n",
      "Test circle -550 trained in 2.7 seconds.\n",
      "Test circle -549 trained in 2.7 seconds.\n",
      "Test circle -548 trained in 2.8 seconds.\n",
      "Test circle -547 trained in 2.75 seconds.\n",
      "Test circle -546 trained in 2.82 seconds.\n",
      "Test circle -545 trained in 2.99 seconds.\n",
      "Test circle -544 trained in 2.91 seconds.\n",
      "Test circle -543 trained in 2.88 seconds.\n",
      "Test circle -542 trained in 2.83 seconds.\n",
      "Test circle -541 trained in 2.79 seconds.\n",
      "Test circle -540 trained in 2.77 seconds.\n",
      "Test circle -539 trained in 2.84 seconds.\n",
      "Test circle -538 trained in 2.78 seconds.\n",
      "Test circle -537 trained in 2.85 seconds.\n",
      "Test circle -536 trained in 2.98 seconds.\n",
      "Test circle -535 trained in 2.98 seconds.\n",
      "Test circle -534 trained in 2.8 seconds.\n",
      "Test circle -533 trained in 2.78 seconds.\n",
      "Test circle -532 trained in 2.78 seconds.\n",
      "Test circle -531 trained in 2.87 seconds.\n",
      "Test circle -530 trained in 2.88 seconds.\n",
      "Test circle -529 trained in 2.77 seconds.\n",
      "Test circle -528 trained in 2.79 seconds.\n",
      "Test circle -527 trained in 2.74 seconds.\n",
      "Test circle -526 trained in 2.85 seconds.\n",
      "Test circle -525 trained in 2.85 seconds.\n",
      "Test circle -524 trained in 2.96 seconds.\n",
      "Test circle -523 trained in 3.07 seconds.\n",
      "Test circle -522 trained in 3.11 seconds.\n",
      "Test circle -521 trained in 2.87 seconds.\n",
      "Test circle -520 trained in 3.02 seconds.\n",
      "Test circle -519 trained in 3.03 seconds.\n",
      "Test circle -518 trained in 2.92 seconds.\n",
      "Test circle -517 trained in 2.99 seconds.\n",
      "Test circle -516 trained in 3.09 seconds.\n",
      "Test circle -515 trained in 2.99 seconds.\n",
      "Test circle -514 trained in 3.09 seconds.\n",
      "Test circle -513 trained in 3.09 seconds.\n",
      "Test circle -512 trained in 3.01 seconds.\n",
      "Test circle -511 trained in 2.97 seconds.\n",
      "Test circle -510 trained in 2.89 seconds.\n",
      "Test circle -509 trained in 3.22 seconds.\n",
      "Test circle -508 trained in 3.09 seconds.\n",
      "Test circle -507 trained in 3.04 seconds.\n",
      "Test circle -506 trained in 3.08 seconds.\n",
      "Test circle -505 trained in 3.04 seconds.\n",
      "Test circle -504 trained in 2.97 seconds.\n",
      "Test circle -503 trained in 3.12 seconds.\n",
      "Test circle -502 trained in 3.09 seconds.\n",
      "Test circle -501 trained in 3.06 seconds.\n",
      "Test circle -500 trained in 3.15 seconds.\n",
      "Test circle -499 trained in 3.19 seconds.\n",
      "Test circle -498 trained in 3.13 seconds.\n",
      "Test circle -497 trained in 3.29 seconds.\n",
      "Test circle -496 trained in 3.29 seconds.\n",
      "Test circle -495 trained in 3.36 seconds.\n",
      "Test circle -494 trained in 3.23 seconds.\n",
      "Test circle -493 trained in 3.25 seconds.\n",
      "Test circle -492 trained in 3.48 seconds.\n",
      "Test circle -491 trained in 3.32 seconds.\n",
      "Test circle -490 trained in 3.45 seconds.\n",
      "Test circle -489 trained in 3.45 seconds.\n",
      "Test circle -488 trained in 4.43 seconds.\n",
      "Test circle -487 trained in 3.77 seconds.\n",
      "Test circle -486 trained in 2.65 seconds.\n",
      "Test circle -485 trained in 2.79 seconds.\n",
      "Test circle -484 trained in 2.69 seconds.\n",
      "Test circle -483 trained in 2.72 seconds.\n",
      "Test circle -482 trained in 2.72 seconds.\n",
      "Test circle -481 trained in 2.64 seconds.\n",
      "Test circle -480 trained in 2.94 seconds.\n",
      "Test circle -479 trained in 3.41 seconds.\n",
      "Test circle -478 trained in 2.74 seconds.\n",
      "Test circle -477 trained in 2.83 seconds.\n",
      "Test circle -476 trained in 2.86 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -475 trained in 3.08 seconds.\n",
      "Test circle -474 trained in 3.22 seconds.\n",
      "Test circle -473 trained in 3.01 seconds.\n",
      "Test circle -472 trained in 3.05 seconds.\n",
      "Test circle -471 trained in 2.97 seconds.\n",
      "Test circle -470 trained in 3.09 seconds.\n",
      "Test circle -469 trained in 2.97 seconds.\n",
      "Test circle -468 trained in 2.97 seconds.\n",
      "Test circle -467 trained in 2.87 seconds.\n",
      "Test circle -466 trained in 2.96 seconds.\n",
      "Test circle -465 trained in 2.94 seconds.\n",
      "Test circle -464 trained in 2.81 seconds.\n",
      "Test circle -463 trained in 2.79 seconds.\n",
      "Test circle -462 trained in 2.76 seconds.\n",
      "Test circle -461 trained in 2.73 seconds.\n",
      "Test circle -460 trained in 2.84 seconds.\n",
      "Test circle -459 trained in 2.8 seconds.\n",
      "Test circle -458 trained in 2.74 seconds.\n",
      "Test circle -457 trained in 2.81 seconds.\n",
      "Test circle -456 trained in 2.79 seconds.\n",
      "Test circle -455 trained in 2.79 seconds.\n",
      "Test circle -454 trained in 2.89 seconds.\n",
      "Test circle -453 trained in 2.79 seconds.\n",
      "Test circle -452 trained in 2.73 seconds.\n",
      "Test circle -451 trained in 2.82 seconds.\n",
      "Test circle -450 trained in 2.76 seconds.\n",
      "Test circle -449 trained in 2.7 seconds.\n",
      "Test circle -448 trained in 2.78 seconds.\n",
      "Test circle -447 trained in 2.79 seconds.\n",
      "Test circle -446 trained in 2.74 seconds.\n",
      "Test circle -445 trained in 2.81 seconds.\n",
      "Test circle -444 trained in 2.84 seconds.\n",
      "Test circle -443 trained in 2.83 seconds.\n",
      "Test circle -442 trained in 2.94 seconds.\n",
      "Test circle -441 trained in 2.97 seconds.\n",
      "Test circle -440 trained in 2.8 seconds.\n",
      "Test circle -439 trained in 2.95 seconds.\n",
      "Test circle -438 trained in 2.98 seconds.\n",
      "Test circle -437 trained in 2.92 seconds.\n",
      "Test circle -436 trained in 3.14 seconds.\n",
      "Test circle -435 trained in 2.92 seconds.\n",
      "Test circle -434 trained in 3.08 seconds.\n",
      "Test circle -433 trained in 3.19 seconds.\n",
      "Test circle -432 trained in 3.28 seconds.\n",
      "Test circle -431 trained in 3.08 seconds.\n",
      "Test circle -430 trained in 3.01 seconds.\n",
      "Test circle -429 trained in 2.95 seconds.\n",
      "Test circle -428 trained in 3.18 seconds.\n",
      "Test circle -427 trained in 3.07 seconds.\n",
      "Test circle -426 trained in 2.98 seconds.\n",
      "Test circle -425 trained in 3.23 seconds.\n",
      "Test circle -424 trained in 3.08 seconds.\n",
      "Test circle -423 trained in 3.17 seconds.\n",
      "Test circle -422 trained in 3.26 seconds.\n",
      "Test circle -421 trained in 3.16 seconds.\n",
      "Test circle -420 trained in 3.2 seconds.\n",
      "Test circle -419 trained in 3.27 seconds.\n",
      "Test circle -418 trained in 3.2 seconds.\n",
      "Test circle -417 trained in 3.29 seconds.\n",
      "Test circle -416 trained in 3.23 seconds.\n",
      "Test circle -415 trained in 3.54 seconds.\n",
      "Test circle -414 trained in 3.48 seconds.\n",
      "Test circle -413 trained in 3.39 seconds.\n",
      "Test circle -412 trained in 2.72 seconds.\n",
      "Test circle -411 trained in 2.61 seconds.\n",
      "Test circle -410 trained in 2.69 seconds.\n",
      "Test circle -409 trained in 2.67 seconds.\n",
      "Test circle -408 trained in 2.64 seconds.\n",
      "Test circle -407 trained in 2.71 seconds.\n",
      "Test circle -406 trained in 2.7 seconds.\n",
      "Test circle -405 trained in 2.73 seconds.\n",
      "Test circle -404 trained in 2.74 seconds.\n",
      "Test circle -403 trained in 2.74 seconds.\n",
      "Test circle -402 trained in 2.74 seconds.\n",
      "Test circle -401 trained in 2.88 seconds.\n",
      "Test circle -400 trained in 2.75 seconds.\n",
      "Test circle -399 trained in 2.73 seconds.\n",
      "Test circle -398 trained in 2.71 seconds.\n",
      "Test circle -397 trained in 2.77 seconds.\n",
      "Test circle -396 trained in 2.71 seconds.\n",
      "Test circle -395 trained in 2.73 seconds.\n",
      "Test circle -394 trained in 2.81 seconds.\n",
      "Test circle -393 trained in 2.71 seconds.\n",
      "Test circle -392 trained in 2.76 seconds.\n",
      "Test circle -391 trained in 2.8 seconds.\n",
      "Test circle -390 trained in 2.79 seconds.\n",
      "Test circle -389 trained in 2.76 seconds.\n",
      "Test circle -388 trained in 2.79 seconds.\n",
      "Test circle -387 trained in 2.73 seconds.\n",
      "Test circle -386 trained in 2.79 seconds.\n",
      "Test circle -385 trained in 2.8 seconds.\n",
      "Test circle -384 trained in 2.75 seconds.\n",
      "Test circle -383 trained in 2.83 seconds.\n",
      "Test circle -382 trained in 2.94 seconds.\n",
      "Test circle -381 trained in 2.77 seconds.\n",
      "Test circle -380 trained in 2.81 seconds.\n",
      "Test circle -379 trained in 3.0 seconds.\n",
      "Test circle -378 trained in 2.78 seconds.\n",
      "Test circle -377 trained in 2.85 seconds.\n",
      "Test circle -376 trained in 2.83 seconds.\n",
      "Test circle -375 trained in 2.77 seconds.\n",
      "Test circle -374 trained in 2.87 seconds.\n",
      "Test circle -373 trained in 2.85 seconds.\n",
      "Test circle -372 trained in 2.78 seconds.\n",
      "Test circle -371 trained in 2.83 seconds.\n",
      "Test circle -370 trained in 2.88 seconds.\n",
      "Test circle -369 trained in 2.85 seconds.\n",
      "Test circle -368 trained in 2.89 seconds.\n",
      "Test circle -367 trained in 2.89 seconds.\n",
      "Test circle -366 trained in 2.83 seconds.\n",
      "Test circle -365 trained in 3.03 seconds.\n",
      "Test circle -364 trained in 2.89 seconds.\n",
      "Test circle -363 trained in 2.94 seconds.\n",
      "Test circle -362 trained in 3.42 seconds.\n",
      "Test circle -361 trained in 3.15 seconds.\n",
      "Test circle -360 trained in 3.13 seconds.\n",
      "Test circle -359 trained in 3.13 seconds.\n",
      "Test circle -358 trained in 2.96 seconds.\n",
      "Test circle -357 trained in 3.17 seconds.\n",
      "Test circle -356 trained in 3.16 seconds.\n",
      "Test circle -355 trained in 3.01 seconds.\n",
      "Test circle -354 trained in 3.22 seconds.\n",
      "Test circle -353 trained in 3.2 seconds.\n",
      "Test circle -352 trained in 3.27 seconds.\n",
      "Test circle -351 trained in 3.27 seconds.\n",
      "Test circle -350 trained in 3.23 seconds.\n",
      "Test circle -349 trained in 3.32 seconds.\n",
      "Test circle -348 trained in 3.19 seconds.\n",
      "Test circle -347 trained in 3.18 seconds.\n",
      "Test circle -346 trained in 3.38 seconds.\n",
      "Test circle -345 trained in 3.33 seconds.\n",
      "Test circle -344 trained in 3.47 seconds.\n",
      "Test circle -343 trained in 3.43 seconds.\n",
      "Test circle -342 trained in 3.49 seconds.\n",
      "Test circle -341 trained in 3.48 seconds.\n",
      "Test circle -340 trained in 2.64 seconds.\n",
      "Test circle -339 trained in 2.73 seconds.\n",
      "Test circle -338 trained in 2.64 seconds.\n",
      "Test circle -337 trained in 2.66 seconds.\n",
      "Test circle -336 trained in 2.71 seconds.\n",
      "Test circle -335 trained in 2.72 seconds.\n",
      "Test circle -334 trained in 2.68 seconds.\n",
      "Test circle -333 trained in 2.78 seconds.\n",
      "Test circle -332 trained in 2.79 seconds.\n",
      "Test circle -331 trained in 2.85 seconds.\n",
      "Test circle -330 trained in 2.76 seconds.\n",
      "Test circle -329 trained in 2.73 seconds.\n",
      "Test circle -328 trained in 2.84 seconds.\n",
      "Test circle -327 trained in 2.79 seconds.\n",
      "Test circle -326 trained in 2.76 seconds.\n",
      "Test circle -325 trained in 2.73 seconds.\n",
      "Test circle -324 trained in 2.76 seconds.\n",
      "Test circle -323 trained in 2.8 seconds.\n",
      "Test circle -322 trained in 2.74 seconds.\n",
      "Test circle -321 trained in 2.77 seconds.\n",
      "Test circle -320 trained in 2.78 seconds.\n",
      "Test circle -319 trained in 2.74 seconds.\n",
      "Test circle -318 trained in 2.77 seconds.\n",
      "Test circle -317 trained in 2.79 seconds.\n",
      "Test circle -316 trained in 2.79 seconds.\n",
      "Test circle -315 trained in 2.83 seconds.\n",
      "Test circle -314 trained in 2.79 seconds.\n",
      "Test circle -313 trained in 2.72 seconds.\n",
      "Test circle -312 trained in 2.81 seconds.\n",
      "Test circle -311 trained in 2.8 seconds.\n",
      "Test circle -310 trained in 2.88 seconds.\n",
      "Test circle -309 trained in 2.78 seconds.\n",
      "Test circle -308 trained in 2.75 seconds.\n",
      "Test circle -307 trained in 2.77 seconds.\n",
      "Test circle -306 trained in 2.86 seconds.\n",
      "Test circle -305 trained in 2.78 seconds.\n",
      "Test circle -304 trained in 2.75 seconds.\n",
      "Test circle -303 trained in 2.85 seconds.\n",
      "Test circle -302 trained in 2.74 seconds.\n",
      "Test circle -301 trained in 2.71 seconds.\n",
      "Test circle -300 trained in 2.81 seconds.\n",
      "Test circle -299 trained in 2.82 seconds.\n",
      "Test circle -298 trained in 2.74 seconds.\n",
      "Test circle -297 trained in 2.85 seconds.\n",
      "Test circle -296 trained in 2.78 seconds.\n",
      "Test circle -295 trained in 2.9 seconds.\n",
      "Test circle -294 trained in 2.97 seconds.\n",
      "Test circle -293 trained in 2.85 seconds.\n",
      "Test circle -292 trained in 3.02 seconds.\n",
      "Test circle -291 trained in 3.09 seconds.\n",
      "Test circle -290 trained in 2.88 seconds.\n",
      "Test circle -289 trained in 3.03 seconds.\n",
      "Test circle -288 trained in 3.02 seconds.\n",
      "Test circle -287 trained in 2.88 seconds.\n",
      "Test circle -286 trained in 2.94 seconds.\n",
      "Test circle -285 trained in 2.96 seconds.\n",
      "Test circle -284 trained in 2.93 seconds.\n",
      "Test circle -283 trained in 3.0 seconds.\n",
      "Test circle -282 trained in 3.05 seconds.\n",
      "Test circle -281 trained in 3.01 seconds.\n",
      "Test circle -280 trained in 3.24 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -279 trained in 3.16 seconds.\n",
      "Test circle -278 trained in 3.28 seconds.\n",
      "Test circle -277 trained in 3.3 seconds.\n",
      "Test circle -276 trained in 3.09 seconds.\n",
      "Test circle -275 trained in 3.42 seconds.\n",
      "Test circle -274 trained in 3.3 seconds.\n",
      "Test circle -273 trained in 3.22 seconds.\n",
      "Test circle -272 trained in 3.45 seconds.\n",
      "Test circle -271 trained in 3.34 seconds.\n",
      "Test circle -270 trained in 3.3 seconds.\n",
      "Test circle -269 trained in 3.28 seconds.\n",
      "Test circle -268 trained in 3.38 seconds.\n",
      "Test circle -267 trained in 3.66 seconds.\n",
      "Test circle -266 trained in 3.46 seconds.\n",
      "Test circle -265 trained in 3.74 seconds.\n",
      "Test circle -264 trained in 3.92 seconds.\n",
      "Test circle -263 trained in 2.75 seconds.\n",
      "Test circle -262 trained in 2.72 seconds.\n",
      "Test circle -261 trained in 2.73 seconds.\n",
      "Test circle -260 trained in 2.72 seconds.\n",
      "Test circle -259 trained in 2.71 seconds.\n",
      "Test circle -258 trained in 2.63 seconds.\n",
      "Test circle -257 trained in 2.79 seconds.\n",
      "Test circle -256 trained in 2.72 seconds.\n",
      "Test circle -255 trained in 2.69 seconds.\n",
      "Test circle -254 trained in 2.72 seconds.\n",
      "Test circle -253 trained in 2.71 seconds.\n",
      "Test circle -252 trained in 2.68 seconds.\n",
      "Test circle -251 trained in 2.77 seconds.\n",
      "Test circle -250 trained in 2.82 seconds.\n",
      "Test circle -249 trained in 2.72 seconds.\n",
      "Test circle -248 trained in 2.77 seconds.\n",
      "Test circle -247 trained in 2.72 seconds.\n",
      "Test circle -246 trained in 2.69 seconds.\n",
      "Test circle -245 trained in 2.84 seconds.\n",
      "Test circle -244 trained in 2.85 seconds.\n",
      "Test circle -243 trained in 2.69 seconds.\n",
      "Test circle -242 trained in 2.8 seconds.\n",
      "Test circle -241 trained in 2.72 seconds.\n",
      "Test circle -240 trained in 2.66 seconds.\n",
      "Test circle -239 trained in 2.76 seconds.\n",
      "Test circle -238 trained in 2.83 seconds.\n",
      "Test circle -237 trained in 2.71 seconds.\n",
      "Test circle -236 trained in 2.82 seconds.\n",
      "Test circle -235 trained in 2.78 seconds.\n",
      "Test circle -234 trained in 2.73 seconds.\n",
      "Test circle -233 trained in 2.86 seconds.\n",
      "Test circle -232 trained in 2.82 seconds.\n",
      "Test circle -231 trained in 2.77 seconds.\n",
      "Test circle -230 trained in 2.87 seconds.\n",
      "Test circle -229 trained in 2.96 seconds.\n",
      "Test circle -228 trained in 2.81 seconds.\n",
      "Test circle -227 trained in 2.87 seconds.\n",
      "Test circle -226 trained in 2.9 seconds.\n",
      "Test circle -225 trained in 2.83 seconds.\n",
      "Test circle -224 trained in 3.07 seconds.\n",
      "Test circle -223 trained in 2.94 seconds.\n",
      "Test circle -222 trained in 2.92 seconds.\n",
      "Test circle -221 trained in 3.04 seconds.\n",
      "Test circle -220 trained in 3.08 seconds.\n",
      "Test circle -219 trained in 3.12 seconds.\n",
      "Test circle -218 trained in 3.18 seconds.\n",
      "Test circle -217 trained in 2.98 seconds.\n",
      "Test circle -216 trained in 3.04 seconds.\n",
      "Test circle -215 trained in 3.13 seconds.\n",
      "Test circle -214 trained in 3.09 seconds.\n",
      "Test circle -213 trained in 3.18 seconds.\n",
      "Test circle -212 trained in 3.05 seconds.\n",
      "Test circle -211 trained in 3.15 seconds.\n",
      "Test circle -210 trained in 3.28 seconds.\n",
      "Test circle -209 trained in 3.13 seconds.\n",
      "Test circle -208 trained in 2.82 seconds.\n",
      "Test circle -207 trained in 2.67 seconds.\n",
      "Test circle -206 trained in 2.64 seconds.\n",
      "Test circle -205 trained in 2.7 seconds.\n",
      "Test circle -204 trained in 2.72 seconds.\n",
      "Test circle -203 trained in 2.74 seconds.\n",
      "Test circle -202 trained in 2.72 seconds.\n",
      "Test circle -201 trained in 2.74 seconds.\n",
      "Test circle -200 trained in 2.67 seconds.\n",
      "Test circle -199 trained in 2.71 seconds.\n",
      "Test circle -198 trained in 2.79 seconds.\n",
      "Test circle -197 trained in 2.71 seconds.\n",
      "Test circle -196 trained in 2.81 seconds.\n",
      "Test circle -195 trained in 2.78 seconds.\n",
      "Test circle -194 trained in 2.74 seconds.\n",
      "Test circle -193 trained in 2.78 seconds.\n",
      "Test circle -192 trained in 2.8 seconds.\n",
      "Test circle -191 trained in 2.76 seconds.\n",
      "Test circle -190 trained in 2.79 seconds.\n",
      "Test circle -189 trained in 2.77 seconds.\n",
      "Test circle -188 trained in 2.72 seconds.\n",
      "Test circle -187 trained in 2.7 seconds.\n",
      "Test circle -186 trained in 2.79 seconds.\n",
      "Test circle -185 trained in 2.75 seconds.\n",
      "Test circle -184 trained in 2.76 seconds.\n",
      "Test circle -183 trained in 2.74 seconds.\n",
      "Test circle -182 trained in 2.72 seconds.\n",
      "Test circle -181 trained in 2.84 seconds.\n",
      "Test circle -180 trained in 2.79 seconds.\n",
      "Test circle -179 trained in 2.75 seconds.\n",
      "Test circle -178 trained in 2.89 seconds.\n",
      "Test circle -177 trained in 2.88 seconds.\n",
      "Test circle -176 trained in 2.88 seconds.\n",
      "Test circle -175 trained in 2.91 seconds.\n",
      "Test circle -174 trained in 2.85 seconds.\n",
      "Test circle -173 trained in 2.78 seconds.\n",
      "Test circle -172 trained in 2.89 seconds.\n",
      "Test circle -171 trained in 2.89 seconds.\n",
      "Test circle -170 trained in 2.91 seconds.\n",
      "Test circle -169 trained in 3.05 seconds.\n",
