search

Bartzi / see

Code for the AAAI 2018 publication "SEE: Towards Semi-Supervised End-to-End Scene Text Recognition"
GNU General Public License v3.0
575 stars 147 forks source link

FSNS demo #103

Open sne21star opened 3 years ago

sne21star commented 3 years ago

When preparing the dataset for the FSNS demo it says to run this command: python swap_classes.py 0 133. What exactly is the gt_file?

Bartzi commented 3 years ago

For the demo, you do not need to do this. However, if you try to prepare data for training, you'll need to do this. gt_file refers to the file that is produced by the script tfrecord_to_image.py

sne21star commented 3 years ago

Yes I was able to get that part working thank you. Also did you ever get this error before and know how to solve it? Traceback (most recent call last): File "/content/drive/My Drive/Colab Notebooks/see/chainer/train_fsns.py", line 169, in updater = MultiprocessParallelUpdater(train_iterators, optimizer, devices=args.gpus) File "/usr/local/lib/python3.6/dist-packages/chainer/training/updaters/multiprocess_parallel_updater.py", line 145, in init assert len(iterators) == len(devices) AssertionError

Sorry I think I fixed that ^ error with using standardupdater but now I can't seem it run it updater = StandardUpdater(train_iterators, optimizer, device=args.gpus) when I changed the multiprocessor to this line.

This is also the command line that I am using:

!python /content/drive/My\ Drive/Colab\ Notebooks/see/chainer/train_fsns.py /content/drive/My\ Drive/Colab\ Notebooks/see/fsns_labels/curriculum.json /content/drive/My\ Drive/Colab\ Notebooks/see/fsns_labels --char-map /content/drive/My\ Drive/Colab\ Notebooks/see/datasets/fsns/fsns_char_map.json. --blank-label 0 -b 2 --gpu 0

Bartzi commented 3 years ago

okay, what is your new error? Is it the one in the other issue?