This project implements a simple handwritten digit detector using machine learning. MNIST dataset is used for the training of the machine learning model.
- Model Comparison & Analysis.ipynb - Compares the performance of different machine learning models to select the best one.
- Detection.ipynb - Trains the best perforning model on the dataset and tests it on test case images (test case images are included in the /images directory). Image pre-processing procedures are also applied in this notebook.
- GUI.ipynb - It implements a simple GUI app using tkinter library of Python in addition to the machine learning and image pre-processing parts.
Python libraries required for the project:
- scikit-learn
- numpy
- pandas
- matplotlib
- tkinter
Download these libraries using pip in the command line(Windows) or hyper terminal(Mac) and run the jupyter notebook.