      "Test circle -168 trained in 2.75 seconds.\n",
      "Test circle -167 trained in 2.73 seconds.\n",
      "Test circle -166 trained in 3.05 seconds.\n",
      "Test circle -165 trained in 2.81 seconds.\n",
      "Test circle -164 trained in 2.74 seconds.\n",
      "Test circle -163 trained in 2.85 seconds.\n",
      "Test circle -162 trained in 2.76 seconds.\n",
      "Test circle -161 trained in 2.75 seconds.\n",
      "Test circle -160 trained in 2.83 seconds.\n",
      "Test circle -159 trained in 2.8 seconds.\n",
      "Test circle -158 trained in 2.8 seconds.\n",
      "Test circle -157 trained in 2.83 seconds.\n",
      "Test circle -156 trained in 2.7 seconds.\n",
      "Test circle -155 trained in 2.72 seconds.\n",
      "Test circle -154 trained in 2.84 seconds.\n",
      "Test circle -153 trained in 2.72 seconds.\n",
      "Test circle -152 trained in 2.82 seconds.\n",
      "Test circle -151 trained in 2.91 seconds.\n",
      "Test circle -150 trained in 2.77 seconds.\n",
      "Test circle -149 trained in 2.77 seconds.\n",
      "Test circle -148 trained in 2.94 seconds.\n",
      "Test circle -147 trained in 2.72 seconds.\n",
      "Test circle -146 trained in 2.79 seconds.\n",
      "Test circle -145 trained in 2.86 seconds.\n",
      "Test circle -144 trained in 2.75 seconds.\n",
      "Test circle -143 trained in 2.79 seconds.\n",
      "Test circle -142 trained in 2.84 seconds.\n",
      "Test circle -141 trained in 2.76 seconds.\n",
      "Test circle -140 trained in 2.84 seconds.\n",
      "Test circle -139 trained in 2.93 seconds.\n",
      "Test circle -138 trained in 2.74 seconds.\n",
      "Test circle -137 trained in 2.8 seconds.\n",
      "Test circle -136 trained in 2.83 seconds.\n",
      "Test circle -135 trained in 2.76 seconds.\n",
      "Test circle -134 trained in 2.83 seconds.\n",
      "Test circle -133 trained in 3.08 seconds.\n",
      "Test circle -132 trained in 2.94 seconds.\n",
      "Test circle -131 trained in 2.88 seconds.\n",
      "Test circle -130 trained in 3.04 seconds.\n",
      "Test circle -129 trained in 2.91 seconds.\n",
      "Test circle -128 trained in 3.06 seconds.\n",
      "Test circle -127 trained in 3.05 seconds.\n",
      "Test circle -126 trained in 2.92 seconds.\n",
      "Test circle -125 trained in 3.1 seconds.\n",
      "Test circle -124 trained in 2.95 seconds.\n",
      "Test circle -123 trained in 3.35 seconds.\n",
      "Test circle -122 trained in 3.11 seconds.\n",
      "Test circle -121 trained in 3.04 seconds.\n",
      "Test circle -120 trained in 3.17 seconds.\n",
      "Test circle -119 trained in 3.1 seconds.\n",
      "Test circle -118 trained in 3.06 seconds.\n",
      "Test circle -117 trained in 3.2 seconds.\n",
      "Test circle -116 trained in 3.16 seconds.\n",
      "Test circle -115 trained in 3.17 seconds.\n",
      "Test circle -114 trained in 3.3 seconds.\n",
      "Test circle -113 trained in 3.27 seconds.\n",
      "Test circle -112 trained in 3.31 seconds.\n",
      "Test circle -111 trained in 3.35 seconds.\n",
      "Test circle -110 trained in 3.33 seconds.\n",
      "Test circle -109 trained in 3.37 seconds.\n",
      "Test circle -108 trained in 3.39 seconds.\n",
      "Test circle -107 trained in 3.52 seconds.\n",
      "Test circle -106 trained in 3.64 seconds.\n",
      "Test circle -105 trained in 3.54 seconds.\n",
      "Test circle -104 trained in 2.82 seconds.\n",
      "Test circle -103 trained in 3.0 seconds.\n",
      "Test circle -102 trained in 3.1 seconds.\n",
      "Test circle -101 trained in 2.83 seconds.\n",
      "Test circle -100 trained in 2.68 seconds.\n",
      "Test circle -99 trained in 2.78 seconds.\n",
      "Test circle -98 trained in 2.68 seconds.\n",
      "Test circle -97 trained in 2.74 seconds.\n",
      "Test circle -96 trained in 2.75 seconds.\n",
      "Test circle -95 trained in 2.75 seconds.\n",
      "Test circle -94 trained in 2.86 seconds.\n",
      "Test circle -93 trained in 2.84 seconds.\n",
      "Test circle -92 trained in 2.73 seconds.\n",
      "Test circle -91 trained in 2.75 seconds.\n",
      "Test circle -90 trained in 2.84 seconds.\n",
      "Test circle -89 trained in 2.75 seconds.\n",
      "Test circle -88 trained in 2.73 seconds.\n",
      "Test circle -87 trained in 2.79 seconds.\n",
      "Test circle -86 trained in 2.76 seconds.\n",
      "Test circle -85 trained in 2.77 seconds.\n",
      "Test circle -84 trained in 2.83 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -83 trained in 2.73 seconds.\n",
      "Test circle -82 trained in 2.76 seconds.\n",
      "Test circle -81 trained in 2.82 seconds.\n",
      "Test circle -80 trained in 2.71 seconds.\n",
      "Test circle -79 trained in 2.78 seconds.\n",
      "Test circle -78 trained in 2.89 seconds.\n",
      "Test circle -77 trained in 2.88 seconds.\n",
      "Test circle -76 trained in 2.73 seconds.\n",
      "Test circle -75 trained in 2.93 seconds.\n",
      "Test circle -74 trained in 2.71 seconds.\n",
      "Test circle -73 trained in 2.76 seconds.\n",
      "Test circle -72 trained in 3.03 seconds.\n",
      "Test circle -71 trained in 2.97 seconds.\n",
      "Test circle -70 trained in 2.88 seconds.\n",
      "Test circle -69 trained in 2.82 seconds.\n",
      "Test circle -68 trained in 2.8 seconds.\n",
      "Test circle -67 trained in 2.88 seconds.\n",
      "Test circle -66 trained in 2.82 seconds.\n",
      "Test circle -65 trained in 2.74 seconds.\n",
      "Test circle -64 trained in 2.89 seconds.\n",
      "Test circle -63 trained in 2.85 seconds.\n",
      "Test circle -62 trained in 2.83 seconds.\n",
      "Test circle -61 trained in 2.91 seconds.\n",
      "Test circle -60 trained in 2.81 seconds.\n",
      "Test circle -59 trained in 2.84 seconds.\n",
      "Test circle -58 trained in 2.9 seconds.\n",
      "Test circle -57 trained in 2.85 seconds.\n",
      "Test circle -56 trained in 2.85 seconds.\n",
      "Test circle -55 trained in 2.92 seconds.\n",
      "Test circle -54 trained in 2.94 seconds.\n",
      "Test circle -53 trained in 2.96 seconds.\n",
      "Test circle -52 trained in 3.07 seconds.\n",
      "Test circle -51 trained in 2.98 seconds.\n",
      "Test circle -50 trained in 2.95 seconds.\n",
      "Test circle -49 trained in 3.08 seconds.\n",
      "Test circle -48 trained in 3.0 seconds.\n",
      "Test circle -47 trained in 3.07 seconds.\n",
      "Test circle -46 trained in 3.08 seconds.\n",
      "Test circle -45 trained in 3.05 seconds.\n",
      "Test circle -44 trained in 3.14 seconds.\n",
      "Test circle -43 trained in 3.17 seconds.\n",
      "Test circle -42 trained in 3.13 seconds.\n",
      "Test circle -41 trained in 3.86 seconds.\n",
      "Test circle -40 trained in 3.63 seconds.\n",
      "Test circle -39 trained in 3.51 seconds.\n",
      "Test circle -38 trained in 3.61 seconds.\n",
      "Test circle -37 trained in 3.31 seconds.\n",
      "Test circle -36 trained in 3.24 seconds.\n",
      "Test circle -35 trained in 3.31 seconds.\n",
      "Test circle -34 trained in 3.43 seconds.\n",
      "Test circle -33 trained in 3.51 seconds.\n",
      "Test circle -32 trained in 3.46 seconds.\n",
      "Test circle -31 trained in 3.37 seconds.\n",
      "Test circle -30 trained in 3.39 seconds.\n",
      "Test circle -29 trained in 3.65 seconds.\n",
      "Test circle -28 trained in 3.57 seconds.\n",
      "Test circle -27 trained in 3.68 seconds.\n",
      "Test circle -26 trained in 3.7 seconds.\n",
      "Test circle -25 trained in 3.61 seconds.\n",
      "Test circle -24 trained in 2.63 seconds.\n",
      "Test circle -23 trained in 2.76 seconds.\n",
      "Test circle -22 trained in 2.7 seconds.\n",
      "Test circle -21 trained in 2.67 seconds.\n",
      "Test circle -20 trained in 2.64 seconds.\n",
      "Test circle -19 trained in 2.7 seconds.\n",
      "Test circle -18 trained in 2.69 seconds.\n",
      "Test circle -17 trained in 2.66 seconds.\n",
      "Test circle -16 trained in 2.73 seconds.\n",
      "Test circle -15 trained in 2.72 seconds.\n",
      "Test circle -14 trained in 2.76 seconds.\n",
      "Test circle -13 trained in 2.78 seconds.\n",
      "Test circle -12 trained in 2.76 seconds.\n",
      "Test circle -11 trained in 2.67 seconds.\n",
      "Test circle -10 trained in 2.77 seconds.\n",
      "Test circle -9 trained in 2.79 seconds.\n",
      "Test circle -8 trained in 2.7 seconds.\n",
      "Test circle -7 trained in 2.78 seconds.\n",
      "Test circle -6 trained in 2.75 seconds.\n",
      "Test circle -5 trained in 2.89 seconds.\n",
      "Test circle -4 trained in 2.8 seconds.\n",
      "Test circle -3 trained in 2.75 seconds.\n",
      "Test circle -2 trained in 2.74 seconds.\n",
      "Test circle -1 trained in 2.8 seconds.\n",
      "Model trained in 2134.66 seconds.\n",
      "Hyperparams (50, 3, 6) trained in 2134.66 seconds.\n",
      "=========================================================================================\n",
      "Model (50, 3, 6) Ended\n",
      "=========================================================================================\n",
      "=========================================================================================\n",
      "Model (50, 6, 3) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 2.77 seconds.\n",
      "Test circle -724 trained in 2.71 seconds.\n",
      "Test circle -723 trained in 2.77 seconds.\n",
      "Test circle -722 trained in 2.77 seconds.\n",
      "Test circle -721 trained in 2.77 seconds.\n",
      "Test circle -720 trained in 2.85 seconds.\n",
      "Test circle -719 trained in 2.74 seconds.\n",
      "Test circle -718 trained in 2.74 seconds.\n",
      "Test circle -717 trained in 2.8 seconds.\n",
      "Test circle -716 trained in 2.8 seconds.\n",
      "Test circle -715 trained in 2.7 seconds.\n",
      "Test circle -714 trained in 2.81 seconds.\n",
      "Test circle -713 trained in 2.77 seconds.\n",
      "Test circle -712 trained in 2.75 seconds.\n",
      "Test circle -711 trained in 2.88 seconds.\n",
      "Test circle -710 trained in 2.79 seconds.\n",
      "Test circle -709 trained in 2.72 seconds.\n",
      "Test circle -708 trained in 2.8 seconds.\n",
      "Test circle -707 trained in 2.77 seconds.\n",
      "Test circle -706 trained in 2.72 seconds.\n",
      "Test circle -705 trained in 2.82 seconds.\n",
      "Test circle -704 trained in 3.05 seconds.\n",
      "Test circle -703 trained in 3.06 seconds.\n",
      "Test circle -702 trained in 2.76 seconds.\n",
      "Test circle -701 trained in 2.87 seconds.\n",
      "Test circle -700 trained in 2.96 seconds.\n",
      "Test circle -699 trained in 3.07 seconds.\n",
      "Test circle -698 trained in 3.14 seconds.\n",
      "Test circle -697 trained in 2.99 seconds.\n",
      "Test circle -696 trained in 3.14 seconds.\n",
      "Test circle -695 trained in 3.42 seconds.\n",
      "Test circle -694 trained in 4.02 seconds.\n",
      "Test circle -693 trained in 3.25 seconds.\n",
      "Test circle -692 trained in 2.93 seconds.\n",
      "Test circle -691 trained in 3.47 seconds.\n",
      "Test circle -690 trained in 2.95 seconds.\n",
      "Test circle -689 trained in 2.98 seconds.\n",
      "Test circle -688 trained in 3.05 seconds.\n",
      "Test circle -687 trained in 3.06 seconds.\n",
      "Test circle -686 trained in 3.18 seconds.\n",
      "Test circle -685 trained in 3.27 seconds.\n",
      "Test circle -684 trained in 3.22 seconds.\n",
      "Test circle -683 trained in 3.28 seconds.\n",
      "Test circle -682 trained in 3.09 seconds.\n",
      "Test circle -681 trained in 3.12 seconds.\n",
      "Test circle -680 trained in 3.18 seconds.\n",
      "Test circle -679 trained in 3.16 seconds.\n",
      "Test circle -678 trained in 3.26 seconds.\n",
      "Test circle -677 trained in 3.2 seconds.\n",
      "Test circle -676 trained in 3.16 seconds.\n",
      "Test circle -675 trained in 3.33 seconds.\n",
      "Test circle -674 trained in 3.17 seconds.\n",
      "Test circle -673 trained in 3.3 seconds.\n",
      "Test circle -672 trained in 3.29 seconds.\n",
      "Test circle -671 trained in 3.28 seconds.\n",
      "Test circle -670 trained in 3.61 seconds.\n",
      "Test circle -669 trained in 3.87 seconds.\n",
      "Test circle -668 trained in 4.33 seconds.\n",
      "Test circle -667 trained in 4.24 seconds.\n",
      "Test circle -666 trained in 4.39 seconds.\n",
      "Test circle -665 trained in 4.33 seconds.\n",
      "Test circle -664 trained in 2.87 seconds.\n",
      "Test circle -663 trained in 2.91 seconds.\n",
      "Test circle -662 trained in 2.78 seconds.\n",
      "Test circle -661 trained in 3.04 seconds.\n",
      "Test circle -660 trained in 2.81 seconds.\n",
      "Test circle -659 trained in 2.71 seconds.\n",
      "Test circle -658 trained in 2.67 seconds.\n",
      "Test circle -657 trained in 2.87 seconds.\n",
      "Test circle -656 trained in 3.24 seconds.\n",
      "Test circle -655 trained in 2.92 seconds.\n",
      "Test circle -654 trained in 2.84 seconds.\n",
      "Test circle -653 trained in 2.82 seconds.\n",
      "Test circle -652 trained in 2.81 seconds.\n",
      "Test circle -651 trained in 2.69 seconds.\n",
      "Test circle -650 trained in 2.79 seconds.\n",
      "Test circle -649 trained in 2.88 seconds.\n",
      "Test circle -648 trained in 2.73 seconds.\n",
      "Test circle -647 trained in 2.8 seconds.\n",
      "Test circle -646 trained in 2.74 seconds.\n",
      "Test circle -645 trained in 2.72 seconds.\n",
      "Test circle -644 trained in 2.75 seconds.\n",
      "Test circle -643 trained in 2.74 seconds.\n",
      "Test circle -642 trained in 2.76 seconds.\n",
      "Test circle -641 trained in 2.79 seconds.\n",
      "Test circle -640 trained in 2.78 seconds.\n",
      "Test circle -639 trained in 2.75 seconds.\n",
      "Test circle -638 trained in 2.9 seconds.\n",
      "Test circle -637 trained in 2.73 seconds.\n",
      "Test circle -636 trained in 2.7 seconds.\n",
      "Test circle -635 trained in 2.78 seconds.\n",
      "Test circle -634 trained in 2.88 seconds.\n",
      "Test circle -633 trained in 2.74 seconds.\n",
      "Test circle -632 trained in 2.76 seconds.\n",
      "Test circle -631 trained in 2.74 seconds.\n",
      "Test circle -630 trained in 2.74 seconds.\n",
      "Test circle -629 trained in 2.79 seconds.\n",
      "Test circle -628 trained in 2.73 seconds.\n",
      "Test circle -627 trained in 2.71 seconds.\n",
      "Test circle -626 trained in 2.85 seconds.\n",
      "Test circle -625 trained in 2.82 seconds.\n",
      "Test circle -624 trained in 2.72 seconds.\n",
      "Test circle -623 trained in 2.75 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -622 trained in 2.77 seconds.\n",
      "Test circle -621 trained in 2.7 seconds.\n",
      "Test circle -620 trained in 2.79 seconds.\n",
      "Test circle -619 trained in 2.95 seconds.\n",
      "Test circle -618 trained in 2.76 seconds.\n",
      "Test circle -617 trained in 2.86 seconds.\n",
      "Test circle -616 trained in 2.77 seconds.\n",
      "Test circle -615 trained in 2.87 seconds.\n",
      "Test circle -614 trained in 3.05 seconds.\n",
      "Test circle -613 trained in 2.74 seconds.\n",
      "Test circle -612 trained in 2.89 seconds.\n",
      "Test circle -611 trained in 2.9 seconds.\n",
      "Test circle -610 trained in 2.96 seconds.\n",
      "Test circle -609 trained in 2.91 seconds.\n",
      "Test circle -608 trained in 2.87 seconds.\n",
      "Test circle -607 trained in 2.78 seconds.\n",
      "Test circle -606 trained in 2.87 seconds.\n",
      "Test circle -605 trained in 2.92 seconds.\n",
      "Test circle -604 trained in 2.89 seconds.\n",
      "Test circle -603 trained in 3.09 seconds.\n",
      "Test circle -602 trained in 3.02 seconds.\n",
      "Test circle -601 trained in 2.89 seconds.\n",
      "Test circle -600 trained in 3.04 seconds.\n",
      "Test circle -599 trained in 2.96 seconds.\n",
      "Test circle -598 trained in 2.96 seconds.\n",
      "Test circle -597 trained in 3.1 seconds.\n",
      "Test circle -596 trained in 3.05 seconds.\n",
      "Test circle -595 trained in 3.0 seconds.\n",
      "Test circle -594 trained in 3.23 seconds.\n",
      "Test circle -593 trained in 3.22 seconds.\n",
      "Test circle -592 trained in 3.38 seconds.\n",
      "Test circle -591 trained in 3.25 seconds.\n",
      "Test circle -590 trained in 3.13 seconds.\n",
      "Test circle -589 trained in 3.38 seconds.\n",
      "Test circle -588 trained in 3.26 seconds.\n",
      "Test circle -587 trained in 3.17 seconds.\n",
      "Test circle -586 trained in 3.18 seconds.\n",
      "Test circle -585 trained in 3.14 seconds.\n",
      "Test circle -584 trained in 3.23 seconds.\n",
      "Test circle -583 trained in 3.21 seconds.\n",
      "Test circle -582 trained in 3.18 seconds.\n",
      "Test circle -581 trained in 3.46 seconds.\n",
      "Test circle -580 trained in 3.33 seconds.\n",
      "Test circle -579 trained in 3.51 seconds.\n",
      "Test circle -578 trained in 3.42 seconds.\n",
      "Test circle -577 trained in 3.39 seconds.\n",
      "Test circle -576 trained in 2.99 seconds.\n",
      "Test circle -575 trained in 2.59 seconds.\n",
      "Test circle -574 trained in 2.64 seconds.\n",
      "Test circle -573 trained in 2.66 seconds.\n",
      "Test circle -572 trained in 2.65 seconds.\n",
      "Test circle -571 trained in 2.74 seconds.\n",
      "Test circle -570 trained in 2.69 seconds.\n",
      "Test circle -569 trained in 2.64 seconds.\n",
      "Test circle -568 trained in 2.68 seconds.\n",
      "Test circle -567 trained in 2.71 seconds.\n",
      "Test circle -566 trained in 2.66 seconds.\n",
      "Test circle -565 trained in 2.69 seconds.\n",
      "Test circle -564 trained in 2.71 seconds.\n",
      "Test circle -563 trained in 2.76 seconds.\n",
      "Test circle -562 trained in 2.69 seconds.\n",
      "Test circle -561 trained in 2.75 seconds.\n",
      "Test circle -560 trained in 2.68 seconds.\n",
      "Test circle -559 trained in 2.75 seconds.\n",
      "Test circle -558 trained in 2.76 seconds.\n",
      "Test circle -557 trained in 2.71 seconds.\n",
      "Test circle -556 trained in 2.72 seconds.\n",
      "Test circle -555 trained in 2.78 seconds.\n",
      "Test circle -554 trained in 2.69 seconds.\n",
      "Test circle -553 trained in 2.72 seconds.\n",
      "Test circle -552 trained in 2.77 seconds.\n",
      "Test circle -551 trained in 2.69 seconds.\n",
      "Test circle -550 trained in 2.69 seconds.\n",
      "Test circle -549 trained in 2.8 seconds.\n",
      "Test circle -548 trained in 2.72 seconds.\n",
      "Test circle -547 trained in 2.7 seconds.\n",
      "Test circle -546 trained in 2.8 seconds.\n",
      "Test circle -545 trained in 2.69 seconds.\n",
      "Test circle -544 trained in 2.8 seconds.\n",
      "Test circle -543 trained in 2.83 seconds.\n",
      "Test circle -542 trained in 2.71 seconds.\n",
      "Test circle -541 trained in 2.7 seconds.\n",
      "Test circle -540 trained in 2.82 seconds.\n",
      "Test circle -539 trained in 2.8 seconds.\n",
      "Test circle -538 trained in 2.88 seconds.\n",
      "Test circle -537 trained in 2.8 seconds.\n",
      "Test circle -536 trained in 2.81 seconds.\n",
      "Test circle -535 trained in 2.73 seconds.\n",
      "Test circle -534 trained in 2.83 seconds.\n",
      "Test circle -533 trained in 2.73 seconds.\n",
      "Test circle -532 trained in 2.85 seconds.\n",
      "Test circle -531 trained in 3.58 seconds.\n",
      "Test circle -530 trained in 2.99 seconds.\n",
      "Test circle -529 trained in 2.97 seconds.\n",
      "Test circle -528 trained in 2.89 seconds.\n",
      "Test circle -527 trained in 3.01 seconds.\n",
      "Test circle -526 trained in 3.09 seconds.\n",
      "Test circle -525 trained in 3.02 seconds.\n",
      "Test circle -524 trained in 3.05 seconds.\n",
      "Test circle -523 trained in 3.19 seconds.\n",
      "Test circle -522 trained in 3.02 seconds.\n",
      "Test circle -521 trained in 3.34 seconds.\n",
      "Test circle -520 trained in 3.09 seconds.\n",
      "Test circle -519 trained in 3.08 seconds.\n",
      "Test circle -518 trained in 3.04 seconds.\n",
