- Load and prepare data: create dummy variables/standardize target variables/Split data into train/valid/test.
- Write out the Neural Network structure and training loop only using numpy from scratch.
- Set hyper parameter and train/valid the model, plot train/valid loss.
- Use test data to see how well it works and plot the results.
wenzhenghe7/Predicting-Bike-Sharing-Trend
Build and train neural network from scratch using only numpy to predict bike sharing trend.
Jupyter Notebook