training_pipelines

Streamlit app for easy classification modeling It's so easy that you should only have to clone this repo, install requirements and run the streamlit app. We'll handle the rest :)

0. Clone the repository, install dependencies and run the app

Clone the repo and go to that folder in the terminal

git clone https://github.com/tatiana-iazykova/training_pipelines.git
cd training_pipelines

Optional (create virtual environment):

python3 -m venv venv 

For Unix-like OS:

source venv/bin/activate

For Windows:

source venv/Scripts/activate

Upgrade pip, wheel and setuptools, install dependencies and the run streamlit app

pip install -U pip wheel setuptools && pip install -r requirements.txt
streamlit run Home_Page.py

1. Head to EDA page to get you data assessed

2. Head to the page with the model you want to try. For now you can choose only the baseline model.

3. After you train you model you can donwload it neatly packed in a zip archive with all the necessary things for running you own streamlit inference application

4. Unzip the archive as a folder, open the folder directory in terminal and run

   1. chmod 777 run.sh

   2. ./run.sh

5. Or you can do all the things specified in run.sh manually

   1. create virtual environment:

      python3 -m venv venv

   2. activate the created virtual environment

      For Unix-like systems:

          source venv/bin/activate

      For Windows:

          source venv/Scripts/activate

   3. upgrade pip, wheel and setuptools

      pip install -U pip wheel setuptools

   4. install the requirements

      pip install -r requirements.txt

   5. run streamlit app

      streamlit run interface.py