Oussamayousre / exam_project

Reducers and KMeans Training App

Repository details

├── .git/ # git logs file 
├── .gitignore # git ignore file 
├── data/  # corpus data stored in  data folder
│   └── ng20_data.joblib
├── dockerfile # Docker file to create an image docker 
├── main.py # main file that contains streamlit app execution
├── models/ # pickles models folder
│   ├── K-means/
│   │   └── kmeans_umap.pkl
│   └── reducers/
│       └── umap_model.pkl
├── notebooks/  # notebook folder for testing
│   └── umap_method_testing.ipynb
├── readme.md
├── requirements.txt
├── src/
│   ├── __pycache__/
│   │   └── utils.cpython-310.pyc
│   └── utils.py
└── tree.py

repo clonning

git clone https://github.com/Oussamayousre/exam_project.git

Requirements

# this command is to install necessary requirements to use the streamlit application
pip install -r requirements.txt

Application run

without docker

python main.py
# then you can easily interact with the streamlit app 

Docker

docker build -t reducer_app .
docker run reduce_app 

Streamlit UI

the entire streamlit page

UMAP reducer projection of data

K means projcection of its results

Project Steps using git and GitHub :

To deploy the project on GitHub and have each group member develop a different method, we followed the following steps:

• Created a new repository named "exam_project" for the project on GitHub.
• Each member cloned the repository to their local machine using the command: git clone https://github.com/Oussamayousre/exam_project
• Each member created a branch.
• Each member switched to their branch using the command: git checkout 'branch_name'
• Each of us created a .ipynb file and developed our method. Then, we saved our changes using the commands: git add . git commit -m 'commit message'
• Each member pushed to their branch using the command: git push -u origin branchName
• Finally, we used the pull request function on GitHub to merge each branch with the main branch "main."