search

THU-KEG / OmniEvent

A comprehensive, unified and modular event extraction toolkit.
https://omnievent.readthedocs.io/
MIT License
336 stars 31 forks source link

ERE: Event Relation Extraction #48

Open EslamAli86 opened 8 months ago

EslamAli86 commented 8 months ago

Hello,

Thank you for this amazing toolkit. I have a question as I tried to run the Code 2 example on page 4 in the paper (https://arxiv.org/pdf/2309.14258.pdf)

event extraction & relation extraction

all_results = infer(text=text, task="EE & ERE")

That part of the code throws the following exception:

Traceback (most recent call last): File "/storage/home/grads/ehussein/OmniEvent/test.py", line 20, in all_results = infer(text=text, task="ERE") File "/storage/home/grads/ehussein/OmniEvent/OmniEvent/infer.py", line 107, in infer assert task in ['ED', 'EAE', 'EE'] AssertionError

The toolkit does not support the ERE part yet. Do you think I need to do something to infer the event relation extraction? or will this part be released soon?

Thank you

h-peng17 commented 7 months ago

Thanks for the reminder! we will release the ERE code and models in 2 weeks. And I'm attending EMNLP 2023 in Singapore. Welcome to our poster to communicate!

EslamAli86 commented 7 months ago

Thank you so much for letting us know, we appreciate it. We are waiting for the new release. And good luck with your poster presentation @ EMNLP :)

EslamAli86 commented 6 months ago

Hi Hao,

I hope this message finds you well. I was wondering if you have any plans to release the ERE module of OmniEvent. Its availability would greatly assist us in our research.

Thank you,

h-peng17 commented 6 months ago

Hello, I'm very sorry for the delay in the release of OmniEvent-ERE due to recent busy schedules. If you urgently need the ERE module, I can send you the original OmniEvent-ERE code via email. Once this busy phase is over, I will integrate OmniEvent-ERE into the current repository.

EslamAli86 commented 5 months ago

Hello Hao,

I appreciate you sharing your current repository for the ERE part. It seems that the model needs training on various datasets to predict different types of relationships between events. Unfortunately, I don't have access to most of these datasets. Could you please share either the trained models or the processed datasets?

Thank you for your assistance; I appreciate it.

h-peng17 commented 4 months ago

You can get some processed datasets at this repository (https://github.com/THU-KEG/MAVEN-ERE). For all the ERE datasets that we use, we abide by their licenses and will open source them together with the codes.

EslamAli86 commented 4 months ago

Hi Hao,

Thank you so much for providing most of the dataset, I understand you can not provide the one that are distributed under licenses.

For the MAVEN-ERE dataset that is processed by the python script maven-ere.py under OmniEvent-ERE/scripts/data_processing/MAVEN-ERE, there seem to be a mismatch between the json format of the data under ( https://github.com/THU-KEG/MAVEN-ERE) and the expected json format read by maven-ere.py. Could you please provide the MAVEN-ERE dataset in the json format processed by the script maven-ere.py?

Thank you and sorry for any inconvenience,

Eslam

On Wed, Feb 28, 2024 at 2:18 AM Hao Peng @.***> wrote:

You can get some processed datasets at this repository ( https://github.com/THU-KEG/MAVEN-ERE). For all the ERE datasets that we use, we abide by their licenses and will open source them together with the codes.

— Reply to this email directly, view it on GitHub https://github.com/THU-KEG/OmniEvent/issues/48#issuecomment-1968372505, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADU5J6MWJDEOKXOWXEP5WZLYV3K5TAVCNFSM6AAAAAA76SJG2KVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNRYGM3TENJQGU . You are receiving this because you authored the thread.Message ID: @.***>