zhangj111 / rencos Goto Github PK

Hi, would it be possible to upload the weights for the model which you trained in your paper? I would like to try out the comment generation without having to train the model, which is very resource and time intensive.

https://docs.python.org/2/library/ast.html
这个可以提供一下代码吗

I want to use my own dataset to run this model.
so could you give me the code of how how to generate rain/test.ast.src.
Thank you very much.

Dear Author:
I am trying to visualize the attention, but I can find where the attention information was stored. Could you tell me where the attention information store? And Which step? Thank you!

Hi, can you please share the dataset used after duplication removal and setting code length and comment length to 300 and 30 respectively?

The test.txt.src has 8714 lines, but the test.spl.src only has 6789 lines.In train or valid set,the lines of txt.src are the same as spl.src
I need your help!!

In line 39
should it be return super(DatasetBase, self).__reduce_ex__(proto)?

In addition, if the trained model is too large, can you please provide the summaries and its reference generated by translate?

I run "python run.py preprocess java" have the error:

Traceback (most recent call last):
File "preprocess.py", line 170, in
main()
File "preprocess.py", line 160, in main
train_dataset_files = build_save_dataset('train', fields, opt)
File "preprocess.py", line 122, in build_save_dataset
corpus_type, opt, ref_corpus)
File "preprocess.py", line 99, in build_save_in_shards
torch.save(dataset, pt_file)
File "/opt/anaconda3/lib/python3.7/site-packages/torch/serialization.py", line 328, in save
_legacy_save(obj, opened_file, pickle_module, pickle_protocol)
File "/opt/anaconda3/lib/python3.7/site-packages/torch/serialization.py", line 401, in _legacy_save
pickler.dump(obj)
File "/data1/chao/rencos/onmt/inputters/dataset_base.py", line 39, in reduce_ex
return super(DatasetBase, self).reduce_ex()
TypeError: reduce_ex() takes exactly one argument (0 given)