      "Test circle -517 trained in 3.17 seconds.\n",
      "Test circle -516 trained in 3.03 seconds.\n",
      "Test circle -515 trained in 3.34 seconds.\n",
      "Test circle -514 trained in 3.2 seconds.\n",
      "Test circle -513 trained in 3.21 seconds.\n",
      "Test circle -512 trained in 3.41 seconds.\n",
      "Test circle -511 trained in 3.14 seconds.\n",
      "Test circle -510 trained in 3.31 seconds.\n",
      "Test circle -509 trained in 3.26 seconds.\n",
      "Test circle -508 trained in 3.33 seconds.\n",
      "Test circle -507 trained in 3.53 seconds.\n",
      "Test circle -506 trained in 3.57 seconds.\n",
      "Test circle -505 trained in 2.82 seconds.\n",
      "Test circle -504 trained in 2.64 seconds.\n",
      "Test circle -503 trained in 2.67 seconds.\n",
      "Test circle -502 trained in 2.79 seconds.\n",
      "Test circle -501 trained in 2.72 seconds.\n",
      "Test circle -500 trained in 2.72 seconds.\n",
      "Test circle -499 trained in 2.76 seconds.\n",
      "Test circle -498 trained in 2.71 seconds.\n",
      "Test circle -497 trained in 2.82 seconds.\n",
      "Test circle -496 trained in 2.9 seconds.\n",
      "Test circle -495 trained in 2.71 seconds.\n",
      "Test circle -494 trained in 2.72 seconds.\n",
      "Test circle -493 trained in 2.76 seconds.\n",
      "Test circle -492 trained in 2.71 seconds.\n",
      "Test circle -491 trained in 2.71 seconds.\n",
      "Test circle -490 trained in 2.8 seconds.\n",
      "Test circle -489 trained in 2.86 seconds.\n",
      "Test circle -488 trained in 2.73 seconds.\n",
      "Test circle -487 trained in 2.93 seconds.\n",
      "Test circle -486 trained in 2.79 seconds.\n",
      "Test circle -485 trained in 2.77 seconds.\n",
      "Test circle -484 trained in 3.34 seconds.\n",
      "Test circle -483 trained in 2.8 seconds.\n",
      "Test circle -482 trained in 2.8 seconds.\n",
      "Test circle -481 trained in 2.79 seconds.\n",
      "Test circle -480 trained in 2.71 seconds.\n",
      "Test circle -479 trained in 2.78 seconds.\n",
      "Test circle -478 trained in 2.76 seconds.\n",
      "Test circle -477 trained in 2.71 seconds.\n",
      "Test circle -476 trained in 2.79 seconds.\n",
      "Test circle -475 trained in 2.75 seconds.\n",
      "Test circle -474 trained in 2.71 seconds.\n",
      "Test circle -473 trained in 2.75 seconds.\n",
      "Test circle -472 trained in 2.75 seconds.\n",
      "Test circle -471 trained in 2.68 seconds.\n",
      "Test circle -470 trained in 2.85 seconds.\n",
      "Test circle -469 trained in 2.82 seconds.\n",
      "Test circle -468 trained in 2.73 seconds.\n",
      "Test circle -467 trained in 2.75 seconds.\n",
      "Test circle -466 trained in 2.83 seconds.\n",
      "Test circle -465 trained in 2.74 seconds.\n",
      "Test circle -464 trained in 2.75 seconds.\n",
      "Test circle -463 trained in 2.81 seconds.\n",
      "Test circle -462 trained in 2.72 seconds.\n",
      "Test circle -461 trained in 2.77 seconds.\n",
      "Test circle -460 trained in 2.91 seconds.\n",
      "Test circle -459 trained in 2.78 seconds.\n",
      "Test circle -458 trained in 2.89 seconds.\n",
      "Test circle -457 trained in 2.93 seconds.\n",
      "Test circle -456 trained in 2.75 seconds.\n",
      "Test circle -455 trained in 2.81 seconds.\n",
      "Test circle -454 trained in 2.84 seconds.\n",
      "Test circle -453 trained in 2.86 seconds.\n",
      "Test circle -452 trained in 3.05 seconds.\n",
      "Test circle -451 trained in 2.94 seconds.\n",
      "Test circle -450 trained in 2.82 seconds.\n",
      "Test circle -449 trained in 2.9 seconds.\n",
      "Test circle -448 trained in 2.92 seconds.\n",
      "Test circle -447 trained in 2.87 seconds.\n",
      "Test circle -446 trained in 2.96 seconds.\n",
      "Test circle -445 trained in 2.93 seconds.\n",
      "Test circle -444 trained in 2.94 seconds.\n",
      "Test circle -443 trained in 3.02 seconds.\n",
      "Test circle -442 trained in 3.16 seconds.\n",
      "Test circle -441 trained in 3.03 seconds.\n",
      "Test circle -440 trained in 3.05 seconds.\n",
      "Test circle -439 trained in 2.98 seconds.\n",
      "Test circle -438 trained in 2.85 seconds.\n",
      "Test circle -437 trained in 2.68 seconds.\n",
      "Test circle -436 trained in 2.67 seconds.\n",
      "Test circle -435 trained in 2.74 seconds.\n",
      "Test circle -434 trained in 2.84 seconds.\n",
      "Test circle -433 trained in 2.72 seconds.\n",
      "Test circle -432 trained in 2.83 seconds.\n",
      "Test circle -431 trained in 2.78 seconds.\n",
      "Test circle -430 trained in 2.75 seconds.\n",
      "Test circle -429 trained in 2.78 seconds.\n",
      "Test circle -428 trained in 3.04 seconds.\n",
      "Test circle -427 trained in 2.71 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -426 trained in 2.81 seconds.\n",
      "Test circle -425 trained in 2.73 seconds.\n",
      "Test circle -424 trained in 2.71 seconds.\n",
      "Test circle -423 trained in 2.81 seconds.\n",
      "Test circle -422 trained in 2.75 seconds.\n",
      "Test circle -421 trained in 2.73 seconds.\n",
      "Test circle -420 trained in 2.77 seconds.\n",
      "Test circle -419 trained in 2.79 seconds.\n",
      "Test circle -418 trained in 3.01 seconds.\n",
      "Test circle -417 trained in 2.89 seconds.\n",
      "Test circle -416 trained in 2.76 seconds.\n",
      "Test circle -415 trained in 2.8 seconds.\n",
      "Test circle -414 trained in 2.74 seconds.\n",
      "Test circle -413 trained in 2.77 seconds.\n",
      "Test circle -412 trained in 2.71 seconds.\n",
      "Test circle -411 trained in 2.86 seconds.\n",
      "Test circle -410 trained in 2.76 seconds.\n",
      "Test circle -409 trained in 2.71 seconds.\n",
      "Test circle -408 trained in 2.76 seconds.\n",
      "Test circle -407 trained in 2.74 seconds.\n",
      "Test circle -406 trained in 2.74 seconds.\n",
      "Test circle -405 trained in 2.74 seconds.\n",
      "Test circle -404 trained in 2.77 seconds.\n",
      "Test circle -403 trained in 2.77 seconds.\n",
      "Test circle -402 trained in 2.85 seconds.\n",
      "Test circle -401 trained in 2.84 seconds.\n",
      "Test circle -400 trained in 2.87 seconds.\n",
      "Test circle -399 trained in 3.03 seconds.\n",
      "Test circle -398 trained in 2.93 seconds.\n",
      "Test circle -397 trained in 2.94 seconds.\n",
      "Test circle -396 trained in 2.75 seconds.\n",
      "Test circle -395 trained in 2.79 seconds.\n",
      "Test circle -394 trained in 2.76 seconds.\n",
      "Test circle -393 trained in 2.77 seconds.\n",
      "Test circle -392 trained in 2.79 seconds.\n",
      "Test circle -391 trained in 2.85 seconds.\n",
      "Test circle -390 trained in 2.8 seconds.\n",
      "Test circle -389 trained in 2.87 seconds.\n",
      "Test circle -388 trained in 2.92 seconds.\n",
      "Test circle -387 trained in 2.95 seconds.\n",
      "Test circle -386 trained in 3.03 seconds.\n",
      "Test circle -385 trained in 2.9 seconds.\n",
      "Test circle -384 trained in 2.97 seconds.\n",
      "Test circle -383 trained in 3.53 seconds.\n",
      "Test circle -382 trained in 3.3 seconds.\n",
      "Test circle -381 trained in 2.99 seconds.\n",
      "Test circle -380 trained in 2.84 seconds.\n",
      "Test circle -379 trained in 2.9 seconds.\n",
      "Test circle -378 trained in 2.98 seconds.\n",
      "Test circle -377 trained in 3.03 seconds.\n",
      "Test circle -376 trained in 3.38 seconds.\n",
      "Test circle -375 trained in 2.93 seconds.\n",
      "Test circle -374 trained in 3.51 seconds.\n",
      "Test circle -373 trained in 3.46 seconds.\n",
      "Test circle -372 trained in 3.07 seconds.\n",
      "Test circle -371 trained in 2.9 seconds.\n",
      "Test circle -370 trained in 2.89 seconds.\n",
      "Test circle -369 trained in 2.86 seconds.\n",
      "Test circle -368 trained in 2.85 seconds.\n",
      "Test circle -367 trained in 2.85 seconds.\n",
      "Test circle -366 trained in 2.79 seconds.\n",
      "Test circle -365 trained in 2.84 seconds.\n",
      "Test circle -364 trained in 2.87 seconds.\n",
      "Test circle -363 trained in 2.93 seconds.\n",
      "Test circle -362 trained in 2.87 seconds.\n",
      "Test circle -361 trained in 2.94 seconds.\n",
      "Test circle -360 trained in 2.93 seconds.\n",
      "Test circle -359 trained in 2.82 seconds.\n",
      "Test circle -358 trained in 2.81 seconds.\n",
      "Test circle -357 trained in 2.77 seconds.\n",
      "Test circle -356 trained in 2.84 seconds.\n",
      "Test circle -355 trained in 3.02 seconds.\n",
      "Test circle -354 trained in 3.78 seconds.\n",
      "Test circle -353 trained in 3.75 seconds.\n",
      "Test circle -352 trained in 3.76 seconds.\n",
      "Test circle -351 trained in 3.74 seconds.\n",
      "Test circle -350 trained in 2.91 seconds.\n",
      "Test circle -349 trained in 2.82 seconds.\n",
      "Test circle -348 trained in 2.77 seconds.\n",
      "Test circle -347 trained in 2.74 seconds.\n",
      "Test circle -346 trained in 2.75 seconds.\n",
      "Test circle -345 trained in 2.75 seconds.\n",
      "Test circle -344 trained in 2.76 seconds.\n",
      "Test circle -343 trained in 2.91 seconds.\n",
      "Test circle -342 trained in 2.77 seconds.\n",
      "Test circle -341 trained in 2.82 seconds.\n",
      "Test circle -340 trained in 2.78 seconds.\n",
      "Test circle -339 trained in 2.77 seconds.\n",
      "Test circle -338 trained in 2.77 seconds.\n",
      "Test circle -337 trained in 2.76 seconds.\n",
      "Test circle -336 trained in 2.79 seconds.\n",
      "Test circle -335 trained in 2.82 seconds.\n",
      "Test circle -334 trained in 2.8 seconds.\n",
      "Test circle -333 trained in 2.76 seconds.\n",
      "Test circle -332 trained in 2.75 seconds.\n",
      "Test circle -331 trained in 2.78 seconds.\n",
      "Test circle -330 trained in 2.79 seconds.\n",
      "Test circle -329 trained in 2.8 seconds.\n",
      "Test circle -328 trained in 2.79 seconds.\n",
      "Test circle -327 trained in 2.77 seconds.\n",
      "Test circle -326 trained in 2.84 seconds.\n",
      "Test circle -325 trained in 2.81 seconds.\n",
      "Test circle -324 trained in 2.88 seconds.\n",
      "Test circle -323 trained in 2.75 seconds.\n",
      "Test circle -322 trained in 2.76 seconds.\n",
      "Test circle -321 trained in 2.8 seconds.\n",
      "Test circle -320 trained in 2.76 seconds.\n",
      "Test circle -319 trained in 2.85 seconds.\n",
      "Test circle -318 trained in 2.75 seconds.\n",
      "Test circle -317 trained in 2.76 seconds.\n",
      "Test circle -316 trained in 2.75 seconds.\n",
      "Test circle -315 trained in 2.74 seconds.\n",
      "Test circle -314 trained in 2.75 seconds.\n",
      "Test circle -313 trained in 2.76 seconds.\n",
      "Test circle -312 trained in 2.75 seconds.\n",
      "Test circle -311 trained in 2.75 seconds.\n",
      "Test circle -310 trained in 2.73 seconds.\n",
      "Test circle -309 trained in 2.75 seconds.\n",
      "Test circle -308 trained in 2.75 seconds.\n",
      "Test circle -307 trained in 2.78 seconds.\n",
      "Test circle -306 trained in 2.75 seconds.\n",
      "Test circle -305 trained in 2.86 seconds.\n",
      "Test circle -304 trained in 2.75 seconds.\n",
      "Test circle -303 trained in 2.76 seconds.\n",
      "Test circle -302 trained in 2.75 seconds.\n",
      "Test circle -301 trained in 2.75 seconds.\n",
      "Test circle -300 trained in 2.77 seconds.\n",
      "Test circle -299 trained in 2.74 seconds.\n",
      "Test circle -298 trained in 2.77 seconds.\n",
      "Test circle -297 trained in 2.76 seconds.\n",
      "Test circle -296 trained in 2.75 seconds.\n",
      "Test circle -295 trained in 2.76 seconds.\n",
      "Test circle -294 trained in 2.74 seconds.\n",
      "Test circle -293 trained in 2.74 seconds.\n",
      "Test circle -292 trained in 2.74 seconds.\n",
      "Test circle -291 trained in 2.74 seconds.\n",
      "Test circle -290 trained in 2.74 seconds.\n",
      "Test circle -289 trained in 2.93 seconds.\n",
      "Test circle -288 trained in 3.43 seconds.\n",
      "Test circle -287 trained in 3.35 seconds.\n",
      "Test circle -286 trained in 3.86 seconds.\n",
      "Test circle -285 trained in 4.75 seconds.\n",
      "Test circle -284 trained in 3.5 seconds.\n",
      "Test circle -283 trained in 3.38 seconds.\n",
      "Test circle -282 trained in 2.95 seconds.\n",
      "Test circle -281 trained in 2.79 seconds.\n",
      "Test circle -280 trained in 2.77 seconds.\n",
      "Test circle -279 trained in 2.75 seconds.\n",
      "Test circle -278 trained in 2.82 seconds.\n",
      "Test circle -277 trained in 2.75 seconds.\n",
      "Test circle -276 trained in 2.84 seconds.\n",
      "Test circle -275 trained in 2.8 seconds.\n",
      "Test circle -274 trained in 2.78 seconds.\n",
      "Test circle -273 trained in 2.75 seconds.\n",
      "Test circle -272 trained in 2.77 seconds.\n",
      "Test circle -271 trained in 2.77 seconds.\n",
      "Test circle -270 trained in 2.77 seconds.\n",
      "Test circle -269 trained in 2.75 seconds.\n",
      "Test circle -268 trained in 2.84 seconds.\n",
      "Test circle -267 trained in 2.76 seconds.\n",
      "Test circle -266 trained in 2.78 seconds.\n",
      "Test circle -265 trained in 2.78 seconds.\n",
      "Test circle -264 trained in 2.76 seconds.\n",
      "Test circle -263 trained in 2.76 seconds.\n",
      "Test circle -262 trained in 2.74 seconds.\n",
      "Test circle -261 trained in 2.74 seconds.\n",
      "Test circle -260 trained in 2.78 seconds.\n",
      "Test circle -259 trained in 2.74 seconds.\n",
      "Test circle -258 trained in 2.74 seconds.\n",
      "Test circle -257 trained in 2.77 seconds.\n",
      "Test circle -256 trained in 2.76 seconds.\n",
      "Test circle -255 trained in 2.76 seconds.\n",
      "Test circle -254 trained in 2.75 seconds.\n",
      "Test circle -253 trained in 2.76 seconds.\n",
      "Test circle -252 trained in 2.74 seconds.\n",
      "Test circle -251 trained in 2.75 seconds.\n",
      "Test circle -250 trained in 2.84 seconds.\n",
      "Test circle -249 trained in 2.76 seconds.\n",
      "Test circle -248 trained in 2.75 seconds.\n",
      "Test circle -247 trained in 2.75 seconds.\n",
      "Test circle -246 trained in 2.77 seconds.\n",
      "Test circle -245 trained in 2.75 seconds.\n",
      "Test circle -244 trained in 2.87 seconds.\n",
      "Test circle -243 trained in 2.76 seconds.\n",
      "Test circle -242 trained in 2.76 seconds.\n",
      "Test circle -241 trained in 2.74 seconds.\n",
      "Test circle -240 trained in 2.76 seconds.\n",
      "Test circle -239 trained in 2.76 seconds.\n",
      "Test circle -238 trained in 2.76 seconds.\n",
      "Test circle -237 trained in 2.74 seconds.\n",
      "Test circle -236 trained in 2.74 seconds.\n",
      "Test circle -235 trained in 2.78 seconds.\n",
      "Test circle -234 trained in 2.73 seconds.\n",
      "Test circle -233 trained in 2.75 seconds.\n",
      "Test circle -232 trained in 2.77 seconds.\n",
      "Test circle -231 trained in 2.84 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -230 trained in 2.76 seconds.\n",
      "Test circle -229 trained in 2.74 seconds.\n",
      "Test circle -228 trained in 2.77 seconds.\n",
      "Test circle -227 trained in 2.75 seconds.\n",
      "Test circle -226 trained in 2.74 seconds.\n",
      "Test circle -225 trained in 2.74 seconds.\n",
      "Test circle -224 trained in 2.75 seconds.\n",
      "Test circle -223 trained in 2.74 seconds.\n",
      "Test circle -222 trained in 2.78 seconds.\n",
      "Test circle -221 trained in 3.18 seconds.\n",
      "Test circle -220 trained in 3.24 seconds.\n",
      "Test circle -219 trained in 3.66 seconds.\n",
      "Test circle -218 trained in 3.46 seconds.\n",
      "Test circle -217 trained in 2.99 seconds.\n",
      "Test circle -216 trained in 2.88 seconds.\n",
      "Test circle -215 trained in 3.25 seconds.\n",
      "Test circle -214 trained in 3.98 seconds.\n",
      "Test circle -213 trained in 3.1 seconds.\n",
      "Test circle -212 trained in 2.85 seconds.\n",
      "Test circle -211 trained in 2.83 seconds.\n",
      "Test circle -210 trained in 2.91 seconds.\n",
      "Test circle -209 trained in 2.82 seconds.\n",
      "Test circle -208 trained in 2.76 seconds.\n",
      "Test circle -207 trained in 2.76 seconds.\n",
      "Test circle -206 trained in 2.75 seconds.\n",
      "Test circle -205 trained in 2.77 seconds.\n",
      "Test circle -204 trained in 2.85 seconds.\n",
      "Test circle -203 trained in 2.81 seconds.\n",
      "Test circle -202 trained in 2.86 seconds.\n",
      "Test circle -201 trained in 2.88 seconds.\n",
      "Test circle -200 trained in 2.92 seconds.\n",
      "Test circle -199 trained in 3.25 seconds.\n",
      "Test circle -198 trained in 2.91 seconds.\n",
      "Test circle -197 trained in 2.78 seconds.\n",
      "Test circle -196 trained in 2.92 seconds.\n",
      "Test circle -195 trained in 2.76 seconds.\n",
      "Test circle -194 trained in 2.85 seconds.\n",
      "Test circle -193 trained in 2.88 seconds.\n",
      "Test circle -192 trained in 2.88 seconds.\n",
      "Test circle -191 trained in 2.79 seconds.\n",
      "Test circle -190 trained in 2.75 seconds.\n",
      "Test circle -189 trained in 2.76 seconds.\n",
      "Test circle -188 trained in 2.81 seconds.\n",
      "Test circle -187 trained in 2.88 seconds.\n",
      "Test circle -186 trained in 2.76 seconds.\n",
      "Test circle -185 trained in 2.79 seconds.\n",
      "Test circle -184 trained in 2.75 seconds.\n",
      "Test circle -183 trained in 2.88 seconds.\n",
      "Test circle -182 trained in 2.82 seconds.\n",
      "Test circle -181 trained in 2.85 seconds.\n",
      "Test circle -180 trained in 2.86 seconds.\n",
      "Test circle -179 trained in 2.76 seconds.\n",
      "Test circle -178 trained in 2.74 seconds.\n",
      "Test circle -177 trained in 2.75 seconds.\n",
      "Test circle -176 trained in 2.73 seconds.\n",
      "Test circle -175 trained in 2.83 seconds.\n",
      "Test circle -174 trained in 2.75 seconds.\n",
      "Test circle -173 trained in 2.75 seconds.\n",
      "Test circle -172 trained in 2.8 seconds.\n",
      "Test circle -171 trained in 2.74 seconds.\n",
      "Test circle -170 trained in 2.77 seconds.\n",
      "Test circle -169 trained in 2.8 seconds.\n",
      "Test circle -168 trained in 3.25 seconds.\n",
      "Test circle -167 trained in 2.84 seconds.\n",
      "Test circle -166 trained in 2.78 seconds.\n",
      "Test circle -165 trained in 2.76 seconds.\n",
      "Test circle -164 trained in 2.76 seconds.\n",
      "Test circle -163 trained in 2.79 seconds.\n",
      "Test circle -162 trained in 2.76 seconds.\n",
      "Test circle -161 trained in 2.82 seconds.\n",
      "Test circle -160 trained in 2.76 seconds.\n",
      "Test circle -159 trained in 2.75 seconds.\n",
      "Test circle -158 trained in 2.75 seconds.\n",
      "Test circle -157 trained in 2.83 seconds.\n",
      "Test circle -156 trained in 2.77 seconds.\n",
      "Test circle -155 trained in 2.75 seconds.\n",
      "Test circle -154 trained in 2.74 seconds.\n",
      "Test circle -153 trained in 2.74 seconds.\n",
      "Test circle -152 trained in 2.74 seconds.\n",
      "Test circle -151 trained in 2.79 seconds.\n",
      "Test circle -150 trained in 2.73 seconds.\n",
      "Test circle -149 trained in 2.75 seconds.\n",
      "Test circle -148 trained in 2.73 seconds.\n",
      "Test circle -147 trained in 2.81 seconds.\n",
      "Test circle -146 trained in 2.77 seconds.\n",
      "Test circle -145 trained in 2.73 seconds.\n",
      "Test circle -144 trained in 2.74 seconds.\n",
      "Test circle -143 trained in 2.73 seconds.\n",
      "Test circle -142 trained in 2.73 seconds.\n",
      "Test circle -141 trained in 2.74 seconds.\n",
      "Test circle -140 trained in 2.73 seconds.\n",
      "Test circle -139 trained in 2.72 seconds.\n",
      "Test circle -138 trained in 2.85 seconds.\n",
      "Test circle -137 trained in 2.76 seconds.\n",
      "Test circle -136 trained in 2.74 seconds.\n",
