Detailed Description
- Desktop application to classify breast cancer patients
- The application is able to classify patients tumor type based on their features
- Using the dataset from the scikit-learn library it trains a neural network to classify tumors
- The model has configurable parameters to adjust the learning rate and the number of epochs in the GUI, and in the code it leave the option to change the number of neurons in the hidden layer, the activation function and the cost function
- It has a GUI to plot the data, choosing the features to use to see the correlation between them, and to adjust the parameters of the model
- It has a GUI to train the model, and to see the results of the training, showing the error rate and the decision boundary in a real time plot
- The GUI is built using tkinter and matplotlib
- The neural network is implemented using raw python with numpy arrays
- The code is written using the pythonic way, with the use of type hints and the use of the attrs module
- The design use a multi-paradigm approach, using both object-oriented and functional programming