In this project we will go through some basic data pre-processing with Python, visualize our dataset with Seaborn and Matplotlib, split it with sklearn, and train a simple Multi-Layer Perceptron (MLP) using PyTorch to solve the WIDS22 Challenge prediction problem. The implementation is based on the Kaggle notebook.
We recommend using a python virtual environment
python3 -m venv WIDS python=3.10
Install the requirements
pip3 install -r requirements.txt
Train the model
python train.py
You can visualize the dataset by passing the command -v True and change the epoch number by setting -e <number_of_epochs>.
The training statistics can be found on tensorboard log directory and can be accessed by running:
tensorboard --logdir runs