text-machine-lab / entity-coref Goto Github PK

Hi,
Thank you for the source codes. It is giving error while running the train_triad.py file, while calling train function.
Please suggest the required modification.
Thanks in advance.

File "/Data/chanchal_1821cs11/entity-coref-master/src/build_data.py", line 563, in worker
X[i].append(X_triad[i])
IndexError: list index out of range

The README says:

We developed the system with Python 3.6. You also need pytorch. We used 0.4.0. You could also use Keras with TensorFlow, but we did not test the final performance on Keras. We used Keras 2.1.5 and TensorFlow 1.7.0

This is confusing - at first it seems to say you used pytorch and didn't try Keras with Tensorflow, but then it says you did use Keras with Tensorflow?

I have trained on the ".gold_conll" and tested on the ".gold_conll". However, my result is higher than the result in your paper.
On the other hand , I find the training method you introduced in readme is different from the method in your paper - the output is 2. Class CorefTaggerReview means what?

I find the output is 12 vector in Class CorefTagger，and the final output y is a 13 vector in your paper. Are there any differences?

another question: did you test on ".auto_conll" in you paper (CONLLING 2018)

I have get the data sets from CoNLL-2012 and the rest data sets in LDC, which don't include the files that have the suffix with *_gold_conll. However I can get three files that have the suffix with *english.v4_gold_conll if I run the code in conll-2012. but it seem not suit for your profile. so ,where can i get the data sets that include the files that your profile request?

No entities found in file: bc/phoenix/00/phoenix_0003-0
No entities found in file: bc/msnbc/00/msnbc_0007-11
No entities found in file: bc/phoenix/00/phoenix_0003-1
No entities found in file: bc/phoenix/00/phoenix_0003-2
No entities found in file: bc/phoenix/00/phoenix_0003-4
No entities found in file: bc/phoenix/00/phoenix_0003-5
No entities found in file: bc/phoenix/00/phoenix_0003-3
No entities found in file: bc/cnn/00/cnn_0008-0
No entities found in file: bc/cnn/00/cnn_0008-1
No entities found in file: bc/phoenix/00/phoenix_0003-6
No entities found in file: bc/cnn/00/cnn_0008-3
No entities found in file: bc/cnn/00/cnn_0008-2
No entities found in file: bc/cnn/00/cnn_0008-6
No entities found in file: bc/cnn/00/cnn_0008-8
No entities found in file: bc/cnn/00/cnn_0008-4
No entities found in file: bc/cnn/00/cnn_0008-7
No entities found in file: bc/cnn/00/cnn_0008-5
No entities found in file: bc/cnn/00/cnn_0008-9
No entities found in file: bc/cnn/00/cnn_0008-13
No entities found in file: bc/cnn/00/cnn_0008-12
No entities found in file: bc/cnn/00/cnn_0008-10
No entities found in file: bc/cnn/00/cnn_0008-15
No entities found in file: bc/cnn/00/cnn_0008-11
No entities found in file: bc/cnn/00/cnn_0008-14
Completed saving results!
Pairwise evaluation:
True histogram (array([0, 0, 0, 0]), array([0. , 0.25, 0.5 , 0.75, 1. ]))
Prediction histogram (array([0, 0, 0, 0]), array([0. , 0.25, 0.5 , 0.75, 1. ]))
No entities found in file: mz/sinorama/10/ectb_1009-0
No entities found in file: mz/sinorama/10/ectb_1009-3
No entities found in file: mz/sinorama/10/ectb_1009-1
Traceback (most recent call last):
File "predict.py", line 126, in
main()
File "predict.py", line 110, in main
clustering_only=args.clustering_only, compute_linkage=args.compute_linkage)
File "/home/wlk/entity-coref-master/src/evaluator.py", line 295, in write_results
print(classification_report(all_pairs_true, np.round(all_pairs_pred), digits=3))
File "/usr/local/lib/python3.5/dist-packages/sklearn/metrics/classification.py", line 1568, in classification_report
name_width = max(len(cn) for cn in target_names)
ValueError: max() arg is an empty sequence