FilterAI

LLM content filtering module

How to use

1. To re-train the modele uncomment the training code and execute the script with:

python3 Bert/pages/notebook_pytorch.py

2. To execute the streamlit app with Bert model with:

streamlit run Bert/st_main.py

3. To execute the Llama prompted model:

python3 Llama/Llama_model.py

To contribute

1. Create a fork repo then you can clone your fork on your machine with:

git clone git@github.com:'yourfork-repo'/FilterAI.git

2. Then add another Git repository as a remote repository to keep your fork synchronized with the original repository with:

cd FilterAI
git remote add upstream git@github.com:arezki4/FilterAI.git

3. To retrieve changes from the remote repository, use:

git fetch upstream

4. then you can create a branch in your fork repo

git branch branche_name
git checkout branche_name

5. To update your repo from the remote repo

git fetch upstream/Master
git merge upstream/Master

6. After pushing your code you should create a Pull request on github and put the zone manager as a reviewer in order to review the code and validate your changes.

7. If you're the manager of the zone you've changed, you'll still need to ask a contributor for a code review, as no code can be pushed to master without a code review.