This repository contains a comprehensive project that performs Exploratory Data Analysis (EDA) and classification tasks on the Iris dataset. The project utilizes FastAPI for creating a robust API endpoints, and Streamlit for building an interactive web application.
The goal of this project is to provide a hands-on approach to understanding data analysis and machine learning workflows while demonstrating the use of FastAPI and Streamlit in a practical context.
The following links were used to implement the multi-page functionality in streamlit:
- https://docs.streamlit.io/get-started/tutorials/create-a-multipage-app
- https://docs.streamlit.io/library/advanced-features/multipage-apps
To run the streamlit app, first FastAPI app should be running. To run the FastAPI app, run the following command:
uvicorn main:app --reload
Once the FastAPI app is running, run the following command to run the streamlit app:
streamlit run app/Home.py
OR, you can run the app using make command:
make run_app
It will run both FastAPI and Streamlit apps, and open the streamlit app in the browser.
This streamlit app is deployed on streamlit cloud. You can access the deployed app using the following link:
The classification results are displayed in the streamlit app on a separate page and look like the following:
requirements.yml file contains the list of all the packages required to run the code in this repository. requirements.yml is generated using the following command:
conda env export --no-builds | grep -v "prefix" > requirements.yml
To create a conda environment using the requirements.yml file, run the following command:
conda env create -f requirements.yml