      "Test circle -135 trained in 2.75 seconds.\n",
      "Test circle -134 trained in 2.74 seconds.\n",
      "Test circle -133 trained in 2.74 seconds.\n",
      "Test circle -132 trained in 2.75 seconds.\n",
      "Test circle -131 trained in 2.73 seconds.\n",
      "Test circle -130 trained in 2.74 seconds.\n",
      "Test circle -129 trained in 2.78 seconds.\n",
      "Test circle -128 trained in 2.74 seconds.\n",
      "Test circle -127 trained in 2.74 seconds.\n",
      "Test circle -126 trained in 2.75 seconds.\n",
      "Test circle -125 trained in 2.73 seconds.\n",
      "Test circle -124 trained in 2.74 seconds.\n",
      "Test circle -123 trained in 2.73 seconds.\n",
      "Test circle -122 trained in 2.74 seconds.\n",
      "Test circle -121 trained in 2.73 seconds.\n",
      "Test circle -120 trained in 2.82 seconds.\n",
      "Test circle -119 trained in 2.75 seconds.\n",
      "Test circle -118 trained in 2.73 seconds.\n",
      "Test circle -117 trained in 2.76 seconds.\n",
      "Test circle -116 trained in 2.8 seconds.\n",
      "Test circle -115 trained in 2.74 seconds.\n",
      "Test circle -114 trained in 2.73 seconds.\n",
      "Test circle -113 trained in 2.73 seconds.\n",
      "Test circle -112 trained in 2.74 seconds.\n",
      "Test circle -111 trained in 2.74 seconds.\n",
      "Test circle -110 trained in 2.74 seconds.\n",
      "Test circle -109 trained in 2.75 seconds.\n",
      "Test circle -108 trained in 2.73 seconds.\n",
      "Test circle -107 trained in 2.78 seconds.\n",
      "Test circle -106 trained in 2.75 seconds.\n",
      "Test circle -105 trained in 2.73 seconds.\n",
      "Test circle -104 trained in 2.74 seconds.\n",
      "Test circle -103 trained in 2.74 seconds.\n",
      "Test circle -102 trained in 2.74 seconds.\n",
      "Test circle -101 trained in 2.84 seconds.\n",
      "Test circle -100 trained in 2.74 seconds.\n",
      "Test circle -99 trained in 2.73 seconds.\n",
      "Test circle -98 trained in 2.73 seconds.\n",
      "Test circle -97 trained in 2.75 seconds.\n",
      "Test circle -96 trained in 2.8 seconds.\n",
      "Test circle -95 trained in 2.78 seconds.\n",
      "Test circle -94 trained in 2.76 seconds.\n",
      "Test circle -93 trained in 2.74 seconds.\n",
      "Test circle -92 trained in 2.74 seconds.\n",
      "Test circle -91 trained in 2.73 seconds.\n",
      "Test circle -90 trained in 2.76 seconds.\n",
      "Test circle -89 trained in 2.75 seconds.\n",
      "Test circle -88 trained in 2.74 seconds.\n",
      "Test circle -87 trained in 2.74 seconds.\n",
      "Test circle -86 trained in 2.74 seconds.\n",
      "Test circle -85 trained in 2.77 seconds.\n",
      "Test circle -84 trained in 2.74 seconds.\n",
      "Test circle -83 trained in 2.8 seconds.\n",
      "Test circle -82 trained in 2.76 seconds.\n",
      "Test circle -81 trained in 2.73 seconds.\n",
      "Test circle -80 trained in 2.73 seconds.\n",
      "Test circle -79 trained in 2.73 seconds.\n",
      "Test circle -78 trained in 2.86 seconds.\n",
      "Test circle -77 trained in 2.73 seconds.\n",
      "Test circle -76 trained in 2.91 seconds.\n",
      "Test circle -75 trained in 3.02 seconds.\n",
      "Test circle -74 trained in 3.2 seconds.\n",
      "Test circle -73 trained in 2.93 seconds.\n",
      "Test circle -72 trained in 2.88 seconds.\n",
      "Test circle -71 trained in 2.93 seconds.\n",
      "Test circle -70 trained in 2.99 seconds.\n",
      "Test circle -69 trained in 2.91 seconds.\n",
      "Test circle -68 trained in 2.81 seconds.\n",
      "Test circle -67 trained in 2.85 seconds.\n",
      "Test circle -66 trained in 2.78 seconds.\n",
      "Test circle -65 trained in 2.88 seconds.\n",
      "Test circle -64 trained in 3.5 seconds.\n",
      "Test circle -63 trained in 2.85 seconds.\n",
      "Test circle -62 trained in 3.19 seconds.\n",
      "Test circle -61 trained in 3.1 seconds.\n",
      "Test circle -60 trained in 3.09 seconds.\n",
      "Test circle -59 trained in 2.83 seconds.\n",
      "Test circle -58 trained in 2.99 seconds.\n",
      "Test circle -57 trained in 2.99 seconds.\n",
      "Test circle -56 trained in 2.85 seconds.\n",
      "Test circle -55 trained in 2.88 seconds.\n",
      "Test circle -54 trained in 2.86 seconds.\n",
      "Test circle -53 trained in 3.02 seconds.\n",
      "Test circle -52 trained in 3.24 seconds.\n",
      "Test circle -51 trained in 3.1 seconds.\n",
      "Test circle -50 trained in 2.87 seconds.\n",
      "Test circle -49 trained in 2.97 seconds.\n",
      "Test circle -48 trained in 2.81 seconds.\n",
      "Test circle -47 trained in 2.83 seconds.\n",
      "Test circle -46 trained in 2.98 seconds.\n",
      "Test circle -45 trained in 3.07 seconds.\n",
      "Test circle -44 trained in 2.89 seconds.\n",
      "Test circle -43 trained in 2.89 seconds.\n",
      "Test circle -42 trained in 2.91 seconds.\n",
      "Test circle -41 trained in 2.78 seconds.\n",
      "Test circle -40 trained in 2.76 seconds.\n",
      "Test circle -39 trained in 2.74 seconds.\n",
      "Test circle -38 trained in 2.76 seconds.\n",
      "Test circle -37 trained in 2.76 seconds.\n",
      "Test circle -36 trained in 2.75 seconds.\n",
      "Test circle -35 trained in 2.74 seconds.\n",
      "Test circle -34 trained in 2.93 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -33 trained in 3.19 seconds.\n",
      "Test circle -32 trained in 3.06 seconds.\n",
      "Test circle -31 trained in 2.98 seconds.\n",
      "Test circle -30 trained in 2.8 seconds.\n",
      "Test circle -29 trained in 2.81 seconds.\n",
      "Test circle -28 trained in 2.77 seconds.\n",
      "Test circle -27 trained in 2.84 seconds.\n",
      "Test circle -26 trained in 2.77 seconds.\n",
      "Test circle -25 trained in 2.74 seconds.\n",
      "Test circle -24 trained in 2.75 seconds.\n",
      "Test circle -23 trained in 2.79 seconds.\n",
      "Test circle -22 trained in 2.74 seconds.\n",
      "Test circle -21 trained in 2.73 seconds.\n",
      "Test circle -20 trained in 2.75 seconds.\n",
      "Test circle -19 trained in 2.73 seconds.\n",
      "Test circle -18 trained in 2.73 seconds.\n",
      "Test circle -17 trained in 2.97 seconds.\n",
      "Test circle -16 trained in 2.84 seconds.\n",
      "Test circle -15 trained in 2.89 seconds.\n",
      "Test circle -14 trained in 2.84 seconds.\n",
      "Test circle -13 trained in 2.9 seconds.\n",
      "Test circle -12 trained in 2.87 seconds.\n",
      "Test circle -11 trained in 3.59 seconds.\n",
      "Test circle -10 trained in 3.8 seconds.\n",
      "Test circle -9 trained in 3.65 seconds.\n",
      "Test circle -8 trained in 3.09 seconds.\n",
      "Test circle -7 trained in 3.42 seconds.\n",
      "Test circle -6 trained in 3.22 seconds.\n",
      "Test circle -5 trained in 3.46 seconds.\n",
      "Test circle -4 trained in 3.29 seconds.\n",
      "Test circle -3 trained in 3.05 seconds.\n",
      "Test circle -2 trained in 3.03 seconds.\n",
      "Test circle -1 trained in 2.86 seconds.\n",
      "Model trained in 2096.99 seconds.\n",
      "Hyperparams (50, 6, 3) trained in 2096.99 seconds.\n",
      "=========================================================================================\n",
      "Model (50, 6, 3) Ended\n",
      "=========================================================================================\n",
      "=========================================================================================\n",
      "Model (50, 6, 6) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 2.91 seconds.\n",
      "Test circle -724 trained in 3.15 seconds.\n",
      "Test circle -723 trained in 2.99 seconds.\n",
      "Test circle -722 trained in 3.0 seconds.\n",
      "Test circle -721 trained in 2.95 seconds.\n",
      "Test circle -720 trained in 3.01 seconds.\n",
      "Test circle -719 trained in 3.04 seconds.\n",
      "Test circle -718 trained in 3.09 seconds.\n",
      "Test circle -717 trained in 3.17 seconds.\n",
      "Test circle -716 trained in 3.08 seconds.\n",
      "Test circle -715 trained in 3.16 seconds.\n",
      "Test circle -714 trained in 2.88 seconds.\n",
      "Test circle -713 trained in 2.98 seconds.\n",
      "Test circle -712 trained in 2.98 seconds.\n",
      "Test circle -711 trained in 2.89 seconds.\n",
      "Test circle -710 trained in 2.89 seconds.\n",
      "Test circle -709 trained in 2.92 seconds.\n",
      "Test circle -708 trained in 3.07 seconds.\n",
      "Test circle -707 trained in 2.91 seconds.\n",
      "Test circle -706 trained in 3.06 seconds.\n",
      "Test circle -705 trained in 2.97 seconds.\n",
      "Test circle -704 trained in 2.98 seconds.\n",
      "Test circle -703 trained in 2.99 seconds.\n",
      "Test circle -702 trained in 3.02 seconds.\n",
      "Test circle -701 trained in 3.02 seconds.\n",
      "Test circle -700 trained in 3.5 seconds.\n",
      "Test circle -699 trained in 2.9 seconds.\n",
      "Test circle -698 trained in 2.98 seconds.\n",
      "Test circle -697 trained in 2.97 seconds.\n",
      "Test circle -696 trained in 2.96 seconds.\n",
      "Test circle -695 trained in 2.91 seconds.\n",
      "Test circle -694 trained in 2.87 seconds.\n",
      "Test circle -693 trained in 2.84 seconds.\n",
      "Test circle -692 trained in 2.85 seconds.\n",
      "Test circle -691 trained in 2.86 seconds.\n",
      "Test circle -690 trained in 2.83 seconds.\n",
      "Test circle -689 trained in 2.91 seconds.\n",
      "Test circle -688 trained in 2.88 seconds.\n",
      "Test circle -687 trained in 2.85 seconds.\n",
      "Test circle -686 trained in 2.86 seconds.\n",
      "Test circle -685 trained in 2.87 seconds.\n",
      "Test circle -684 trained in 2.86 seconds.\n",
      "Test circle -683 trained in 2.86 seconds.\n",
      "Test circle -682 trained in 2.85 seconds.\n",
      "Test circle -681 trained in 2.85 seconds.\n",
      "Test circle -680 trained in 2.9 seconds.\n",
      "Test circle -679 trained in 2.86 seconds.\n",
      "Test circle -678 trained in 3.06 seconds.\n",
      "Test circle -677 trained in 2.98 seconds.\n",
      "Test circle -676 trained in 2.9 seconds.\n",
      "Test circle -675 trained in 2.87 seconds.\n",
      "Test circle -674 trained in 2.87 seconds.\n",
      "Test circle -673 trained in 2.86 seconds.\n",
      "Test circle -672 trained in 2.88 seconds.\n",
      "Test circle -671 trained in 2.87 seconds.\n",
      "Test circle -670 trained in 2.85 seconds.\n",
      "Test circle -669 trained in 2.86 seconds.\n",
      "Test circle -668 trained in 2.89 seconds.\n",
      "Test circle -667 trained in 2.9 seconds.\n",
      "Test circle -666 trained in 2.87 seconds.\n",
      "Test circle -665 trained in 2.86 seconds.\n",
      "Test circle -664 trained in 2.85 seconds.\n",
      "Test circle -663 trained in 2.9 seconds.\n",
      "Test circle -662 trained in 2.87 seconds.\n",
      "Test circle -661 trained in 2.86 seconds.\n",
      "Test circle -660 trained in 2.83 seconds.\n",
      "Test circle -659 trained in 2.98 seconds.\n",
      "Test circle -658 trained in 2.97 seconds.\n",
      "Test circle -657 trained in 2.88 seconds.\n",
      "Test circle -656 trained in 2.89 seconds.\n",
      "Test circle -655 trained in 2.9 seconds.\n",
      "Test circle -654 trained in 3.02 seconds.\n",
      "Test circle -653 trained in 2.86 seconds.\n",
      "Test circle -652 trained in 2.87 seconds.\n",
      "Test circle -651 trained in 2.87 seconds.\n",
      "Test circle -650 trained in 2.92 seconds.\n",
      "Test circle -649 trained in 2.88 seconds.\n",
      "Test circle -648 trained in 2.88 seconds.\n",
      "Test circle -647 trained in 2.91 seconds.\n",
      "Test circle -646 trained in 2.86 seconds.\n",
      "Test circle -645 trained in 2.9 seconds.\n",
      "Test circle -644 trained in 2.89 seconds.\n",
      "Test circle -643 trained in 2.84 seconds.\n",
      "Test circle -642 trained in 2.87 seconds.\n",
      "Test circle -641 trained in 2.84 seconds.\n",
      "Test circle -640 trained in 2.94 seconds.\n",
      "Test circle -639 trained in 2.91 seconds.\n",
      "Test circle -638 trained in 2.86 seconds.\n",
      "Test circle -637 trained in 2.88 seconds.\n",
      "Test circle -636 trained in 2.99 seconds.\n",
      "Test circle -635 trained in 2.87 seconds.\n",
      "Test circle -634 trained in 2.9 seconds.\n",
      "Test circle -633 trained in 2.86 seconds.\n",
      "Test circle -632 trained in 2.85 seconds.\n",
      "Test circle -631 trained in 2.91 seconds.\n",
      "Test circle -630 trained in 2.88 seconds.\n",
      "Test circle -629 trained in 2.9 seconds.\n",
      "Test circle -628 trained in 2.88 seconds.\n",
      "Test circle -627 trained in 2.9 seconds.\n",
      "Test circle -626 trained in 3.0 seconds.\n",
      "Test circle -625 trained in 2.88 seconds.\n",
      "Test circle -624 trained in 2.86 seconds.\n",
      "Test circle -623 trained in 2.89 seconds.\n",
      "Test circle -622 trained in 2.83 seconds.\n",
      "Test circle -621 trained in 2.95 seconds.\n",
      "Test circle -620 trained in 2.88 seconds.\n",
      "Test circle -619 trained in 2.86 seconds.\n",
      "Test circle -618 trained in 2.86 seconds.\n",
      "Test circle -617 trained in 2.86 seconds.\n",
      "Test circle -616 trained in 2.85 seconds.\n",
      "Test circle -615 trained in 2.87 seconds.\n",
      "Test circle -614 trained in 2.99 seconds.\n",
      "Test circle -613 trained in 2.85 seconds.\n",
      "Test circle -612 trained in 2.86 seconds.\n",
      "Test circle -611 trained in 2.84 seconds.\n",
      "Test circle -610 trained in 2.9 seconds.\n",
      "Test circle -609 trained in 2.86 seconds.\n",
      "Test circle -608 trained in 2.85 seconds.\n",
      "Test circle -607 trained in 2.86 seconds.\n",
      "Test circle -606 trained in 2.86 seconds.\n",
      "Test circle -605 trained in 2.87 seconds.\n",
      "Test circle -604 trained in 2.97 seconds.\n",
      "Test circle -603 trained in 2.91 seconds.\n",
      "Test circle -602 trained in 2.89 seconds.\n",
      "Test circle -601 trained in 2.86 seconds.\n",
      "Test circle -600 trained in 2.87 seconds.\n",
      "Test circle -599 trained in 2.86 seconds.\n",
      "Test circle -598 trained in 2.87 seconds.\n",
      "Test circle -597 trained in 2.86 seconds.\n",
      "Test circle -596 trained in 2.86 seconds.\n",
      "Test circle -595 trained in 2.88 seconds.\n",
      "Test circle -594 trained in 2.97 seconds.\n",
      "Test circle -593 trained in 3.45 seconds.\n",
      "Test circle -592 trained in 3.61 seconds.\n",
      "Test circle -591 trained in 3.6 seconds.\n",
      "Test circle -590 trained in 3.64 seconds.\n",
      "Test circle -589 trained in 3.58 seconds.\n",
      "Test circle -588 trained in 3.57 seconds.\n",
      "Test circle -587 trained in 3.57 seconds.\n",
      "Test circle -586 trained in 3.57 seconds.\n",
      "Test circle -585 trained in 3.63 seconds.\n",
      "Test circle -584 trained in 3.58 seconds.\n",
      "Test circle -583 trained in 3.51 seconds.\n",
      "Test circle -582 trained in 3.48 seconds.\n",
      "Test circle -581 trained in 3.56 seconds.\n",
      "Test circle -580 trained in 3.56 seconds.\n",
      "Test circle -579 trained in 3.64 seconds.\n",
      "Test circle -578 trained in 3.59 seconds.\n",
      "Test circle -577 trained in 3.85 seconds.\n",
      "Test circle -576 trained in 3.61 seconds.\n",
      "Test circle -575 trained in 3.57 seconds.\n",
      "Test circle -574 trained in 3.58 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -573 trained in 3.56 seconds.\n",
      "Test circle -572 trained in 3.57 seconds.\n",
      "Test circle -571 trained in 3.59 seconds.\n",
      "Test circle -570 trained in 3.6 seconds.\n",
      "Test circle -569 trained in 3.58 seconds.\n",
      "Test circle -568 trained in 3.59 seconds.\n",
      "Test circle -567 trained in 3.67 seconds.\n",
      "Test circle -566 trained in 3.59 seconds.\n",
      "Test circle -565 trained in 3.92 seconds.\n",
      "Test circle -564 trained in 3.58 seconds.\n",
      "Test circle -563 trained in 3.54 seconds.\n",
      "Test circle -562 trained in 3.52 seconds.\n",
      "Test circle -561 trained in 3.59 seconds.\n",
      "Test circle -560 trained in 3.59 seconds.\n",
      "Test circle -559 trained in 3.57 seconds.\n",
      "Test circle -558 trained in 3.57 seconds.\n",
      "Test circle -557 trained in 3.58 seconds.\n",
      "Test circle -556 trained in 3.53 seconds.\n",
      "Test circle -555 trained in 3.57 seconds.\n",
      "Test circle -554 trained in 3.54 seconds.\n",
      "Test circle -553 trained in 3.61 seconds.\n",
      "Test circle -552 trained in 3.64 seconds.\n",
      "Test circle -551 trained in 3.71 seconds.\n",
      "Test circle -550 trained in 3.6 seconds.\n",
      "Test circle -549 trained in 3.95 seconds.\n",
      "Test circle -548 trained in 3.95 seconds.\n",
      "Test circle -547 trained in 3.9 seconds.\n",
      "Test circle -546 trained in 4.1 seconds.\n",
      "Test circle -545 trained in 3.51 seconds.\n",
      "Test circle -544 trained in 3.46 seconds.\n",
      "Test circle -543 trained in 3.42 seconds.\n",
      "Test circle -542 trained in 3.45 seconds.\n",
      "Test circle -541 trained in 3.28 seconds.\n",
      "Test circle -540 trained in 3.23 seconds.\n",
      "Test circle -539 trained in 3.24 seconds.\n",
      "Test circle -538 trained in 3.32 seconds.\n",
      "Test circle -537 trained in 3.32 seconds.\n",
      "Test circle -536 trained in 3.44 seconds.\n",
      "Test circle -535 trained in 2.95 seconds.\n",
      "Test circle -534 trained in 2.75 seconds.\n",
      "Test circle -533 trained in 2.63 seconds.\n",
      "Test circle -532 trained in 2.63 seconds.\n",
      "Test circle -531 trained in 2.83 seconds.\n",
      "Test circle -530 trained in 2.69 seconds.\n",
      "Test circle -529 trained in 2.65 seconds.\n",
      "Test circle -528 trained in 2.72 seconds.\n",
      "Test circle -527 trained in 2.71 seconds.\n",
      "Test circle -526 trained in 2.67 seconds.\n",
      "Test circle -525 trained in 2.8 seconds.\n",
      "Test circle -524 trained in 2.73 seconds.\n",
      "Test circle -523 trained in 2.68 seconds.\n",
      "Test circle -522 trained in 2.99 seconds.\n",
      "Test circle -521 trained in 2.72 seconds.\n",
      "Test circle -520 trained in 2.7 seconds.\n",
      "Test circle -519 trained in 2.75 seconds.\n",
      "Test circle -518 trained in 2.76 seconds.\n",
      "Test circle -517 trained in 2.69 seconds.\n",
      "Test circle -516 trained in 2.7 seconds.\n",
      "Test circle -515 trained in 2.76 seconds.\n",
      "Test circle -514 trained in 2.71 seconds.\n",
      "Test circle -513 trained in 2.82 seconds.\n",
      "Test circle -512 trained in 2.74 seconds.\n",
      "Test circle -511 trained in 2.7 seconds.\n",
      "Test circle -510 trained in 2.75 seconds.\n",
      "Test circle -509 trained in 2.76 seconds.\n",
      "Test circle -508 trained in 2.69 seconds.\n",
      "Test circle -507 trained in 2.72 seconds.\n",
      "Test circle -506 trained in 2.72 seconds.\n",
      "Test circle -505 trained in 2.7 seconds.\n",
      "Test circle -504 trained in 2.77 seconds.\n",
      "Test circle -503 trained in 2.71 seconds.\n",
      "Test circle -502 trained in 2.71 seconds.\n",
      "Test circle -501 trained in 2.86 seconds.\n",
      "Test circle -500 trained in 2.76 seconds.\n",
      "Test circle -499 trained in 2.7 seconds.\n",
      "Test circle -498 trained in 2.75 seconds.\n",
      "Test circle -497 trained in 2.74 seconds.\n",
      "Test circle -496 trained in 2.71 seconds.\n",
      "Test circle -495 trained in 2.78 seconds.\n",
      "Test circle -494 trained in 2.84 seconds.\n",
      "Test circle -493 trained in 2.75 seconds.\n",
      "Test circle -492 trained in 2.8 seconds.\n",
      "Test circle -491 trained in 2.76 seconds.\n",
      "Test circle -490 trained in 2.7 seconds.\n",
      "Test circle -489 trained in 2.75 seconds.\n",
      "Test circle -488 trained in 2.76 seconds.\n",
      "Test circle -487 trained in 2.74 seconds.\n",
      "Test circle -486 trained in 2.77 seconds.\n",
      "Test circle -485 trained in 2.75 seconds.\n",
