Neural network usage for solving binary classification problem

This repo contents the two following sections:

1. Visualization of training process and dataset analysis

Dataset analysis, preprocessing, metrics and other analyses are shown in Jupyter notebook binary_classification.ipynb.

2. Source files for neural network deployment

Three Python files can be found in src/ folder:

1. train.py - contains logic for neural network training,
2. inference.py - contains logic for predicting classes after training is completed,
3. helpers.py - contains some useful functions and variables, which are used by train.py and inference.py.

NOTE: after train.py is completed, the output/ folder is created. It contains three files:

1. neural_network.keras (TRAINED_MODEL variable in helpers.py) - contains trained neural network,
2. standard_scaler.json (STANDARD_SCALER variable in helpers.py) - contains Standard Scaler configs,
3. encoder.json (ENCODER variable in helpers.py) - contains Encoder configs.

Usage

All requirements need for running scripts can be found in requirements.txt.

Train the neural network

To train the neural network run:

python src/train.py data/p1_train.csv

Predict classes with trained neural network

To predict classes with trained neural network run:

python src/inference.py data/p1_test_student.csv