abhuse/polyloss-pytorch

Polyloss Pytorch Implementation

38
/ 100
Emerging

This project offers an implementation of PolyLoss within the PyTorch framework, specifically designed to enhance classification tasks in machine learning. It takes raw prediction scores (logits) and corresponding true labels as input, and outputs a refined loss value used to train more accurate classification models. This is for machine learning engineers and researchers who are building and training deep learning models for categorization problems.

No commits in the last 6 months.

Use this if you are training deep learning classification models in PyTorch and want to experiment with advanced loss functions (Poly-Cross-Entropy or Poly-Focal Loss) to improve model performance, especially in scenarios with imbalanced classes or challenging samples.

Not ideal if you are working with non-PyTorch frameworks, regression problems, or if you prefer a simpler, out-of-the-box loss function without fine-tuning parameters like epsilon, alpha, or gamma.

deep-learning image-classification natural-language-processing model-training machine-learning-research
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

45

Forks

7

Language

Python

License

MIT

Last pushed

May 09, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/abhuse/polyloss-pytorch"

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