      "Test circle -484 trained in 2.72 seconds.\n",
      "Test circle -483 trained in 2.77 seconds.\n",
      "Test circle -482 trained in 2.77 seconds.\n",
      "Test circle -481 trained in 2.75 seconds.\n",
      "Test circle -480 trained in 2.77 seconds.\n",
      "Test circle -479 trained in 2.76 seconds.\n",
      "Test circle -478 trained in 2.71 seconds.\n",
      "Test circle -477 trained in 2.75 seconds.\n",
      "Test circle -476 trained in 2.74 seconds.\n",
      "Test circle -475 trained in 2.81 seconds.\n",
      "Test circle -474 trained in 2.82 seconds.\n",
      "Test circle -473 trained in 2.76 seconds.\n",
      "Test circle -472 trained in 2.73 seconds.\n",
      "Test circle -471 trained in 2.78 seconds.\n",
      "Test circle -470 trained in 2.75 seconds.\n",
      "Test circle -469 trained in 2.72 seconds.\n",
      "Test circle -468 trained in 2.75 seconds.\n",
      "Test circle -467 trained in 2.76 seconds.\n",
      "Test circle -466 trained in 2.71 seconds.\n",
      "Test circle -465 trained in 2.76 seconds.\n",
      "Test circle -464 trained in 2.75 seconds.\n",
      "Test circle -463 trained in 2.72 seconds.\n",
      "Test circle -462 trained in 2.8 seconds.\n",
      "Test circle -461 trained in 2.76 seconds.\n",
      "Test circle -460 trained in 2.73 seconds.\n",
      "Test circle -459 trained in 2.75 seconds.\n",
      "Test circle -458 trained in 2.78 seconds.\n",
      "Test circle -457 trained in 2.8 seconds.\n",
      "Test circle -456 trained in 2.78 seconds.\n",
      "Test circle -455 trained in 2.76 seconds.\n",
      "Test circle -454 trained in 2.72 seconds.\n",
      "Test circle -453 trained in 2.78 seconds.\n",
      "Test circle -452 trained in 2.77 seconds.\n",
      "Test circle -451 trained in 2.72 seconds.\n",
      "Test circle -450 trained in 2.78 seconds.\n",
      "Test circle -449 trained in 2.8 seconds.\n",
      "Test circle -448 trained in 2.73 seconds.\n",
      "Test circle -447 trained in 2.76 seconds.\n",
      "Test circle -446 trained in 2.77 seconds.\n",
      "Test circle -445 trained in 2.74 seconds.\n",
      "Test circle -444 trained in 2.78 seconds.\n",
      "Test circle -443 trained in 2.76 seconds.\n",
      "Test circle -442 trained in 2.75 seconds.\n",
      "Test circle -441 trained in 2.77 seconds.\n",
      "Test circle -440 trained in 2.76 seconds.\n",
      "Test circle -439 trained in 2.74 seconds.\n",
      "Test circle -438 trained in 2.95 seconds.\n",
      "Test circle -437 trained in 2.8 seconds.\n",
      "Test circle -436 trained in 2.74 seconds.\n",
      "Test circle -435 trained in 2.79 seconds.\n",
      "Test circle -434 trained in 2.79 seconds.\n",
      "Test circle -433 trained in 2.72 seconds.\n",
      "Test circle -432 trained in 2.77 seconds.\n",
      "Test circle -431 trained in 2.76 seconds.\n",
      "Test circle -430 trained in 2.73 seconds.\n",
      "Test circle -429 trained in 2.76 seconds.\n",
      "Test circle -428 trained in 2.75 seconds.\n",
      "Test circle -427 trained in 2.78 seconds.\n",
      "Test circle -426 trained in 2.78 seconds.\n",
      "Test circle -425 trained in 2.77 seconds.\n",
      "Test circle -424 trained in 2.74 seconds.\n",
      "Test circle -423 trained in 2.8 seconds.\n",
      "Test circle -422 trained in 2.78 seconds.\n",
      "Test circle -421 trained in 2.73 seconds.\n",
      "Test circle -420 trained in 2.86 seconds.\n",
      "Test circle -419 trained in 2.79 seconds.\n",
      "Test circle -418 trained in 2.74 seconds.\n",
      "Test circle -417 trained in 2.84 seconds.\n",
      "Test circle -416 trained in 2.82 seconds.\n",
      "Test circle -415 trained in 2.79 seconds.\n",
      "Test circle -414 trained in 2.83 seconds.\n",
      "Test circle -413 trained in 2.87 seconds.\n",
      "Test circle -412 trained in 2.8 seconds.\n",
      "Test circle -411 trained in 2.9 seconds.\n",
      "Test circle -410 trained in 2.92 seconds.\n",
      "Test circle -409 trained in 2.85 seconds.\n",
      "Test circle -408 trained in 2.92 seconds.\n",
      "Test circle -407 trained in 2.94 seconds.\n",
      "Test circle -406 trained in 2.94 seconds.\n",
      "Test circle -405 trained in 3.05 seconds.\n",
      "Test circle -404 trained in 2.95 seconds.\n",
      "Test circle -403 trained in 2.94 seconds.\n",
      "Test circle -402 trained in 3.16 seconds.\n",
      "Test circle -401 trained in 2.96 seconds.\n",
      "Test circle -400 trained in 3.01 seconds.\n",
      "Test circle -399 trained in 2.97 seconds.\n",
      "Test circle -398 trained in 2.63 seconds.\n",
      "Test circle -397 trained in 2.7 seconds.\n",
      "Test circle -396 trained in 2.72 seconds.\n",
      "Test circle -395 trained in 2.72 seconds.\n",
      "Test circle -394 trained in 2.76 seconds.\n",
      "Test circle -393 trained in 2.75 seconds.\n",
      "Test circle -392 trained in 2.7 seconds.\n",
      "Test circle -391 trained in 2.77 seconds.\n",
      "Test circle -390 trained in 2.75 seconds.\n",
      "Test circle -389 trained in 2.72 seconds.\n",
      "Test circle -388 trained in 2.75 seconds.\n",
      "Test circle -387 trained in 2.74 seconds.\n",
      "Test circle -386 trained in 2.72 seconds.\n",
      "Test circle -385 trained in 2.82 seconds.\n",
      "Test circle -384 trained in 2.78 seconds.\n",
      "Test circle -383 trained in 2.81 seconds.\n",
      "Test circle -382 trained in 2.74 seconds.\n",
      "Test circle -381 trained in 2.76 seconds.\n",
      "Test circle -380 trained in 2.73 seconds.\n",
      "Test circle -379 trained in 2.79 seconds.\n",
      "Test circle -378 trained in 2.8 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -377 trained in 2.72 seconds.\n",
      "Test circle -376 trained in 2.77 seconds.\n",
      "Test circle -375 trained in 2.79 seconds.\n",
      "Test circle -374 trained in 2.73 seconds.\n",
      "Test circle -373 trained in 2.75 seconds.\n",
      "Test circle -372 trained in 2.76 seconds.\n",
      "Test circle -371 trained in 2.71 seconds.\n",
      "Test circle -370 trained in 2.75 seconds.\n",
      "Test circle -369 trained in 2.8 seconds.\n",
      "Test circle -368 trained in 2.73 seconds.\n",
      "Test circle -367 trained in 2.75 seconds.\n",
      "Test circle -366 trained in 2.85 seconds.\n",
      "Test circle -365 trained in 2.84 seconds.\n",
      "Test circle -364 trained in 2.82 seconds.\n",
      "Test circle -363 trained in 2.91 seconds.\n",
      "Test circle -362 trained in 2.81 seconds.\n",
      "Test circle -361 trained in 2.88 seconds.\n",
      "Test circle -360 trained in 2.89 seconds.\n",
      "Test circle -359 trained in 2.85 seconds.\n",
      "Test circle -358 trained in 2.92 seconds.\n",
      "Test circle -357 trained in 2.92 seconds.\n",
      "Test circle -356 trained in 2.91 seconds.\n",
      "Test circle -355 trained in 2.97 seconds.\n",
      "Test circle -354 trained in 2.99 seconds.\n",
      "Test circle -353 trained in 3.04 seconds.\n",
      "Test circle -352 trained in 3.28 seconds.\n",
      "Test circle -351 trained in 2.99 seconds.\n",
      "Test circle -350 trained in 3.02 seconds.\n",
      "Test circle -349 trained in 3.12 seconds.\n",
      "Test circle -348 trained in 3.02 seconds.\n",
      "Test circle -347 trained in 2.75 seconds.\n",
      "Test circle -346 trained in 2.78 seconds.\n",
      "Test circle -345 trained in 2.66 seconds.\n",
      "Test circle -344 trained in 2.72 seconds.\n",
      "Test circle -343 trained in 2.76 seconds.\n",
      "Test circle -342 trained in 2.7 seconds.\n",
      "Test circle -341 trained in 2.74 seconds.\n",
      "Test circle -340 trained in 2.76 seconds.\n",
      "Test circle -339 trained in 2.7 seconds.\n",
      "Test circle -338 trained in 2.75 seconds.\n",
      "Test circle -337 trained in 2.77 seconds.\n",
      "Test circle -336 trained in 2.69 seconds.\n",
      "Test circle -335 trained in 2.77 seconds.\n",
      "Test circle -334 trained in 2.76 seconds.\n",
      "Test circle -333 trained in 2.69 seconds.\n",
      "Test circle -332 trained in 2.77 seconds.\n",
      "Test circle -331 trained in 2.76 seconds.\n",
      "Test circle -330 trained in 2.73 seconds.\n",
      "Test circle -329 trained in 2.79 seconds.\n",
      "Test circle -328 trained in 2.8 seconds.\n",
      "Test circle -327 trained in 2.73 seconds.\n",
      "Test circle -326 trained in 2.73 seconds.\n",
      "Test circle -325 trained in 2.73 seconds.\n",
      "Test circle -324 trained in 2.71 seconds.\n",
      "Test circle -323 trained in 2.71 seconds.\n",
      "Test circle -322 trained in 2.69 seconds.\n",
      "Test circle -321 trained in 2.72 seconds.\n",
      "Test circle -320 trained in 2.69 seconds.\n",
      "Test circle -319 trained in 2.69 seconds.\n",
      "Test circle -318 trained in 2.7 seconds.\n",
      "Test circle -317 trained in 2.71 seconds.\n",
      "Test circle -316 trained in 2.68 seconds.\n",
      "Test circle -315 trained in 2.7 seconds.\n",
      "Test circle -314 trained in 2.7 seconds.\n",
      "Test circle -313 trained in 2.7 seconds.\n",
      "Test circle -312 trained in 2.66 seconds.\n",
      "Test circle -311 trained in 2.66 seconds.\n",
      "Test circle -310 trained in 2.68 seconds.\n",
      "Test circle -309 trained in 2.85 seconds.\n",
      "Test circle -308 trained in 2.72 seconds.\n",
      "Test circle -307 trained in 2.72 seconds.\n",
      "Test circle -306 trained in 2.72 seconds.\n",
      "Test circle -305 trained in 2.72 seconds.\n",
      "Test circle -304 trained in 2.72 seconds.\n",
      "Test circle -303 trained in 2.75 seconds.\n",
      "Test circle -302 trained in 2.74 seconds.\n",
      "Test circle -301 trained in 2.76 seconds.\n",
      "Test circle -300 trained in 2.76 seconds.\n",
      "Test circle -299 trained in 2.81 seconds.\n",
      "Test circle -298 trained in 2.82 seconds.\n",
      "Test circle -297 trained in 2.81 seconds.\n",
      "Test circle -296 trained in 2.81 seconds.\n",
      "Test circle -295 trained in 2.79 seconds.\n",
      "Test circle -294 trained in 2.82 seconds.\n",
      "Test circle -293 trained in 2.8 seconds.\n",
      "Test circle -292 trained in 2.8 seconds.\n",
      "Test circle -291 trained in 2.9 seconds.\n",
      "Test circle -290 trained in 2.81 seconds.\n",
      "Test circle -289 trained in 2.8 seconds.\n",
      "Test circle -288 trained in 2.81 seconds.\n",
      "Test circle -287 trained in 2.82 seconds.\n",
      "Test circle -286 trained in 2.81 seconds.\n",
      "Test circle -285 trained in 2.8 seconds.\n",
      "Test circle -284 trained in 2.8 seconds.\n",
      "Test circle -283 trained in 2.79 seconds.\n",
      "Test circle -282 trained in 2.8 seconds.\n",
      "Test circle -281 trained in 2.8 seconds.\n",
      "Test circle -280 trained in 2.84 seconds.\n",
      "Test circle -279 trained in 2.8 seconds.\n",
      "Test circle -278 trained in 2.8 seconds.\n",
      "Test circle -277 trained in 2.82 seconds.\n",
      "Test circle -276 trained in 2.8 seconds.\n",
      "Test circle -275 trained in 2.78 seconds.\n",
      "Test circle -274 trained in 2.8 seconds.\n",
      "Test circle -273 trained in 2.88 seconds.\n",
      "Test circle -272 trained in 2.81 seconds.\n",
      "Test circle -271 trained in 2.8 seconds.\n",
      "Test circle -270 trained in 2.8 seconds.\n",
      "Test circle -269 trained in 2.8 seconds.\n",
      "Test circle -268 trained in 2.79 seconds.\n",
      "Test circle -267 trained in 2.8 seconds.\n",
      "Test circle -266 trained in 3.0 seconds.\n",
      "Test circle -265 trained in 2.81 seconds.\n",
      "Test circle -264 trained in 2.81 seconds.\n",
      "Test circle -263 trained in 2.8 seconds.\n",
      "Test circle -262 trained in 2.8 seconds.\n",
      "Test circle -261 trained in 2.8 seconds.\n",
      "Test circle -260 trained in 2.8 seconds.\n",
      "Test circle -259 trained in 2.81 seconds.\n",
      "Test circle -258 trained in 2.79 seconds.\n",
      "Test circle -257 trained in 2.79 seconds.\n",
      "Test circle -256 trained in 2.83 seconds.\n",
      "Test circle -255 trained in 2.8 seconds.\n",
      "Test circle -254 trained in 2.88 seconds.\n",
      "Test circle -253 trained in 2.81 seconds.\n",
      "Test circle -252 trained in 2.79 seconds.\n",
      "Test circle -251 trained in 2.79 seconds.\n",
      "Test circle -250 trained in 2.79 seconds.\n",
      "Test circle -249 trained in 2.8 seconds.\n",
      "Test circle -248 trained in 2.78 seconds.\n",
      "Test circle -247 trained in 2.8 seconds.\n",
      "Test circle -246 trained in 2.79 seconds.\n",
      "Test circle -245 trained in 2.79 seconds.\n",
      "Test circle -244 trained in 2.8 seconds.\n",
      "Test circle -243 trained in 2.79 seconds.\n",
      "Test circle -242 trained in 2.79 seconds.\n",
      "Test circle -241 trained in 2.8 seconds.\n",
      "Test circle -240 trained in 2.79 seconds.\n",
      "Test circle -239 trained in 2.78 seconds.\n",
      "Test circle -238 trained in 2.79 seconds.\n",
      "Test circle -237 trained in 2.78 seconds.\n",
      "Test circle -236 trained in 2.78 seconds.\n",
      "Test circle -235 trained in 2.91 seconds.\n",
      "Test circle -234 trained in 2.8 seconds.\n",
      "Test circle -233 trained in 2.79 seconds.\n",
      "Test circle -232 trained in 2.79 seconds.\n",
      "Test circle -231 trained in 2.79 seconds.\n",
      "Test circle -230 trained in 2.78 seconds.\n",
      "Test circle -229 trained in 2.78 seconds.\n",
      "Test circle -228 trained in 2.8 seconds.\n",
      "Test circle -227 trained in 2.82 seconds.\n",
      "Test circle -226 trained in 2.79 seconds.\n",
      "Test circle -225 trained in 2.79 seconds.\n",
      "Test circle -224 trained in 2.82 seconds.\n",
      "Test circle -223 trained in 2.85 seconds.\n",
      "Test circle -222 trained in 2.8 seconds.\n",
      "Test circle -221 trained in 2.8 seconds.\n",
      "Test circle -220 trained in 2.79 seconds.\n",
      "Test circle -219 trained in 2.78 seconds.\n",
      "Test circle -218 trained in 2.78 seconds.\n",
      "Test circle -217 trained in 2.87 seconds.\n",
      "Test circle -216 trained in 2.79 seconds.\n",
      "Test circle -215 trained in 2.81 seconds.\n",
      "Test circle -214 trained in 2.78 seconds.\n",
      "Test circle -213 trained in 2.84 seconds.\n",
      "Test circle -212 trained in 2.8 seconds.\n",
      "Test circle -211 trained in 2.79 seconds.\n",
      "Test circle -210 trained in 2.79 seconds.\n",
      "Test circle -209 trained in 2.77 seconds.\n",
      "Test circle -208 trained in 2.79 seconds.\n",
      "Test circle -207 trained in 2.81 seconds.\n",
      "Test circle -206 trained in 2.8 seconds.\n",
      "Test circle -205 trained in 2.78 seconds.\n",
      "Test circle -204 trained in 2.83 seconds.\n",
      "Test circle -203 trained in 2.78 seconds.\n",
      "Test circle -202 trained in 2.79 seconds.\n",
      "Test circle -201 trained in 2.79 seconds.\n",
      "Test circle -200 trained in 2.78 seconds.\n",
      "Test circle -199 trained in 2.78 seconds.\n",
      "Test circle -198 trained in 2.87 seconds.\n",
      "Test circle -197 trained in 2.79 seconds.\n",
      "Test circle -196 trained in 2.8 seconds.\n",
      "Test circle -195 trained in 2.8 seconds.\n",
      "Test circle -194 trained in 2.79 seconds.\n",
      "Test circle -193 trained in 2.78 seconds.\n",
      "Test circle -192 trained in 2.84 seconds.\n",
      "Test circle -191 trained in 2.8 seconds.\n",
      "Test circle -190 trained in 2.79 seconds.\n",
      "Test circle -189 trained in 2.78 seconds.\n",
      "Test circle -188 trained in 2.79 seconds.\n",
      "Test circle -187 trained in 2.77 seconds.\n",
      "Test circle -186 trained in 2.79 seconds.\n",
      "Test circle -185 trained in 2.79 seconds.\n",
      "Test circle -184 trained in 2.79 seconds.\n",
      "Test circle -183 trained in 2.76 seconds.\n",
      "Test circle -182 trained in 2.78 seconds.\n",
      "Test circle -181 trained in 2.81 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -180 trained in 2.85 seconds.\n",
      "Test circle -179 trained in 2.88 seconds.\n",
      "Test circle -178 trained in 2.79 seconds.\n",
      "Test circle -177 trained in 2.81 seconds.\n",
      "Test circle -176 trained in 2.79 seconds.\n",
      "Test circle -175 trained in 2.79 seconds.\n",
      "Test circle -174 trained in 2.79 seconds.\n",
      "Test circle -173 trained in 2.79 seconds.\n",
      "Test circle -172 trained in 2.78 seconds.\n",
      "Test circle -171 trained in 2.78 seconds.\n",
      "Test circle -170 trained in 2.83 seconds.\n",
      "Test circle -169 trained in 2.79 seconds.\n",
      "Test circle -168 trained in 2.79 seconds.\n",
      "Test circle -167 trained in 2.79 seconds.\n",
      "Test circle -166 trained in 2.78 seconds.\n",
      "Test circle -165 trained in 2.79 seconds.\n",
      "Test circle -164 trained in 2.78 seconds.\n",
      "Test circle -163 trained in 2.79 seconds.\n",
      "Test circle -162 trained in 2.77 seconds.\n",
      "Test circle -161 trained in 2.86 seconds.\n",
      "Test circle -160 trained in 2.78 seconds.\n",
      "Test circle -159 trained in 2.79 seconds.\n",
      "Test circle -158 trained in 2.77 seconds.\n",
      "Test circle -157 trained in 2.79 seconds.\n",
      "Test circle -156 trained in 2.79 seconds.\n",
      "Test circle -155 trained in 2.78 seconds.\n",
      "Test circle -154 trained in 2.79 seconds.\n",
      "Test circle -153 trained in 2.78 seconds.\n",
      "Test circle -152 trained in 2.78 seconds.\n",
      "Test circle -151 trained in 2.78 seconds.\n",
      "Test circle -150 trained in 2.78 seconds.\n",
      "Test circle -149 trained in 2.83 seconds.\n",
      "Test circle -148 trained in 2.77 seconds.\n",
      "Test circle -147 trained in 2.78 seconds.\n",
      "Test circle -146 trained in 2.79 seconds.\n",
      "Test circle -145 trained in 2.78 seconds.\n",
      "Test circle -144 trained in 2.77 seconds.\n",
      "Test circle -143 trained in 2.87 seconds.\n",
      "Test circle -142 trained in 2.79 seconds.\n",
      "Test circle -141 trained in 2.79 seconds.\n",
      "Test circle -140 trained in 2.78 seconds.\n",
      "Test circle -139 trained in 2.8 seconds.\n",
      "Test circle -138 trained in 2.83 seconds.\n",
      "Test circle -137 trained in 2.86 seconds.\n",
      "Test circle -136 trained in 2.81 seconds.\n",
      "Test circle -135 trained in 2.78 seconds.\n",
      "Test circle -134 trained in 2.78 seconds.\n",
      "Test circle -133 trained in 2.78 seconds.\n",
      "Test circle -132 trained in 2.79 seconds.\n",
      "Test circle -131 trained in 2.77 seconds.\n",
      "Test circle -130 trained in 2.79 seconds.\n",
      "Test circle -129 trained in 2.76 seconds.\n",
      "Test circle -128 trained in 2.78 seconds.\n",
      "Test circle -127 trained in 2.84 seconds.\n",
      "Test circle -126 trained in 2.76 seconds.\n",
      "Test circle -125 trained in 2.79 seconds.\n",
      "Test circle -124 trained in 2.86 seconds.\n",
      "Test circle -123 trained in 2.78 seconds.\n",
      "Test circle -122 trained in 2.8 seconds.\n",
      "Test circle -121 trained in 2.78 seconds.\n",
      "Test circle -120 trained in 2.78 seconds.\n",
      "Test circle -119 trained in 2.78 seconds.\n",
      "Test circle -118 trained in 2.78 seconds.\n",
      "Test circle -117 trained in 2.8 seconds.\n",
      "Test circle -116 trained in 2.77 seconds.\n",
      "Test circle -115 trained in 2.79 seconds.\n",
      "Test circle -114 trained in 2.78 seconds.\n",
      "Test circle -113 trained in 2.78 seconds.\n",
      "Test circle -112 trained in 2.77 seconds.\n",
      "Test circle -111 trained in 2.78 seconds.\n",
      "Test circle -110 trained in 2.79 seconds.\n",
      "Test circle -109 trained in 2.78 seconds.\n",
      "Test circle -108 trained in 2.8 seconds.\n",
      "Test circle -107 trained in 2.79 seconds.\n",
      "Test circle -106 trained in 2.91 seconds.\n",
      "Test circle -105 trained in 2.8 seconds.\n",
      "Test circle -104 trained in 2.79 seconds.\n",
      "Test circle -103 trained in 2.79 seconds.\n",
      "Test circle -102 trained in 2.78 seconds.\n",
