New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Pythia model usage with VizWiz #891
Comments
Hi,
Pythia model also requires ResNet152 grid features for which there is a script as well in tools. Alternatively, kick out the encoder from the model's config. |
Closing this as resolved. If you have more questions, please open up a new issue. |
@apsdehal OverviewI can test the pretrained vqa2 model on vqa2 dataset; can train pythia model for the vizwiz dataset, but cannot test the pretrained vqa2 model on the vizwiz dataset. command line I runmmf_predict dataset=vizwiz model=pythia config=projects/pythia/configs/vizwiz/defaults.yaml run_type=test checkpoint.resume_file=/data/vqa/mmf/data/models/pythia_train_val.pth The pythia_train_val.pth is provided by #697 (comment) error messageThe first error I encounter is To address it, I referred to issue #207 and changed mmf/mmf/configs/datasets/vizwiz/defaults.yaml' s answer_processor vocab_file from "vizwiz/v2019/extras/vocabs/answers_vizwiz_7k.txt" to "vqa2/defaults/extras/vocabs/answers_vqa.txt" To address this, I referred this issue and removed " Thanks for the help in advance! You have resolved many issues posted by me hahaha, although I know many of the issues I posted are naive and silly. Thank you again. @apsdehal |
I want to convert my dataset to the 'vizwiz' format and use Pythia model. I found several difficulties.
I have extracted features as the intuition, and get a folder composed of "xxx.npy" and "xxx_info.npy". How to use "tools/scripts/features/lmdb_conversion.py" to convert npy feature to lmdb feature. What is the argument "--mode"? What enviroment it dpends?
I use the original vizwiz dataset, download every thing automatically, and run "mmf_run config=projects/pythia/configs/vizwiz/defaults.yaml run_type=train dataset=vizwiz model=pythia". Assert Error happens:
Traceback (most recent call last):
File "/home/wulonglyx/miniconda3/envs/mmf/bin/mmf_run", line 33, in
sys.exit(load_entry_point('mmf', 'console_scripts', 'mmf_run')())
File "/home/wulonglyx/competition/imagecaption/mmf/mmf_cli/run.py", line 133, in run
main(configuration, predict=predict)
File "/home/wulonglyx/competition/imagecaption/mmf/mmf_cli/run.py", line 56, in main
trainer.train()
File "/home/wulonglyx/competition/imagecaption/mmf/mmf/trainers/mmf_trainer.py", line 141, in train
self.training_loop()
File "/home/wulonglyx/competition/imagecaption/mmf/mmf/trainers/core/training_loop.py", line 33, in training_loop
self.run_training_epoch()
File "/home/wulonglyx/competition/imagecaption/mmf/mmf/trainers/core/training_loop.py", line 91, in run_training_epoch
report = self.run_training_batch(batch, num_batches_for_this_update)
File "/home/wulonglyx/competition/imagecaption/mmf/mmf/trainers/core/training_loop.py", line 165, in run_training_batch
report = self._forward(batch)
File "/home/wulonglyx/competition/imagecaption/mmf/mmf/trainers/core/training_loop.py", line 182, in _forward
model_output = self.model(prepared_batch)
File "/home/wulonglyx/competition/imagecaption/mmf/mmf/models/base_model.py", line 236, in call
model_output = super().call(sample_list, *args, **kwargs)
File "/home/wulonglyx/miniconda3/envs/mmf/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/wulonglyx/competition/imagecaption/mmf/mmf/models/pythia.py", line 294, in forward
"image", sample_list, text_embedding_total
File "/home/wulonglyx/competition/imagecaption/mmf/mmf/models/pythia.py", line 241, in process_feature_embedding
"to number of features, {}.".format(len(feature_encoders), len(features))
AssertionError: Number of feature encoders, 2 are not equal to number of features, 1.
The text was updated successfully, but these errors were encountered: