Merge branch 'master' into safeprint

ocrolib/common.py

@@ -279,7 +279,6 @@ def read_page_segmentation(fname):
     segmentation = make_seg_black(segmentation)
     return segmentation
 
-@checks(str,PAGESEG)
 def write_page_segmentation(fname,image):
     """Writes a page segmentation, that is an RGB image whose values
     encode the segmentation of a page."""

ocrolib/lstm.py

@@ -380,7 +380,8 @@ def weights(self):
 
 def ffunc(x):
     "Nonlinearity used for gates."
-    return 1.0/(1.0+exp(-x))
+    # cliping to avoid overflows
+    return 1.0/(1.0+exp(clip(-x,-20,20)))
 def fprime(x,y=None):
     "Derivative of nonlinearity used for gates."
     if y is None: y = sigmoid(x)

ocropus-dewarp

@@ -2,20 +2,17 @@
 
 from __future__ import print_function
 
-import random as pyrandom
 import re
-from pylab import *
-import os.path
-import ocrolib
 import argparse
-import matplotlib
-import numpy
+import sys
+
+import numpy as np
+from scipy.misc import imsave
+
+import ocrolib
 from ocrolib import lineest
-import ocrolib.lstm as lstm
-import traceback
-import scipy
 
-numpy.seterr(divide='raise',over='raise',invalid='raise',under='ignore')
+np.seterr(divide='raise',over='raise',invalid='raise',under='ignore')
 
 parser = argparse.ArgumentParser("dewarp lines as used for the recognizer")
 
@@ -36,9 +33,9 @@ if len(inputs)==0:
 print("# inputs", len(inputs))
 
 if args.lineest=="center":
-  lnorm = lineest.CenterNormalizer()
+    lnorm = lineest.CenterNormalizer()
 else:
-  raise Exception(args.lineest+": unknown line normalizer")
+    raise Exception(args.lineest+": unknown line normalizer")
 
 lnorm.setHeight(args.height)
 
@@ -47,9 +44,9 @@ for fname in inputs:
     base,_ = ocrolib.allsplitext(fname)
     try:
         line = ocrolib.read_image_gray(fname)
-        lnorm.measure(amax(line)-line)
-        line = lnorm.normalize(line,cval=amax(line))
-        scipy.misc.imsave(base+".dew.png",line)
+        lnorm.measure(np.amax(line)-line)
+        line = lnorm.normalize(line,cval=np.amax(line))
+        imsave(base+".dew.png",line)
     except Exception as e:
         print("ERROR", e)
         continue

ocropus-econf

@@ -3,14 +3,21 @@
 
 from __future__ import print_function
 
-import warnings,numpy,argparse,sys,os,os.path,multiprocessing,codecs
+import warnings
+import argparse
+import sys
+import os.path
+import multiprocessing
+import codecs
 from collections import Counter
+
+import numpy as np
+
 import ocrolib
-from pylab import *
 from ocrolib import edist
 
 # disable rank warnings from polyfit
-warnings.simplefilter('ignore',numpy.RankWarning)
+warnings.simplefilter('ignore',np.RankWarning)
 
 parser = argparse.ArgumentParser(description = """
 Compute the edit distances between ground truth and recognizer output.

ocropus-errs

@@ -3,7 +3,12 @@
 
 from __future__ import print_function
 
-import argparse,sys,os,os.path,multiprocessing
+import argparse
+import sys
+import os
+import os.path
+import multiprocessing
+
 import ocrolib
 from ocrolib import edist
 
@@ -27,6 +32,7 @@ args.files = ocrolib.glob_all(args.files)
 if not ".gt." in args.files[0]:
     sys.stderr.write("warning: compare on .gt.txt files, not .txt files\n")
 
+
 def process1(fname):
     # fgt = ocrolib.allsplitext(fname)[0]+args.gtextension
     gt = ocrolib.project_text(ocrolib.read_text(fname),kind=args.kind)