felipelodur/Bike-Sharing-Prediction-with-NeuralNet

Developed as coursework for Udacity "Deep Learning Fundamentals" Nanodegree

23
/ 100
Experimental

Implements a fully connected neural network from scratch using only NumPy, including sigmoid activation, forward/backward propagation, and gradient descent optimization—no deep learning frameworks. Trained on daily bike rental data to perform regression prediction, with careful hyperparameter tuning (hidden units, learning rate, epochs) to balance accuracy and generalization while achieving sub-0.09 training loss and sub-0.18 validation loss.

No commits in the last 6 months.

No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 1 / 25
Community 17 / 25

How are scores calculated?

Stars

12

Forks

11

Language

Jupyter Notebook

License

Last pushed

Dec 21, 2017

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/felipelodur/Bike-Sharing-Prediction-with-NeuralNet"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.