Handwritten Digit Recognition with Machine Learning

• Implemented with 6 classification algorithms
• Algorithms: KNN, SVM, RFC, MLP, RNN, CNN
  • K-Nearest Neighbors(KNN)
  • Support Vector Machines(SVM)
  • Random Forest Classification(RFC)
  • Multi-layer Perceptron(MLP)
  • Recurrent Neural Network(RNN)
  • Convolutional Neural Network(CNN)

Getting Started

• Install dependencies

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Training

Use -s to select a strategy for training. The available strategies are KNN, SVM, RFC, MLP, RNN, CNN

python main.py -s knn

Use -m to select a method to run. For example with init_trained_model you can pick up training from where you left off with a saved pickle inside of tmp/models

python main.py -m init_trained_model

Renders

Confusion matrices of the results of training using KNN as well as summaries of other models & their strategies.

References

• Classification Strategies - Scikit-Learn
• Charts - Matplotlib