How to change the code to work for multi-label classification? #138

Activity

1. urialon commented on Nov 15, 2021

   

   urialon

   on Nov 15, 2021

   Collaborator

   More actions

   Hi @allomy ,
   Thank you for your interest in code2vec!

   I think that you can loss here:
   https://github.com/tech-srl/code2vec/blob/master/tensorflow_model.py#L228
   from the standard cross entropy to sigmoid cross entropy: https://www.tensorflow.org/api_docs/python/tf/compat/v1/nn/sigmoid_cross_entropy_with_logits

   But you will also need to change the pipeline to support reading multi-labeled examples. Follow the variable target_index here: https://github.com/tech-srl/code2vec/blob/master/path_context_reader.py
   and modify it to get a list of targets for every example.

   Best,
   Uri
2. allomy commented on Nov 17, 2021

   

   allomy

   on Nov 17, 2021

   Author

   More actions

   Hi @urialon , thank you for your quick response. I'll try it soon.
3. allomy commented on Dec 3, 2021

   

   allomy

   on Dec 3, 2021

   Author

   More actions

   Hi @urialon , sorry for the delay response that I have tried to modify the code related to target_index, but was lost in the code... Could you give more information about modifying it to get a list of targets for every sample? Thank you in advance for your help.
4. urialon commented on Dec 6, 2021

   

   urialon

   on Dec 6, 2021

   Collaborator

   More actions

   Hi @allomy ,
   Actually it might be easiest for you to use https://code2seq.org/ .
   It predicts a sequence of labels and not multi-label, but it may either be a good approximation, or easier to adapt for multi-label (just change the loss computation, not the entire data reading pipeline).

   Best,
   Uri
5. allomy commented on Dec 7, 2021

   

   allomy

   on Dec 7, 2021

   Author

   More actions

   Thank you @urialon , I will take a look at code2seq.