      "Test circle -101 trained in 2.79 seconds.\n",
      "Test circle -100 trained in 2.79 seconds.\n",
      "Test circle -99 trained in 2.81 seconds.\n",
      "Test circle -98 trained in 2.79 seconds.\n",
      "Test circle -97 trained in 2.78 seconds.\n",
      "Test circle -96 trained in 2.79 seconds.\n",
      "Test circle -95 trained in 3.17 seconds.\n",
      "Test circle -94 trained in 2.89 seconds.\n",
      "Test circle -93 trained in 2.78 seconds.\n",
      "Test circle -92 trained in 2.79 seconds.\n",
      "Test circle -91 trained in 2.78 seconds.\n",
      "Test circle -90 trained in 2.76 seconds.\n",
      "Test circle -89 trained in 2.8 seconds.\n",
      "Test circle -88 trained in 2.86 seconds.\n",
      "Test circle -87 trained in 2.78 seconds.\n",
      "Test circle -86 trained in 2.79 seconds.\n",
      "Test circle -85 trained in 2.81 seconds.\n",
      "Test circle -84 trained in 2.8 seconds.\n",
      "Test circle -83 trained in 2.79 seconds.\n",
      "Test circle -82 trained in 2.78 seconds.\n",
      "Test circle -81 trained in 2.77 seconds.\n",
      "Test circle -80 trained in 2.77 seconds.\n",
      "Test circle -79 trained in 2.79 seconds.\n",
      "Test circle -78 trained in 2.78 seconds.\n",
      "Test circle -77 trained in 2.77 seconds.\n",
      "Test circle -76 trained in 2.78 seconds.\n",
      "Test circle -75 trained in 2.76 seconds.\n",
      "Test circle -74 trained in 2.78 seconds.\n",
      "Test circle -73 trained in 2.79 seconds.\n",
      "Test circle -72 trained in 2.8 seconds.\n",
      "Test circle -71 trained in 2.78 seconds.\n",
      "Test circle -70 trained in 2.78 seconds.\n",
      "Test circle -69 trained in 2.85 seconds.\n",
      "Test circle -68 trained in 2.78 seconds.\n",
      "Test circle -67 trained in 2.78 seconds.\n",
      "Test circle -66 trained in 2.77 seconds.\n",
      "Test circle -65 trained in 2.78 seconds.\n",
      "Test circle -64 trained in 2.78 seconds.\n",
      "Test circle -63 trained in 2.83 seconds.\n",
      "Test circle -62 trained in 2.78 seconds.\n",
      "Test circle -61 trained in 2.78 seconds.\n",
      "Test circle -60 trained in 2.78 seconds.\n",
      "Test circle -59 trained in 2.77 seconds.\n",
      "Test circle -58 trained in 2.79 seconds.\n",
      "Test circle -57 trained in 2.77 seconds.\n",
      "Test circle -56 trained in 2.79 seconds.\n",
      "Test circle -55 trained in 2.78 seconds.\n",
      "Test circle -54 trained in 2.77 seconds.\n",
      "Test circle -53 trained in 2.78 seconds.\n",
      "Test circle -52 trained in 2.84 seconds.\n",
      "Test circle -51 trained in 2.82 seconds.\n",
      "Test circle -50 trained in 2.88 seconds.\n",
      "Test circle -49 trained in 2.79 seconds.\n",
      "Test circle -48 trained in 2.78 seconds.\n",
      "Test circle -47 trained in 2.77 seconds.\n",
      "Test circle -46 trained in 2.78 seconds.\n",
      "Test circle -45 trained in 2.77 seconds.\n",
      "Test circle -44 trained in 2.78 seconds.\n",
      "Test circle -43 trained in 2.77 seconds.\n",
      "Test circle -42 trained in 2.81 seconds.\n",
      "Test circle -41 trained in 2.79 seconds.\n",
      "Test circle -40 trained in 2.77 seconds.\n",
      "Test circle -39 trained in 2.77 seconds.\n",
      "Test circle -38 trained in 2.77 seconds.\n",
      "Test circle -37 trained in 2.81 seconds.\n",
      "Test circle -36 trained in 2.78 seconds.\n",
      "Test circle -35 trained in 2.77 seconds.\n",
      "Test circle -34 trained in 2.78 seconds.\n",
      "Test circle -33 trained in 2.77 seconds.\n",
      "Test circle -32 trained in 2.77 seconds.\n",
      "Test circle -31 trained in 2.91 seconds.\n",
      "Test circle -30 trained in 2.79 seconds.\n",
      "Test circle -29 trained in 2.79 seconds.\n",
      "Test circle -28 trained in 2.78 seconds.\n",
      "Test circle -27 trained in 2.77 seconds.\n",
      "Test circle -26 trained in 2.78 seconds.\n",
      "Test circle -25 trained in 2.78 seconds.\n",
      "Test circle -24 trained in 2.89 seconds.\n",
      "Test circle -23 trained in 2.77 seconds.\n",
      "Test circle -22 trained in 2.77 seconds.\n",
      "Test circle -21 trained in 2.77 seconds.\n",
      "Test circle -20 trained in 2.82 seconds.\n",
      "Test circle -19 trained in 2.78 seconds.\n",
      "Test circle -18 trained in 2.77 seconds.\n",
      "Test circle -17 trained in 2.78 seconds.\n",
      "Test circle -16 trained in 2.77 seconds.\n",
      "Test circle -15 trained in 2.77 seconds.\n",
      "Test circle -14 trained in 2.78 seconds.\n",
      "Test circle -13 trained in 2.85 seconds.\n",
      "Test circle -12 trained in 2.77 seconds.\n",
      "Test circle -11 trained in 3.11 seconds.\n",
      "Test circle -10 trained in 2.79 seconds.\n",
      "Test circle -9 trained in 2.9 seconds.\n",
      "Test circle -8 trained in 2.83 seconds.\n",
      "Test circle -7 trained in 2.84 seconds.\n",
      "Test circle -6 trained in 2.79 seconds.\n",
      "Test circle -5 trained in 2.77 seconds.\n",
      "Test circle -4 trained in 2.77 seconds.\n",
      "Test circle -3 trained in 2.78 seconds.\n",
      "Test circle -2 trained in 2.78 seconds.\n",
      "Test circle -1 trained in 2.77 seconds.\n",
      "Model trained in 2085.31 seconds.\n",
      "Hyperparams (50, 6, 6) trained in 2085.32 seconds.\n",
      "=========================================================================================\n",
      "Model (50, 6, 6) Ended\n",
      "=========================================================================================\n"
     ]
    }
   ],
   "source": [
    "random_forest_val_results = []\n",
    "time_to_train_RF = {}\n",
    "\n",
    "for i in RF_params:\n",
    "    total_time = time.time()\n",
    "    \n",
    "    print('=========================================================================================')\n",
    "    print(f'Model {i} starts')\n",
    "    print('=========================================================================================')\n",
    "    direction = model_validation.rolling_window(model_validation.tech_val, \n",
    "                                            model_validation.binary_val, \n",
    "                                            model_validation.window, \n",
    "                                            model_validation.val_size, \n",
    "                                            model_validation.returns_val,\n",
    "                                            val_model='RF',\n",
    "                                            RF_params=i,\n",
    "                                            features=model_validation.feat_val)\n",
    "    \n",
    "    random_forest_val_results.append(direction)\n",
    "    \n",
    "    elapsed_time = round(time.time() - total_time, 2)\n",
    "    print(f\"Hyperparams {i} trained in {elapsed_time} seconds.\")    \n",
    "    time_to_train_RF[i] = f'{elapsed_time} seconds'\n",
    "    print('=========================================================================================')\n",
    "    print(f'Model {i} Ended')\n",
    "    print('=========================================================================================')\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdcd107c",
   "metadata": {},
   "source": [
    "# GARCH Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "64904256",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Random forest params\n",
    "GARCH_params = []\n",
    "\n",
    "# param_grid_garch = {\n",
    "#     'p': [1, 2, 3, 4],\n",
    "#     'q': [1, 2, 3, 4]\n",
    "# }\n",
    "param_grid_garch = {\n",
    "    'p': [1],\n",
    "    'q': [1]\n",
    "}\n",
    "for i in itertools.product(param_grid_garch['p'],\n",
    "                           param_grid_garch['q']):\n",
    "    \n",
    "    GARCH_params.append(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "115d51ab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "=========================================================================================\n",
      "Model (1, 1) starts\n",
      "=========================================================================================\n",
      "Test circle -725 trained in 0.49 seconds.\n",
      "Test circle -724 trained in 0.47 seconds.\n",
      "Test circle -723 trained in 0.47 seconds.\n",
      "Test circle -722 trained in 0.46 seconds.\n",
      "Test circle -721 trained in 0.48 seconds.\n",
      "Test circle -720 trained in 0.47 seconds.\n",
      "Test circle -719 trained in 0.47 seconds.\n",
      "Test circle -718 trained in 0.48 seconds.\n",
      "Test circle -717 trained in 0.47 seconds.\n",
      "Test circle -716 trained in 0.47 seconds.\n",
      "Test circle -715 trained in 0.46 seconds.\n",
      "Test circle -714 trained in 0.45 seconds.\n",
      "Test circle -713 trained in 0.47 seconds.\n",
      "Test circle -712 trained in 0.47 seconds.\n",
      "Test circle -711 trained in 0.47 seconds.\n",
      "Test circle -710 trained in 0.48 seconds.\n",
      "Test circle -709 trained in 0.44 seconds.\n",
      "Test circle -708 trained in 0.45 seconds.\n",
      "Test circle -707 trained in 0.45 seconds.\n",
      "Test circle -706 trained in 0.46 seconds.\n",
      "Test circle -705 trained in 0.47 seconds.\n",
      "Test circle -704 trained in 0.44 seconds.\n",
      "Test circle -703 trained in 0.47 seconds.\n",
      "Test circle -702 trained in 0.47 seconds.\n",
      "Test circle -701 trained in 0.44 seconds.\n",
      "Test circle -700 trained in 0.45 seconds.\n",
      "Test circle -699 trained in 0.47 seconds.\n",
      "Test circle -698 trained in 0.44 seconds.\n",
      "Test circle -697 trained in 0.45 seconds.\n",
      "Test circle -696 trained in 0.46 seconds.\n",
      "Test circle -695 trained in 0.45 seconds.\n",
      "Test circle -694 trained in 0.43 seconds.\n",
      "Test circle -693 trained in 0.45 seconds.\n",
      "Test circle -692 trained in 0.43 seconds.\n",
      "Test circle -691 trained in 0.43 seconds.\n",
      "Test circle -690 trained in 0.45 seconds.\n",
      "Test circle -689 trained in 0.44 seconds.\n",
      "Test circle -688 trained in 0.45 seconds.\n",
      "Test circle -687 trained in 0.43 seconds.\n",
      "Test circle -686 trained in 0.45 seconds.\n",
      "Test circle -685 trained in 0.45 seconds.\n",
      "Test circle -684 trained in 0.43 seconds.\n",
      "Test circle -683 trained in 0.48 seconds.\n",
      "Test circle -682 trained in 0.44 seconds.\n",
      "Test circle -681 trained in 0.45 seconds.\n",
      "Test circle -680 trained in 0.47 seconds.\n",
      "Test circle -679 trained in 0.45 seconds.\n",
      "Test circle -678 trained in 0.46 seconds.\n",
      "Test circle -677 trained in 0.44 seconds.\n",
      "Test circle -676 trained in 0.47 seconds.\n",
      "Test circle -675 trained in 0.47 seconds.\n",
      "Test circle -674 trained in 0.5 seconds.\n",
      "Test circle -673 trained in 0.5 seconds.\n",
      "Test circle -672 trained in 0.48 seconds.\n",
      "Test circle -671 trained in 0.49 seconds.\n",
      "Test circle -670 trained in 0.5 seconds.\n",
      "Test circle -669 trained in 0.47 seconds.\n",
      "Test circle -668 trained in 0.47 seconds.\n",
      "Test circle -667 trained in 0.46 seconds.\n",
      "Test circle -666 trained in 0.45 seconds.\n",
      "Test circle -665 trained in 0.45 seconds.\n",
      "Test circle -664 trained in 0.43 seconds.\n",
      "Test circle -663 trained in 0.46 seconds.\n",
      "Test circle -662 trained in 0.46 seconds.\n",
      "Test circle -661 trained in 0.46 seconds.\n",
      "Test circle -660 trained in 0.46 seconds.\n",
      "Test circle -659 trained in 0.47 seconds.\n",
      "Test circle -658 trained in 0.46 seconds.\n",
      "Test circle -657 trained in 0.45 seconds.\n",
      "Test circle -656 trained in 0.47 seconds.\n",
      "Test circle -655 trained in 0.47 seconds.\n",
      "Test circle -654 trained in 0.47 seconds.\n",
      "Test circle -653 trained in 0.47 seconds.\n",
      "Test circle -652 trained in 0.47 seconds.\n",
      "Test circle -651 trained in 0.48 seconds.\n",
      "Test circle -650 trained in 0.43 seconds.\n",
      "Test circle -649 trained in 0.45 seconds.\n",
      "Test circle -648 trained in 0.44 seconds.\n",
      "Test circle -647 trained in 0.45 seconds.\n",
      "Test circle -646 trained in 0.45 seconds.\n",
      "Test circle -645 trained in 0.45 seconds.\n",
      "Test circle -644 trained in 0.48 seconds.\n",
      "Test circle -643 trained in 0.47 seconds.\n",
      "Test circle -642 trained in 0.45 seconds.\n",
      "Test circle -641 trained in 0.45 seconds.\n",
      "Test circle -640 trained in 0.45 seconds.\n",
      "Test circle -639 trained in 0.45 seconds.\n",
      "Test circle -638 trained in 0.45 seconds.\n",
      "Test circle -637 trained in 0.44 seconds.\n",
      "Test circle -636 trained in 0.46 seconds.\n",
      "Test circle -635 trained in 0.45 seconds.\n",
      "Test circle -634 trained in 0.45 seconds.\n",
      "Test circle -633 trained in 0.45 seconds.\n",
      "Test circle -632 trained in 0.46 seconds.\n",
      "Test circle -631 trained in 0.45 seconds.\n",
      "Test circle -630 trained in 0.45 seconds.\n",
      "Test circle -629 trained in 0.48 seconds.\n",
      "Test circle -628 trained in 0.47 seconds.\n",
      "Test circle -627 trained in 0.47 seconds.\n",
      "Test circle -626 trained in 0.45 seconds.\n",
      "Test circle -625 trained in 0.43 seconds.\n",
      "Test circle -624 trained in 0.45 seconds.\n",
      "Test circle -623 trained in 0.44 seconds.\n",
      "Test circle -622 trained in 0.46 seconds.\n",
      "Test circle -621 trained in 0.45 seconds.\n",
      "Test circle -620 trained in 0.45 seconds.\n",
      "Test circle -619 trained in 0.45 seconds.\n",
      "Test circle -618 trained in 0.45 seconds.\n",
      "Test circle -617 trained in 0.45 seconds.\n",
      "Test circle -616 trained in 0.46 seconds.\n",
      "Test circle -615 trained in 0.44 seconds.\n",
      "Test circle -614 trained in 0.47 seconds.\n",
      "Test circle -613 trained in 0.45 seconds.\n",
      "Test circle -612 trained in 0.47 seconds.\n",
      "Test circle -611 trained in 0.45 seconds.\n",
      "Test circle -610 trained in 0.47 seconds.\n",
      "Test circle -609 trained in 0.45 seconds.\n",
      "Test circle -608 trained in 0.45 seconds.\n",
      "Test circle -607 trained in 0.45 seconds.\n",
      "Test circle -606 trained in 0.44 seconds.\n",
      "Test circle -605 trained in 0.43 seconds.\n",
      "Test circle -604 trained in 0.45 seconds.\n",
      "Test circle -603 trained in 0.43 seconds.\n",
      "Test circle -602 trained in 0.45 seconds.\n",
      "Test circle -601 trained in 0.45 seconds.\n",
      "Test circle -600 trained in 0.47 seconds.\n",
      "Test circle -599 trained in 0.45 seconds.\n",
      "Test circle -598 trained in 0.46 seconds.\n",
      "Test circle -597 trained in 0.45 seconds.\n",
      "Test circle -596 trained in 0.45 seconds.\n",
      "Test circle -595 trained in 0.47 seconds.\n",
      "Test circle -594 trained in 0.48 seconds.\n",
      "Test circle -593 trained in 0.45 seconds.\n",
      "Test circle -592 trained in 0.47 seconds.\n",
      "Test circle -591 trained in 0.45 seconds.\n",
      "Test circle -590 trained in 0.48 seconds.\n",
      "Test circle -589 trained in 0.49 seconds.\n",
      "Test circle -588 trained in 0.5 seconds.\n",
      "Test circle -587 trained in 0.5 seconds.\n",
      "Test circle -586 trained in 0.48 seconds.\n",
      "Test circle -585 trained in 0.48 seconds.\n",
      "Test circle -584 trained in 0.49 seconds.\n",
      "Test circle -583 trained in 0.49 seconds.\n",
      "Test circle -582 trained in 0.51 seconds.\n",
      "Test circle -581 trained in 0.53 seconds.\n",
      "Test circle -580 trained in 0.53 seconds.\n",
      "Test circle -579 trained in 0.51 seconds.\n",
      "Test circle -578 trained in 0.47 seconds.\n",
      "Test circle -577 trained in 0.5 seconds.\n",
      "Test circle -576 trained in 0.57 seconds.\n",
      "Test circle -575 trained in 0.55 seconds.\n",
      "Test circle -574 trained in 0.55 seconds.\n",
      "Test circle -573 trained in 0.53 seconds.\n",
      "Test circle -572 trained in 0.52 seconds.\n",
      "Test circle -571 trained in 0.54 seconds.\n",
      "Test circle -570 trained in 0.54 seconds.\n",
      "Test circle -569 trained in 0.52 seconds.\n",
      "Test circle -568 trained in 0.55 seconds.\n",
      "Test circle -567 trained in 0.55 seconds.\n",
      "Test circle -566 trained in 0.57 seconds.\n",
      "Test circle -565 trained in 0.55 seconds.\n",
      "Test circle -564 trained in 0.58 seconds.\n",
      "Test circle -563 trained in 0.57 seconds.\n",
      "Test circle -562 trained in 0.57 seconds.\n",
      "Test circle -561 trained in 0.57 seconds.\n",
      "Test circle -560 trained in 0.54 seconds.\n",
      "Test circle -559 trained in 0.55 seconds.\n",
      "Test circle -558 trained in 0.52 seconds.\n",
      "Test circle -557 trained in 0.52 seconds.\n",
      "Test circle -556 trained in 0.55 seconds.\n",
      "Test circle -555 trained in 0.57 seconds.\n",
      "Test circle -554 trained in 0.52 seconds.\n",
      "Test circle -553 trained in 0.53 seconds.\n",
      "Test circle -552 trained in 0.52 seconds.\n",
      "Test circle -551 trained in 0.55 seconds.\n",
      "Test circle -550 trained in 0.52 seconds.\n",
      "Test circle -549 trained in 0.55 seconds.\n",
      "Test circle -548 trained in 0.54 seconds.\n",
      "Test circle -547 trained in 0.52 seconds.\n",
      "Test circle -546 trained in 0.52 seconds.\n",
      "Test circle -545 trained in 0.53 seconds.\n",
      "Test circle -544 trained in 0.55 seconds.\n",
      "Test circle -543 trained in 0.55 seconds.\n",
      "Test circle -542 trained in 0.55 seconds.\n",
      "Test circle -541 trained in 0.54 seconds.\n",
      "Test circle -540 trained in 0.52 seconds.\n",
      "Test circle -539 trained in 0.52 seconds.\n",
      "Test circle -538 trained in 0.53 seconds.\n",
      "Test circle -537 trained in 0.52 seconds.\n",
      "Test circle -536 trained in 0.5 seconds.\n",
      "Test circle -535 trained in 0.53 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -534 trained in 0.48 seconds.\n",
      "Test circle -533 trained in 0.52 seconds.\n",
      "Test circle -532 trained in 0.52 seconds.\n",
      "Test circle -531 trained in 0.5 seconds.\n",
      "Test circle -530 trained in 0.47 seconds.\n",
      "Test circle -529 trained in 0.48 seconds.\n",
      "Test circle -528 trained in 0.55 seconds.\n",
      "Test circle -527 trained in 0.59 seconds.\n",
      "Test circle -526 trained in 0.56 seconds.\n",
      "Test circle -525 trained in 0.44 seconds.\n",
      "Test circle -524 trained in 0.46 seconds.\n",
      "Test circle -523 trained in 0.48 seconds.\n",
      "Test circle -522 trained in 0.49 seconds.\n",
      "Test circle -521 trained in 0.48 seconds.\n",
      "Test circle -520 trained in 0.48 seconds.\n",
      "Test circle -519 trained in 0.5 seconds.\n",
      "Test circle -518 trained in 0.47 seconds.\n",
      "Test circle -517 trained in 0.45 seconds.\n",
      "Test circle -516 trained in 0.5 seconds.\n",
      "Test circle -515 trained in 0.48 seconds.\n",
      "Test circle -514 trained in 0.48 seconds.\n",
      "Test circle -513 trained in 0.48 seconds.\n",
      "Test circle -512 trained in 0.47 seconds.\n",
      "Test circle -511 trained in 0.46 seconds.\n",
      "Test circle -510 trained in 0.45 seconds.\n",
      "Test circle -509 trained in 0.48 seconds.\n",
      "Test circle -508 trained in 0.5 seconds.\n",
      "Test circle -507 trained in 0.47 seconds.\n",
      "Test circle -506 trained in 0.48 seconds.\n",
      "Test circle -505 trained in 0.48 seconds.\n",
      "Test circle -504 trained in 0.47 seconds.\n",
      "Test circle -503 trained in 0.49 seconds.\n",
      "Test circle -502 trained in 0.49 seconds.\n",
      "Test circle -501 trained in 0.48 seconds.\n",
      "Test circle -500 trained in 0.49 seconds.\n",
      "Test circle -499 trained in 0.47 seconds.\n",
      "Test circle -498 trained in 0.47 seconds.\n",
