How to train MTV?

[haratena9]

I want to train by below command, but an error has occured.
So I tried to download vit-tiny, vit-small, and vit-base weights.
But I couldn't found here and here.

• how can I get pretrained weight?
• And when after getting pretrained weight, what concrete steps should I take?

command from here
$ python -m scenic.projects.mtv.main \ --config=scenic/projects/mtv/configs/kinetics/k400_mtv_b2_cva.py \ --workdir=mtv_base_cva/

error
I0921 11:00:20.737020 140608837556032 pretrain_utils.py:328] Loading big_vision checkpoint from /path/to/vit-tiny Traceback (most recent call last): File "/usr/local/lib/python3.10/runpy.py", line 196, in _run_module_as_main return _run_code(code, main_globals, None, File "/usr/local/lib/python3.10/runpy.py", line 86, in _run_code exec(code, run_globals) File "/mnt/scenic/scenic/projects/mtv/main.py", line 58, in <module> app.run(main=main) File "/mnt/scenic/scenic/app.py", line 68, in run app.run(functools.partial(_run_main, main=main)) File "/usr/local/lib/python3.10/site-packages/absl/app.py", line 308, in run _run_main(main, args) File "/usr/local/lib/python3.10/site-packages/absl/app.py", line 254, in _run_main sys.exit(main(argv)) File "/mnt/scenic/scenic/app.py", line 109, in _run_main main(rng=rng, config=config, workdir=workdir, writer=writer) File "/mnt/scenic/scenic/projects/mtv/main.py", line 48, in main trainer( File "/mnt/scenic/scenic/projects/mtv/trainer.py", line 194, in train train_state = init_from_vit_checkpoints(config, model, train_state) File "/mnt/scenic/scenic/projects/mtv/trainer.py", line 101, in init_from_vit_checkpoints pretrain_utils.convert_big_vision_to_scenic_checkpoint( File "/mnt/scenic/scenic/train_lib/pretrain_utils.py", line 332, in convert_big_vision_to_scenic_checkpoint checkpoint_data = np.load(gfile.GFile(checkpoint_path, 'rb')) File "/usr/local/lib/python3.10/site-packages/numpy/lib/npyio.py", line 434, in load magic = fid.read(N) File "/usr/local/lib/python3.10/site-packages/tensorflow/python/lib/io/file_io.py", line 116, in read self._preread_check() File "/usr/local/lib/python3.10/site-packages/tensorflow/python/lib/io/file_io.py", line 77, in _preread_check self._read_buf = _pywrap_file_io.BufferedInputStream( tensorflow.python.framework.errors_impl.NotFoundError: /path/to/vit-tiny; No such file or directory

k400_mtv_b2_cva.py