A simple digit recognition demo using keras and streamlit. It uses streamlit-drawable-canvas for drawing on canvas.
streamlit is an open-source app framework, which is the easiest way for data scientists and machine learning engineers to create beautiful, performant apps. All in pure Python, no longer fiddling with javascript.
This demo contains two parts: training a simple digit recognition model using mnist dataset and a webapp to live demo that model.
-
First install all the dependencies
pip install -r requirements.txt -
Train model
Run all the cells of train.ipynb manually or run it using runipy.
runipy train.ipynb -
Run demo web-app
Demo your model by running app.py
streamlit run app.py