      "Test circle -497 trained in 0.48 seconds.\n",
      "Test circle -496 trained in 0.48 seconds.\n",
      "Test circle -495 trained in 0.5 seconds.\n",
      "Test circle -494 trained in 0.48 seconds.\n",
      "Test circle -493 trained in 0.53 seconds.\n",
      "Test circle -492 trained in 0.54 seconds.\n",
      "Test circle -491 trained in 0.46 seconds.\n",
      "Test circle -490 trained in 0.44 seconds.\n",
      "Test circle -489 trained in 0.49 seconds.\n",
      "Test circle -488 trained in 0.47 seconds.\n",
      "Test circle -487 trained in 0.5 seconds.\n",
      "Test circle -486 trained in 0.5 seconds.\n",
      "Test circle -485 trained in 0.5 seconds.\n",
      "Test circle -484 trained in 0.47 seconds.\n",
      "Test circle -483 trained in 0.48 seconds.\n",
      "Test circle -482 trained in 0.45 seconds.\n",
      "Test circle -481 trained in 0.47 seconds.\n",
      "Test circle -480 trained in 0.48 seconds.\n",
      "Test circle -479 trained in 0.49 seconds.\n",
      "Test circle -478 trained in 0.48 seconds.\n",
      "Test circle -477 trained in 0.47 seconds.\n",
      "Test circle -476 trained in 0.46 seconds.\n",
      "Test circle -475 trained in 0.47 seconds.\n",
      "Test circle -474 trained in 0.45 seconds.\n",
      "Test circle -473 trained in 0.42 seconds.\n",
      "Test circle -472 trained in 0.46 seconds.\n",
      "Test circle -471 trained in 0.45 seconds.\n",
      "Test circle -470 trained in 0.47 seconds.\n",
      "Test circle -469 trained in 0.47 seconds.\n",
      "Test circle -468 trained in 0.51 seconds.\n",
      "Test circle -467 trained in 0.49 seconds.\n",
      "Test circle -466 trained in 0.48 seconds.\n",
      "Test circle -465 trained in 0.5 seconds.\n",
      "Test circle -464 trained in 0.47 seconds.\n",
      "Test circle -463 trained in 0.48 seconds.\n",
      "Test circle -462 trained in 0.49 seconds.\n",
      "Test circle -461 trained in 0.48 seconds.\n",
      "Test circle -460 trained in 0.48 seconds.\n",
      "Test circle -459 trained in 0.49 seconds.\n",
      "Test circle -458 trained in 0.46 seconds.\n",
      "Test circle -457 trained in 0.58 seconds.\n",
      "Test circle -456 trained in 0.53 seconds.\n",
      "Test circle -455 trained in 0.5 seconds.\n",
      "Test circle -454 trained in 0.49 seconds.\n",
      "Test circle -453 trained in 0.53 seconds.\n",
      "Test circle -452 trained in 0.54 seconds.\n",
      "Test circle -451 trained in 0.49 seconds.\n",
      "Test circle -450 trained in 0.48 seconds.\n",
      "Test circle -449 trained in 0.51 seconds.\n",
      "Test circle -448 trained in 0.46 seconds.\n",
      "Test circle -447 trained in 0.5 seconds.\n",
      "Test circle -446 trained in 0.46 seconds.\n",
      "Test circle -445 trained in 0.49 seconds.\n",
      "Test circle -444 trained in 0.47 seconds.\n",
      "Test circle -443 trained in 0.48 seconds.\n",
      "Test circle -442 trained in 0.54 seconds.\n",
      "Test circle -441 trained in 0.53 seconds.\n",
      "Test circle -440 trained in 0.58 seconds.\n",
      "Test circle -439 trained in 0.56 seconds.\n",
      "Test circle -438 trained in 0.55 seconds.\n",
      "Test circle -437 trained in 0.56 seconds.\n",
      "Test circle -436 trained in 0.58 seconds.\n",
      "Test circle -435 trained in 0.53 seconds.\n",
      "Test circle -434 trained in 0.55 seconds.\n",
      "Test circle -433 trained in 0.54 seconds.\n",
      "Test circle -432 trained in 0.59 seconds.\n",
      "Test circle -431 trained in 0.64 seconds.\n",
      "Test circle -430 trained in 0.58 seconds.\n",
      "Test circle -429 trained in 0.57 seconds.\n",
      "Test circle -428 trained in 0.58 seconds.\n",
      "Test circle -427 trained in 0.56 seconds.\n",
      "Test circle -426 trained in 0.56 seconds.\n",
      "Test circle -425 trained in 0.55 seconds.\n",
      "Test circle -424 trained in 0.56 seconds.\n",
      "Test circle -423 trained in 0.56 seconds.\n",
      "Test circle -422 trained in 0.63 seconds.\n",
      "Test circle -421 trained in 0.58 seconds.\n",
      "Test circle -420 trained in 0.64 seconds.\n",
      "Test circle -419 trained in 0.58 seconds.\n",
      "Test circle -418 trained in 0.6 seconds.\n",
      "Test circle -417 trained in 0.61 seconds.\n",
      "Test circle -416 trained in 0.56 seconds.\n",
      "Test circle -415 trained in 0.54 seconds.\n",
      "Test circle -414 trained in 0.53 seconds.\n",
      "Test circle -413 trained in 0.53 seconds.\n",
      "Test circle -412 trained in 0.53 seconds.\n",
      "Test circle -411 trained in 0.54 seconds.\n",
      "Test circle -410 trained in 0.52 seconds.\n",
      "Test circle -409 trained in 0.54 seconds.\n",
      "Test circle -408 trained in 0.54 seconds.\n",
      "Test circle -407 trained in 0.49 seconds.\n",
      "Test circle -406 trained in 0.53 seconds.\n",
      "Test circle -405 trained in 0.49 seconds.\n",
      "Test circle -404 trained in 0.57 seconds.\n",
      "Test circle -403 trained in 0.52 seconds.\n",
      "Test circle -402 trained in 0.51 seconds.\n",
      "Test circle -401 trained in 0.51 seconds.\n",
      "Test circle -400 trained in 0.53 seconds.\n",
      "Test circle -399 trained in 0.5 seconds.\n",
      "Test circle -398 trained in 0.54 seconds.\n",
      "Test circle -397 trained in 0.5 seconds.\n",
      "Test circle -396 trained in 0.49 seconds.\n",
      "Test circle -395 trained in 0.51 seconds.\n",
      "Test circle -394 trained in 0.49 seconds.\n",
      "Test circle -393 trained in 0.5 seconds.\n",
      "Test circle -392 trained in 0.52 seconds.\n",
      "Test circle -391 trained in 0.52 seconds.\n",
      "Test circle -390 trained in 0.81 seconds.\n",
      "Test circle -389 trained in 0.52 seconds.\n",
      "Test circle -388 trained in 0.52 seconds.\n",
      "Test circle -387 trained in 0.53 seconds.\n",
      "Test circle -386 trained in 0.52 seconds.\n",
      "Test circle -385 trained in 0.53 seconds.\n",
      "Test circle -384 trained in 0.53 seconds.\n",
      "Test circle -383 trained in 0.49 seconds.\n",
      "Test circle -382 trained in 0.48 seconds.\n",
      "Test circle -381 trained in 0.47 seconds.\n",
      "Test circle -380 trained in 0.51 seconds.\n",
      "Test circle -379 trained in 0.57 seconds.\n",
      "Test circle -378 trained in 0.55 seconds.\n",
      "Test circle -377 trained in 0.53 seconds.\n",
      "Test circle -376 trained in 0.55 seconds.\n",
      "Test circle -375 trained in 0.52 seconds.\n",
      "Test circle -374 trained in 0.55 seconds.\n",
      "Test circle -373 trained in 0.5 seconds.\n",
      "Test circle -372 trained in 0.52 seconds.\n",
      "Test circle -371 trained in 0.54 seconds.\n",
      "Test circle -370 trained in 0.54 seconds.\n",
      "Test circle -369 trained in 0.5 seconds.\n",
      "Test circle -368 trained in 0.5 seconds.\n",
      "Test circle -367 trained in 0.52 seconds.\n",
      "Test circle -366 trained in 0.51 seconds.\n",
      "Test circle -365 trained in 0.52 seconds.\n",
      "Test circle -364 trained in 0.52 seconds.\n",
      "Test circle -363 trained in 0.61 seconds.\n",
      "Test circle -362 trained in 0.56 seconds.\n",
      "Test circle -361 trained in 0.6 seconds.\n",
      "Test circle -360 trained in 0.61 seconds.\n",
      "Test circle -359 trained in 0.52 seconds.\n",
      "Test circle -358 trained in 0.52 seconds.\n",
      "Test circle -357 trained in 0.48 seconds.\n",
      "Test circle -356 trained in 0.48 seconds.\n",
      "Test circle -355 trained in 0.52 seconds.\n",
      "Test circle -354 trained in 0.49 seconds.\n",
      "Test circle -353 trained in 0.5 seconds.\n",
      "Test circle -352 trained in 0.5 seconds.\n",
      "Test circle -351 trained in 0.49 seconds.\n",
      "Test circle -350 trained in 0.5 seconds.\n",
      "Test circle -349 trained in 0.5 seconds.\n",
      "Test circle -348 trained in 0.5 seconds.\n",
      "Test circle -347 trained in 0.5 seconds.\n",
      "Test circle -346 trained in 0.49 seconds.\n",
      "Test circle -345 trained in 0.5 seconds.\n",
      "Test circle -344 trained in 0.48 seconds.\n",
      "Test circle -343 trained in 0.47 seconds.\n",
      "Test circle -342 trained in 0.48 seconds.\n",
      "Test circle -341 trained in 0.48 seconds.\n",
      "Test circle -340 trained in 0.47 seconds.\n",
      "Test circle -339 trained in 0.5 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -338 trained in 0.52 seconds.\n",
      "Test circle -337 trained in 0.48 seconds.\n",
      "Test circle -336 trained in 0.5 seconds.\n",
      "Test circle -335 trained in 0.47 seconds.\n",
      "Test circle -334 trained in 0.5 seconds.\n",
      "Test circle -333 trained in 0.49 seconds.\n",
      "Test circle -332 trained in 0.52 seconds.\n",
      "Test circle -331 trained in 0.49 seconds.\n",
      "Test circle -330 trained in 0.51 seconds.\n",
      "Test circle -329 trained in 0.48 seconds.\n",
      "Test circle -328 trained in 0.48 seconds.\n",
      "Test circle -327 trained in 0.52 seconds.\n",
      "Test circle -326 trained in 0.53 seconds.\n",
      "Test circle -325 trained in 0.52 seconds.\n",
      "Test circle -324 trained in 0.5 seconds.\n",
      "Test circle -323 trained in 0.48 seconds.\n",
      "Test circle -322 trained in 0.48 seconds.\n",
      "Test circle -321 trained in 0.53 seconds.\n",
      "Test circle -320 trained in 0.5 seconds.\n",
      "Test circle -319 trained in 0.5 seconds.\n",
      "Test circle -318 trained in 0.5 seconds.\n",
      "Test circle -317 trained in 0.53 seconds.\n",
      "Test circle -316 trained in 0.49 seconds.\n",
      "Test circle -315 trained in 0.5 seconds.\n",
      "Test circle -314 trained in 0.52 seconds.\n",
      "Test circle -313 trained in 0.54 seconds.\n",
      "Test circle -312 trained in 0.5 seconds.\n",
      "Test circle -311 trained in 0.45 seconds.\n",
      "Test circle -310 trained in 0.5 seconds.\n",
      "Test circle -309 trained in 0.49 seconds.\n",
      "Test circle -308 trained in 0.52 seconds.\n",
      "Test circle -307 trained in 0.48 seconds.\n",
      "Test circle -306 trained in 0.5 seconds.\n",
      "Test circle -305 trained in 0.52 seconds.\n",
      "Test circle -304 trained in 0.53 seconds.\n",
      "Test circle -303 trained in 0.47 seconds.\n",
      "Test circle -302 trained in 0.48 seconds.\n",
      "Test circle -301 trained in 0.48 seconds.\n",
      "Test circle -300 trained in 0.45 seconds.\n",
      "Test circle -299 trained in 0.5 seconds.\n",
      "Test circle -298 trained in 0.48 seconds.\n",
      "Test circle -297 trained in 0.48 seconds.\n",
      "Test circle -296 trained in 0.48 seconds.\n",
      "Test circle -295 trained in 0.47 seconds.\n",
      "Test circle -294 trained in 0.5 seconds.\n",
      "Test circle -293 trained in 0.47 seconds.\n",
      "Test circle -292 trained in 0.49 seconds.\n",
      "Test circle -291 trained in 0.48 seconds.\n",
      "Test circle -290 trained in 0.48 seconds.\n",
      "Test circle -289 trained in 0.45 seconds.\n",
      "Test circle -288 trained in 0.44 seconds.\n",
      "Test circle -287 trained in 0.44 seconds.\n",
      "Test circle -286 trained in 0.49 seconds.\n",
      "Test circle -285 trained in 0.47 seconds.\n",
      "Test circle -284 trained in 0.48 seconds.\n",
      "Test circle -283 trained in 0.56 seconds.\n",
      "Test circle -282 trained in 0.52 seconds.\n",
      "Test circle -281 trained in 0.49 seconds.\n",
      "Test circle -280 trained in 0.48 seconds.\n",
      "Test circle -279 trained in 0.51 seconds.\n",
      "Test circle -278 trained in 0.5 seconds.\n",
      "Test circle -277 trained in 0.51 seconds.\n",
      "Test circle -276 trained in 0.52 seconds.\n",
      "Test circle -275 trained in 0.5 seconds.\n",
      "Test circle -274 trained in 0.49 seconds.\n",
      "Test circle -273 trained in 0.51 seconds.\n",
      "Test circle -272 trained in 0.51 seconds.\n",
      "Test circle -271 trained in 0.48 seconds.\n",
      "Test circle -270 trained in 0.49 seconds.\n",
      "Test circle -269 trained in 0.5 seconds.\n",
      "Test circle -268 trained in 0.51 seconds.\n",
      "Test circle -267 trained in 0.51 seconds.\n",
      "Test circle -266 trained in 0.47 seconds.\n",
      "Test circle -265 trained in 0.5 seconds.\n",
      "Test circle -264 trained in 0.49 seconds.\n",
      "Test circle -263 trained in 0.49 seconds.\n",
      "Test circle -262 trained in 0.53 seconds.\n",
      "Test circle -261 trained in 0.54 seconds.\n",
      "Test circle -260 trained in 0.5 seconds.\n",
      "Test circle -259 trained in 0.52 seconds.\n",
      "Test circle -258 trained in 0.52 seconds.\n",
      "Test circle -257 trained in 0.51 seconds.\n",
      "Test circle -256 trained in 0.51 seconds.\n",
      "Test circle -255 trained in 0.46 seconds.\n",
      "Test circle -254 trained in 0.47 seconds.\n",
      "Test circle -253 trained in 0.48 seconds.\n",
      "Test circle -252 trained in 0.48 seconds.\n",
      "Test circle -251 trained in 0.47 seconds.\n",
      "Test circle -250 trained in 0.49 seconds.\n",
      "Test circle -249 trained in 0.49 seconds.\n",
      "Test circle -248 trained in 0.47 seconds.\n",
      "Test circle -247 trained in 0.47 seconds.\n",
      "Test circle -246 trained in 0.47 seconds.\n",
      "Test circle -245 trained in 0.45 seconds.\n",
      "Test circle -244 trained in 0.47 seconds.\n",
      "Test circle -243 trained in 0.47 seconds.\n",
      "Test circle -242 trained in 0.45 seconds.\n",
      "Test circle -241 trained in 0.45 seconds.\n",
      "Test circle -240 trained in 0.45 seconds.\n",
      "Test circle -239 trained in 0.48 seconds.\n",
      "Test circle -238 trained in 0.47 seconds.\n",
      "Test circle -237 trained in 0.47 seconds.\n",
      "Test circle -236 trained in 0.48 seconds.\n",
      "Test circle -235 trained in 0.5 seconds.\n",
      "Test circle -234 trained in 0.45 seconds.\n",
      "Test circle -233 trained in 0.49 seconds.\n",
      "Test circle -232 trained in 0.48 seconds.\n",
      "Test circle -231 trained in 0.47 seconds.\n",
      "Test circle -230 trained in 0.46 seconds.\n",
      "Test circle -229 trained in 0.47 seconds.\n",
      "Test circle -228 trained in 0.48 seconds.\n",
      "Test circle -227 trained in 0.47 seconds.\n",
      "Test circle -226 trained in 0.47 seconds.\n",
      "Test circle -225 trained in 0.5 seconds.\n",
      "Test circle -224 trained in 0.49 seconds.\n",
      "Test circle -223 trained in 0.5 seconds.\n",
      "Test circle -222 trained in 0.48 seconds.\n",
      "Test circle -221 trained in 0.45 seconds.\n",
      "Test circle -220 trained in 0.5 seconds.\n",
      "Test circle -219 trained in 0.48 seconds.\n",
      "Test circle -218 trained in 0.49 seconds.\n",
      "Test circle -217 trained in 0.5 seconds.\n",
      "Test circle -216 trained in 0.5 seconds.\n",
      "Test circle -215 trained in 0.47 seconds.\n",
      "Test circle -214 trained in 0.49 seconds.\n",
      "Test circle -213 trained in 0.5 seconds.\n",
      "Test circle -212 trained in 0.49 seconds.\n",
      "Test circle -211 trained in 0.48 seconds.\n",
      "Test circle -210 trained in 0.5 seconds.\n",
      "Test circle -209 trained in 0.47 seconds.\n",
      "Test circle -208 trained in 0.51 seconds.\n",
      "Test circle -207 trained in 0.49 seconds.\n",
      "Test circle -206 trained in 0.5 seconds.\n",
      "Test circle -205 trained in 0.52 seconds.\n",
      "Test circle -204 trained in 0.51 seconds.\n",
      "Test circle -203 trained in 0.48 seconds.\n",
      "Test circle -202 trained in 0.49 seconds.\n",
      "Test circle -201 trained in 0.48 seconds.\n",
      "Test circle -200 trained in 0.49 seconds.\n",
      "Test circle -199 trained in 0.47 seconds.\n",
      "Test circle -198 trained in 0.48 seconds.\n",
      "Test circle -197 trained in 0.48 seconds.\n",
      "Test circle -196 trained in 0.5 seconds.\n",
      "Test circle -195 trained in 0.47 seconds.\n",
      "Test circle -194 trained in 0.48 seconds.\n",
      "Test circle -193 trained in 0.47 seconds.\n",
      "Test circle -192 trained in 0.49 seconds.\n",
      "Test circle -191 trained in 0.47 seconds.\n",
      "Test circle -190 trained in 0.48 seconds.\n",
      "Test circle -189 trained in 0.45 seconds.\n",
      "Test circle -188 trained in 0.47 seconds.\n",
      "Test circle -187 trained in 0.46 seconds.\n",
      "Test circle -186 trained in 0.47 seconds.\n",
      "Test circle -185 trained in 0.47 seconds.\n",
      "Test circle -184 trained in 0.45 seconds.\n",
      "Test circle -183 trained in 0.47 seconds.\n",
      "Test circle -182 trained in 0.45 seconds.\n",
      "Test circle -181 trained in 0.49 seconds.\n",
      "Test circle -180 trained in 0.47 seconds.\n",
      "Test circle -179 trained in 0.45 seconds.\n",
      "Test circle -178 trained in 0.47 seconds.\n",
      "Test circle -177 trained in 0.47 seconds.\n",
      "Test circle -176 trained in 0.47 seconds.\n",
      "Test circle -175 trained in 0.49 seconds.\n",
      "Test circle -174 trained in 0.54 seconds.\n",
      "Test circle -173 trained in 0.59 seconds.\n",
      "Test circle -172 trained in 0.47 seconds.\n",
      "Test circle -171 trained in 0.47 seconds.\n",
      "Test circle -170 trained in 0.49 seconds.\n",
      "Test circle -169 trained in 0.47 seconds.\n",
      "Test circle -168 trained in 0.49 seconds.\n",
      "Test circle -167 trained in 0.45 seconds.\n",
      "Test circle -166 trained in 0.47 seconds.\n",
      "Test circle -165 trained in 0.45 seconds.\n",
      "Test circle -164 trained in 0.5 seconds.\n",
      "Test circle -163 trained in 0.52 seconds.\n",
      "Test circle -162 trained in 0.47 seconds.\n",
      "Test circle -161 trained in 0.51 seconds.\n",
      "Test circle -160 trained in 0.49 seconds.\n",
      "Test circle -159 trained in 0.49 seconds.\n",
      "Test circle -158 trained in 0.45 seconds.\n",
      "Test circle -157 trained in 0.47 seconds.\n",
      "Test circle -156 trained in 0.45 seconds.\n",
      "Test circle -155 trained in 0.45 seconds.\n",
      "Test circle -154 trained in 0.46 seconds.\n",
      "Test circle -153 trained in 0.47 seconds.\n",
      "Test circle -152 trained in 0.46 seconds.\n",
      "Test circle -151 trained in 0.48 seconds.\n",
      "Test circle -150 trained in 0.45 seconds.\n",
      "Test circle -149 trained in 0.49 seconds.\n",
      "Test circle -148 trained in 0.45 seconds.\n",
      "Test circle -147 trained in 0.49 seconds.\n",
      "Test circle -146 trained in 0.47 seconds.\n",
      "Test circle -145 trained in 0.48 seconds.\n",
      "Test circle -144 trained in 0.47 seconds.\n",
      "Test circle -143 trained in 0.46 seconds.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test circle -142 trained in 0.42 seconds.\n",
      "Test circle -141 trained in 0.44 seconds.\n",
      "Test circle -140 trained in 0.43 seconds.\n",
      "Test circle -139 trained in 0.47 seconds.\n",
      "Test circle -138 trained in 0.48 seconds.\n",
      "Test circle -137 trained in 0.45 seconds.\n",
      "Test circle -136 trained in 0.45 seconds.\n",
      "Test circle -135 trained in 0.44 seconds.\n",
      "Test circle -134 trained in 0.45 seconds.\n",
      "Test circle -133 trained in 0.44 seconds.\n",
      "Test circle -132 trained in 0.44 seconds.\n",
      "Test circle -131 trained in 0.46 seconds.\n",
      "Test circle -130 trained in 0.47 seconds.\n",
      "Test circle -129 trained in 0.49 seconds.\n",
      "Test circle -128 trained in 0.47 seconds.\n",
      "Test circle -127 trained in 0.45 seconds.\n",
      "Test circle -126 trained in 0.46 seconds.\n",
      "Test circle -125 trained in 0.51 seconds.\n",
      "Test circle -124 trained in 0.47 seconds.\n",
      "Test circle -123 trained in 0.53 seconds.\n",
      "Test circle -122 trained in 0.51 seconds.\n",
      "Test circle -121 trained in 0.46 seconds.\n",
      "Test circle -120 trained in 0.46 seconds.\n",
      "Test circle -119 trained in 0.43 seconds.\n",
      "Test circle -118 trained in 0.44 seconds.\n",
      "Test circle -117 trained in 0.48 seconds.\n",
      "Test circle -116 trained in 0.47 seconds.\n",
