nexusapoorvacus / DeepVariationStructuredRL

A PyTorch implementation of the "Deep Variation-structured Reinforcement Learning for Visual Relationship and Attribute Detection" paper by Liang et. al.
63 stars 12 forks source link

"evaluate" function not found #3

Open neka-nat opened 6 years ago

neka-nat commented 6 years ago

Hi @nexusapoorvacus ,

I ran the script in test mode, and the following error occurred.

Are there solutions to this problem?

alicedingyueming commented 5 years ago

the same problem!

berthyf96 commented 5 years ago

I also have the same problem. Also the training script does not save any checkpoints...

CodHeK commented 4 years ago

@nexusapoorvacus any updates on this? I am facing the same problem!

Looking into the earlier commits: https://github.com/nexusapoorvacus/DeepVariationStructuredRL/blob/3a3eab37a8b7be36b25c393fc0b8b9cc0bcf3cdf/main.py#L235

there was no function body at all for the evaluate function.

Only options rests, is to train and produce the image_states.pickle files and then directly run the pickle_to_files.py and then visualize!