mabirck/adaptative-dropout-pytorch

Pytorch implementation of Adaptative Dropout a.ka Standout.

21
/ 100
Experimental

This project helps machine learning researchers or practitioners explore a specific neural network regularization technique called Adaptive Dropout (also known as Standout). It takes a neural network model built with PyTorch and applies this dropout method during training. The output is a trained model that aims to generalize better by reducing overfitting, though the current implementation notes indicate it may not perfectly replicate published performance benchmarks.

No commits in the last 6 months.

Use this if you are a machine learning researcher or deep learning practitioner experimenting with regularization techniques in PyTorch and specifically want to test or understand Adaptive Dropout.

Not ideal if you are looking for an out-of-the-box solution to immediately improve your model's performance without extensive experimentation, or if you require guaranteed state-of-the-art results without further tuning.

deep-learning neural-networks model-training machine-learning-research overfitting-prevention
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

12

Forks

Language

Python

License

MIT

Last pushed

Feb 22, 2018

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mabirck/adaptative-dropout-pytorch"

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