      "Test circle -115 trained in 0.46 seconds.\n",
      "Test circle -114 trained in 0.48 seconds.\n",
      "Test circle -113 trained in 0.45 seconds.\n",
      "Test circle -112 trained in 0.46 seconds.\n",
      "Test circle -111 trained in 0.49 seconds.\n",
      "Test circle -110 trained in 0.46 seconds.\n",
      "Test circle -109 trained in 0.44 seconds.\n",
      "Test circle -108 trained in 0.45 seconds.\n",
      "Test circle -107 trained in 0.47 seconds.\n",
      "Test circle -106 trained in 0.47 seconds.\n",
      "Test circle -105 trained in 0.46 seconds.\n",
      "Test circle -104 trained in 0.47 seconds.\n",
      "Test circle -103 trained in 0.47 seconds.\n",
      "Test circle -102 trained in 0.5 seconds.\n",
      "Test circle -101 trained in 0.5 seconds.\n",
      "Test circle -100 trained in 0.48 seconds.\n",
      "Test circle -99 trained in 0.51 seconds.\n",
      "Test circle -98 trained in 0.47 seconds.\n",
      "Test circle -97 trained in 0.47 seconds.\n",
      "Test circle -96 trained in 0.47 seconds.\n",
      "Test circle -95 trained in 0.5 seconds.\n",
      "Test circle -94 trained in 0.48 seconds.\n",
      "Test circle -93 trained in 0.51 seconds.\n",
      "Test circle -92 trained in 0.48 seconds.\n",
      "Test circle -91 trained in 0.47 seconds.\n",
      "Test circle -90 trained in 0.47 seconds.\n",
      "Test circle -89 trained in 0.44 seconds.\n",
      "Test circle -88 trained in 0.44 seconds.\n",
      "Test circle -87 trained in 0.47 seconds.\n",
      "Test circle -86 trained in 0.48 seconds.\n",
      "Test circle -85 trained in 0.49 seconds.\n",
      "Test circle -84 trained in 0.45 seconds.\n",
      "Test circle -83 trained in 0.45 seconds.\n",
      "Test circle -82 trained in 0.45 seconds.\n",
      "Test circle -81 trained in 0.45 seconds.\n",
      "Test circle -80 trained in 0.47 seconds.\n",
      "Test circle -79 trained in 0.45 seconds.\n",
      "Test circle -78 trained in 0.45 seconds.\n",
      "Test circle -77 trained in 0.46 seconds.\n",
      "Test circle -76 trained in 0.46 seconds.\n",
      "Test circle -75 trained in 0.49 seconds.\n",
      "Test circle -74 trained in 0.42 seconds.\n",
      "Test circle -73 trained in 0.44 seconds.\n",
      "Test circle -72 trained in 0.46 seconds.\n",
      "Test circle -71 trained in 0.48 seconds.\n",
      "Test circle -70 trained in 0.44 seconds.\n",
      "Test circle -69 trained in 0.46 seconds.\n",
      "Test circle -68 trained in 0.43 seconds.\n",
      "Test circle -67 trained in 0.44 seconds.\n",
      "Test circle -66 trained in 0.45 seconds.\n",
      "Test circle -65 trained in 0.43 seconds.\n",
      "Test circle -64 trained in 0.5 seconds.\n",
      "Test circle -63 trained in 0.47 seconds.\n",
      "Test circle -62 trained in 0.44 seconds.\n",
      "Test circle -61 trained in 0.49 seconds.\n",
      "Test circle -60 trained in 0.51 seconds.\n",
      "Test circle -59 trained in 0.49 seconds.\n",
      "Test circle -58 trained in 0.5 seconds.\n",
      "Test circle -57 trained in 0.47 seconds.\n",
      "Test circle -56 trained in 0.47 seconds.\n",
      "Test circle -55 trained in 0.52 seconds.\n",
      "Test circle -54 trained in 0.48 seconds.\n",
      "Test circle -53 trained in 0.46 seconds.\n",
      "Test circle -52 trained in 0.48 seconds.\n",
      "Test circle -51 trained in 0.48 seconds.\n",
      "Test circle -50 trained in 0.46 seconds.\n",
      "Test circle -49 trained in 0.45 seconds.\n",
      "Test circle -48 trained in 0.47 seconds.\n",
      "Test circle -47 trained in 0.47 seconds.\n",
      "Test circle -46 trained in 0.45 seconds.\n",
      "Test circle -45 trained in 0.48 seconds.\n",
      "Test circle -44 trained in 0.47 seconds.\n",
      "Test circle -43 trained in 0.51 seconds.\n",
      "Test circle -42 trained in 0.49 seconds.\n",
      "Test circle -41 trained in 0.5 seconds.\n",
      "Test circle -40 trained in 0.47 seconds.\n",
      "Test circle -39 trained in 0.45 seconds.\n",
      "Test circle -38 trained in 0.5 seconds.\n",
      "Test circle -37 trained in 0.49 seconds.\n",
      "Test circle -36 trained in 0.48 seconds.\n",
      "Test circle -35 trained in 0.49 seconds.\n",
      "Test circle -34 trained in 0.5 seconds.\n",
      "Test circle -33 trained in 0.5 seconds.\n",
      "Test circle -32 trained in 0.48 seconds.\n",
      "Test circle -31 trained in 0.49 seconds.\n",
      "Test circle -30 trained in 0.49 seconds.\n",
      "Test circle -29 trained in 0.48 seconds.\n",
      "Test circle -28 trained in 0.48 seconds.\n",
      "Test circle -27 trained in 0.45 seconds.\n",
      "Test circle -26 trained in 0.5 seconds.\n",
      "Test circle -25 trained in 0.47 seconds.\n",
      "Test circle -24 trained in 0.45 seconds.\n",
      "Test circle -23 trained in 0.48 seconds.\n",
      "Test circle -22 trained in 0.5 seconds.\n",
      "Test circle -21 trained in 0.49 seconds.\n",
      "Test circle -20 trained in 0.47 seconds.\n",
      "Test circle -19 trained in 0.5 seconds.\n",
      "Test circle -18 trained in 0.5 seconds.\n",
      "Test circle -17 trained in 0.5 seconds.\n",
      "Test circle -16 trained in 0.48 seconds.\n",
      "Test circle -15 trained in 0.5 seconds.\n",
      "Test circle -14 trained in 0.5 seconds.\n",
      "Test circle -13 trained in 0.47 seconds.\n",
      "Test circle -12 trained in 0.48 seconds.\n",
      "Test circle -11 trained in 0.53 seconds.\n",
      "Test circle -10 trained in 0.46 seconds.\n",
      "Test circle -9 trained in 0.52 seconds.\n",
      "Test circle -8 trained in 0.49 seconds.\n",
      "Test circle -7 trained in 0.51 seconds.\n",
      "Test circle -6 trained in 0.52 seconds.\n",
      "Test circle -5 trained in 0.5 seconds.\n",
      "Test circle -4 trained in 0.48 seconds.\n",
      "Test circle -3 trained in 0.5 seconds.\n",
      "Test circle -2 trained in 0.51 seconds.\n",
      "Test circle -1 trained in 0.48 seconds.\n",
      "Model trained in 354.23 seconds.\n",
      "Hyperparams (1, 1) trained in 354.26 seconds.\n",
      "=========================================================================================\n",
      "Model (1, 1) Ended\n",
      "=========================================================================================\n"
     ]
    }
   ],
   "source": [
    "GARCH_val_results = []\n",
    "time_to_train_GARCH = {}\n",
    "all_aic = []\n",
    "\n",
    "for i in GARCH_params:\n",
    "    total_time = time.time()\n",
    "    \n",
    "    print('=========================================================================================')\n",
    "    print(f'Model {i} starts')\n",
    "    print('=========================================================================================')\n",
    "    standard_dev, aic = model_validation.rolling_window(model_validation.tech_val, \n",
    "                                                model_validation.binary_val, \n",
    "                                                model_validation.window, \n",
    "                                                model_validation.val_size, \n",
    "                                                model_validation.returns_val,\n",
    "                                                val_model='GARCH',\n",
    "                                                GARCH_params=i,\n",
    "                                                summary_garch=True)\n",
    "    \n",
    "    all_aic.append(aic)\n",
    "    GARCH_val_results.append(standard_dev)\n",
    "    \n",
    "    elapsed_time = round(time.time() - total_time, 2)\n",
    "    print(f\"Hyperparams {i} trained in {elapsed_time} seconds.\")    \n",
    "    time_to_train_GARCH[i] = f'{elapsed_time} seconds'\n",
    "    print('=========================================================================================')\n",
    "    print(f'Model {i} Ended')\n",
    "    print('=========================================================================================')\n",
    "    \n",
    "   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b9a1f722",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fa4f6534",
   "metadata": {},
   "outputs": [],
   "source": [
    "def GARCH_model(self, returns, GARCH_params=(1, 1), summary=False):\n",
    "    assert type(returns) == pd.Series\n",
    "    model = arch_model(\n",
    "        returns,\n",
    "        p=GARCH_params[0],\n",
    "        q=GARCH_params[1],\n",
    "        rescale=False\n",
    "    )\n",
    "\n",
    "    model_res = model.fit(disp=0)\n",
    "    next_pred = model_res.forecast(horizon=1, reindex=False).variance.iloc[0,0] ** 0.5\n",
    "\n",
    "    if summary:\n",
    "        aic = model_res.aic\n",
    "        return next_pred, aic\n",
    "\n",
    "    return next_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "3a0d96ab",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# for i in range(0, len(all_aic)):\n",
    "#     print(f' Model {GARCH_params[i]} AIC Score {pd.Series(all_aic[i].values()).mean()}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "a9a4ed5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save validation results Random Forest\n",
    "\n",
    "\n",
    "# pickle.dump(random_forest_val_results, open(\"random_forest_val_results.pkl\", \"wb\"))\n",
    "# pickle.dump(time_to_train_RF, open(\"time_to_train_RF.pkl\", \"wb\"))\n",
    "# pickle.dump(RF_params, open(\"RF_params.pkl\", \"wb\"))\n",
    "# pickle.dump(param_grid_random_forest, open(\"param_grid_random_forest.pkl\", \"wb\"))\n",
    "\n",
    "# Save validation results GARCH\n",
    "\n",
    "\n",
    "# pickle.dump(GARCH_params, open(\"GARCH_params.pkl\", \"wb\"))\n",
    "# pickle.dump(param_grid_garch, open(\"param_grid_garch.pkl\", \"wb\"))\n",
    "# pickle.dump(time_to_train_GARCH, open(\"time_to_train_GARCH.pkl\", \"wb\"))\n",
    "# pickle.dump(all_aic, open(\"all_aic.pkl\", \"wb\"))\n",
    "# pickle.dump(GARCH_val_results, open(\"GARCH_val_results.pkl\", \"wb\"))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "81683814",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save actual prices and returns classification\n",
    "# pickle.dump(model_validation.binary_val, open(\"binary_realized.pkl\", \"wb\"))\n",
    "# pickle.dump(model_validation.ohlc_val, open(\"ohlc_realized.pkl\", \"wb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5fbda4ee",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c3c9e09b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e83302ce",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ba66d8e8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "15b1fb74",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8596015f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bd7bfe0f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# direction = model_validation.rolling_window(model_validation.tech_val, \n",
    "#                                         model_validation.binary_val, \n",
    "#                                         model_validation.window, \n",
    "#                                         model_validation.val_size, \n",
    "#                                         model_validation.returns_val,\n",
    "#                                         val_model='RF',\n",
    "#                                         RF_params=(50, 3, None),\n",
    "#                                         features=model_validation.feat_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77aac544",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0e0d94e4",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2c8d3222",
   "metadata": {},
   "outputs": [],
   "source": [
    "# train_set = pd.concat([model_validation.tech[company], model_validation.features], axis=1).shift().iloc[1:window]\n",
    "# train_binary = model_validation.binary_val.iloc[:window-1][company]\n",
    "# assert len(train_binary) == len(train_set)\n",
    "# test_set = pd.concat([model_validation.tech[company], model_validation.features], axis=1).shift().iloc[window+1:]\n",
    "# test_binary = model_validation.binary_val.iloc[window:][company]\n",
    "# assert len(test_set) == len(test_binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c9567456",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1991b56e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.linear_model import LinearRegression\n",
    "# reg = LinearRegression().fit(train_set, train_binary)\n",
    "# reg.score(train_set, train_binary)\n",
    "\n",
    "# import statsmodels.api as sm\n",
    "# X_ols = sm.add_constant(train_set) # [['MACD', 'CL=F', 'EURUSD=X']]\n",
    "# model_ols = sm.OLS(train_binary,X_ols)\n",
    "# results_ols = model_ols.fit()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5551ac61",
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn import svm\n",
    "# clf = svm.SVC()\n",
    "# clf.fit(train_set, train_binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "72547408",
   "metadata": {},
   "outputs": [],
   "source": [
    "# res = clf.predict(train_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "608dd540",
   "metadata": {},
   "outputs": [],
   "source": [
    "# res_dict_backtest_SVC = {'f1':None,\n",
    "#                     'recall':None,\n",
    "#                     'precision':None,\n",
    "#                     'accuracy':None}\n",
    "# res_dict_backtest_SVC['f1'] = metrics.f1_score(train_binary, res)\n",
    "# res_dict_backtest_SVC['recall'] = metrics.recall_score(train_binary, res)\n",
    "# res_dict_backtest_SVC['precision'] = metrics.precision_score(train_binary, res)\n",
    "# res_dict_backtest_SVC['accuracy'] = metrics.accuracy_score(train_binary, res)\n",
    "# res_dict_backtest_SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f24c5bc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# res_out = clf.predict(test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1f59d386",
   "metadata": {},
   "outputs": [],
   "source": [
    "# res_dict_out_SVC = {'f1':None,\n",
    "#                     'recall':None,\n",
    "#                     'precision':None,\n",
    "#                     'accuracy':None}\n",
    "# res_dict_out_SVC['f1'] = metrics.f1_score(test_binary, res_out)\n",
    "# res_dict_out_SVC['recall'] = metrics.recall_score(test_binary, res_out)\n",
    "# res_dict_out_SVC['precision'] = metrics.precision_score(test_binary, res_out)\n",
    "# res_dict_out_SVC['accuracy'] = metrics.accuracy_score(test_binary, res_out)\n",
    "# res_dict_out_SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2b393bf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# RF_add_params = []\n",
    "# parameters_check = {\n",
    "#     'n_estimators': [20, 40, 60, 80, 100, 120, 140],\n",
    "#     'max_depth': [3, 5, 7, 9, 11, 13, 15, None],\n",
    "#     'max_leaf_nodes': [3, 5, 7, 9, 11, 13, None],\n",
    "# }\n",
    "\n",
    "# for i in itertools.product(parameters_check['n_estimators'],\n",
    "#                            parameters_check['max_depth'],\n",
    "#                            parameters_check['max_leaf_nodes']):\n",
    "    \n",
    "#     RF_add_params.append(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "41b5c322",
   "metadata": {},
   "outputs": [],
   "source": [
    "# clf = make_pipeline(RandomForestClassifier(random_state=42))\n",
    "# cv_acc_scores = cross_val_score(clf, train_set, train_binary, cv=5, n_jobs=-1)\n",
    "# print(cv_acc_scores)\n",
    "# params = {\n",
    "#     \"randomforestclassifier__n_estimators\": range(20, 140, 20),\n",
    "#     \"randomforestclassifier__max_depth\": range(3, 14, 3),\n",
    "#     \"randomforestclassifier__max_leaf_nodes\": range(3, 14, 3)\n",
    "# }\n",
    "# model = GridSearchCV(\n",
    "#     clf,\n",
    "#     param_grid=params,\n",
    "#     cv=5,\n",
    "#     n_jobs=-1,\n",
    "#     verbose=0\n",
    "# )\n",
    "\n",
    "# model.fit(train_set, train_binary)\n",
    "# cv_results = pd.DataFrame(model.cv_results_)\n",
    "# cv_results.sort_values('rank_test_score')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "275749d5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# company = tickers[1]\n",
    "\n",
    "# company"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6edcc59f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# train_set = pd.concat([model_validation.tech[company], model_validation.features], axis=1).iloc[1:window]\n",
    "# train_binary = model_validation.returns_val.iloc[:window-1][company] / 100\n",
    "# assert len(train_binary) == len(train_set)\n",
    "# test_set = pd.concat([model_validation.tech[company], model_validation.features], axis=1).iloc[window:]\n",
    "# test_binary = model_validation.returns_val.iloc[window-1:][company] / 100\n",
    "# assert len(test_set) == len(test_binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0bdb6630",
   "metadata": {},
   "outputs": [],
   "source": [
    "# model_final = RandomForestRegressor(n_estimators=40, \n",
    "#                                max_depth=3, \n",
    "#                                max_leaf_nodes=3, \n",
    "#                                random_state=42,\n",
    "#                                n_jobs=-1)\n",
    "\n",
    "\n",
    "# model_final.fit(train_set, train_binary)\n",
    "# insample = model_final.predict(train_set)\n",
    "# outofsample = model_final.predict(test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "35ab4835",
   "metadata": {},
   "outputs": [],
   "source": [
    "# print(metrics.mean_squared_error(train_binary, insample))\n",
    "# print(metrics.mean_squared_error(test_binary, outofsample))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3360a64a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# y_pred_baseline = [train_binary.mean()] * len(train_binary)\n",
    "# y_pred_baseline_test = [train_binary.mean()] * len(test_binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "462817ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "# print(metrics.mean_squared_error(train_binary, y_pred_baseline))\n",
    "# print(metrics.mean_squared_error(test_binary, y_pred_baseline_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7d554956",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a457fc58",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a937b801",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9c0af7a6",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "675c73ec",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "01daeb53",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# res_dict_backtest = {'f1':None,\n",
    "#                     'recall':None,\n",
    "#                     'precision':None,\n",
    "#                     'accuracy':None}\n",
    "# res_dict_backtest['f1'] = metrics.f1_score(train_binary, insample)\n",
    "# res_dict_backtest['recall'] = metrics.recall_score(train_binary, insample)\n",
    "# res_dict_backtest['precision'] = metrics.precision_score(train_binary, insample)\n",
    "# res_dict_backtest['accuracy'] = metrics.accuracy_score(train_binary, insample)\n",
    "# res_dict_backtest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0f630d10",
   "metadata": {},
   "outputs": [],
   "source": [
    "# res_dict_backtest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b2a48315",
   "metadata": {},
   "outputs": [],
   "source": [
    "# res_dict_outofsample = {'f1':None,\n",
    "#                     'recall':None,\n",
    "#                     'precision':None,\n",
    "#                     'accuracy':None}\n",
    "# res_dict_outofsample['f1'] = metrics.f1_score(test_binary, outofsample)\n",
    "# res_dict_outofsample['recall'] = metrics.recall_score(test_binary, outofsample)\n",
    "# res_dict_outofsample['precision'] = metrics.precision_score(test_binary, outofsample)\n",
    "# res_dict_outofsample['accuracy'] = metrics.accuracy_score(test_binary, outofsample)\n",
    "# res_dict_outofsample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f90acd8c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# res_dict_outofsample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3aa265f4",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}