This repo hosts the storage point and utilities for the LiRo NLP Hackathon - held on the 5th of November 2022 in Bucharest, Romania
Below is information for participants on how to upload their work at the end of the hackathon for evaluation.
We have the following tasks :
task1
- LiRo Challenge - Task 1 - Named Entity Recognitiontask2
- LiRo Challenge - Task 2 - Tweet Emotion Detectiontask3
- LiRo Challenge - Task 3 - Semantic Textual Similaritytask4
- LiRo Challenge - Task 4 - Sentence Segmentationtask5
- Diac Restore Challenge
Let's assume you want to upload your code for task1, and your team name is MyAwesomeTeam
Please follow these steps:
Please work in the following manner: create a folder with your team name (don't include punctuation please), and a subfolder corresponding to your task.
For our example, please create MyAwesomeTeam/task1/
and work in there.
In the task1
folder please have a file named model.py
that implements all the methods described in the colab for each task.
These are requirements for the automatic evaluation that will be run at the end of the hackathon.
This is required for task 5 in particular, where we won't have time to train your model from scratch, and you'll have to upload the model itself. Because models are usually >50MB , you'll need to install git lfs to be able to upload larger files.
There are several tutorials online on how to install git lfs. Here's a quick tutorial.
Logged in to GitHub with your user, create a fork of this repo.
Then, clone it locally, and in the upload
folder move the MyAwesomeTeam/task1/
with all your code.
This means that your code will now be in nlp-hackathon-2022/upload/MyAwesomeTeam/task1/
If you have large files, init git lfs at this point, and, in the command line, run ``git lfs track
Then run, from nlp-hackathon-2022/upload/MyAwesomeTeam/task1/
:
git add .
git commit -m "<team name here>"
git push origin main
this should upload all your files to your repo fork!
This is the last step, and done online from github.com from your repo fork. Click on the new commit in your fork and click Create a new Pull Request to the LiroBenchmark/nlp-hackathon-2022/ repo.