vt-nlp / event_query_extract Goto Github PK

Dear authors:
I am glad to learn your work. I have a question about the argument detection dataset. In the file run_arg_detection.py, the training dataset(batch data) used in code is different from the dataset in the file run_trigger_detection.py. But I only find the file save_dataset.py which is used to save the trigger detection dataset. Could you tell me hou can I store the argument detection dataset into .pt file? Thank you for your work!

亲爱的作者：
您好！我很高兴能阅读您的论文。我注意到代码中论元抽取代码里导入的.pt文件应该与事件检测导入的.pt文件不同。但是save_dataset.py文件中只有保存事件检测数据集的代码，请问您能告诉我如何保存论元检测的数据到.pt文件中吗？

对作者所写文章非常感兴趣，读了几遍，想复现论文。但是数据集以及数据处理始终困扰，请问raw_data在经过ACE_ERE_Scripts处理过后的.csv文件放进./data/ace_en/processed_data and ./data/ere_en/processed_data respectively之后；运行bssh setup.sh之前。还要经过什么处理呢

Dear Author:

When i use the file you provided to process the data i got this error:

Traceback (most recent call last):
 File "./preprocess/save_dataset.py", line 305, in
  read_from_source(args)
 File "./preprocess/save_dataset.py", line 289, in read_from_source
  raw_data = read_data_from(path, config.tokenizer, ace=args.ace)
 File "F:\Event_Query_Extract-main.\utils\data_to_dataloader.py", line 120, in read_data_from
  output = list(map(partial(unpack_ace_with_vt, tokenizer=tokenizer), output))
 File "F:\Event_Query_Extract-main.\utils\data_to_dataloader.py", line 60, in _unpack_ace_with_vt
  entities = data[-5].split(' ')
IndexError: list index out of range

I checked the code and guessed the possible cause of the problem
After run python preprocess/ace/read_args_with_entity_ace.py
I found the content of the output file as follows:

'Text'
'B-I-O TAG'
'xxxxxxxxxxx.csv'

It is seems to contradict function _unpack_ace_with_vt
Where is the Time and Value arguments and where is the POS-TAG?
What should I do?

Dear Author:

I noticed that the trained entity detection model needs to be used when running eval.py, but there is no file named scripts/run_entity_detection.py.

I found that model/entity_dection.py exists. So how should I train the model?

Dear Author:

I ran into a few problems when trying to run this codebase. Could you help with the following questions? Thanks!

1. In run_arg_detection.py Line8 and Line14, you imported Write2HtmlArg and FactContainer which are not included in the codebase.
2. The preprocess code in save_dataset.py only provides the dataset for run_trigger_detection.py, but not for run_arg_detection.py. In run_arg_detection.py, there are 16 values to be unpacked from a batch, but only 7 is preprocessed. So I don't know how to generate the dataset for EAE task.
3. I think there is a bug when you're preprocessing the arguments for each event. In https://github.com/VT-NLP/Event_Query_Extract/blob/main/preprocess/save_dataset.py#L256, if there is multiple events in a sentence, the arg_list only contains the arguments of the last one. Since the evaluation code of EAE isn't included in this codebase like mentioned in 1., I don't know how you manage to align multiple event mentions with only one arg_list in the end.
4. The paper mentioned zero-shot EE, but zero-shot related code isn't included in the repository. Since zero-shot requires a different data format (in the paper, you said that the prototype triggers aren't included), I think adding the code for this part may help us reproduce the result.
5. We need to add train.doc.txt,dev.doc.txt,test.doc.txt in data/splits/ACE05-E/ before running ./setup.sh, each file representing the document split of the corresponding dataset. The format of each file is that every line contains the document name. I think you could state that in README.md to help.
6. The pos_tag ,time and value wouldn't appear if we stick to your preprocessing procedure in ACE_ERE_scripts repository. So if we want to run save_dataset.py successfully, we need to change _unpack_ace_with_vt like I mentioned in #4.

Dear authors:
I am glad to learn your work. I have a question about the argument detection dataset. In the file run_arg_detection.py, the training dataset(batch data) used in code is different from the dataset in the file run_trigger_detection.py. But I only find the file save_dataset.py which is used to save the trigger detection dataset. Could you tell me hou can I store the argument detection dataset into .pt file? Thank you for your work!

亲爱的作者：
您好！我很高兴能阅读您的论文。我注意到代码中论元抽取代码里导入的.pt文件应该与事件检测导入的.pt文件不同。但是save_dataset.py文件中只有保存事件检测数据集的代码，请问您能告诉我如何保存论元检测的数据到.pt文件中吗？

请问requirements.txt在哪呢

_

Thank you for sharing your work！！I was stuck in the data preprocessing section while training,What are the files placed in the path ,split path and output in the data folder?

Dear Author:

Thanks for sharing you work! I have a question about the usage of trigger information in arg_detection model.

In your arg_detection model,
https://github.com/VT-NLP/Event_Query_Extract/blob/main/model/argument_detection/arg_detection.py#L205
after getting the trigger embeddings, it seems that you did not use them for further calculations, which doesn't align with the definition of "trigger-aware entity representations" in your paper "2.2 Event Argument Extraction, Multiway Attention".
In other words, I could not find where the trigger information is used in arg_detection models.

Could you help me with this problem if possible? Thanks!