rhfeiyang/PPOT
Official implementation of 'P$^2$OT: Progressive Partial Optimal Transport for Deep Imbalanced Clustering'. (Accepted by ICLR 2024)
This tool helps researchers and data scientists improve the accuracy of image classification and recognition when working with datasets where some categories have far fewer examples than others. You input your imbalanced image dataset, and it outputs a more robust clustering model that can better identify objects or features in underrepresented classes. This is ideal for those building computer vision models for diverse real-world scenarios.
No commits in the last 6 months.
Use this if you are performing image clustering on datasets where certain categories are significantly less common than others, and you need a method that performs well across all categories, including the rare ones.
Not ideal if your datasets are well-balanced with similar numbers of examples for each category, or if you are not working with image data.
Stars
18
Forks
2
Language
Python
License
—
Category
Last pushed
Jan 19, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/rhfeiyang/PPOT"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Compare
Higher-rated alternatives
PythonOT/POT
POT : Python Optimal Transport
recursionpharma/gflownet
GFlowNet library specialized for graph & molecular data
IShengFang/SpectralNormalizationKeras
Spectral Normalization for Keras Dense and Convolution Layers
dfdazac/wassdistance
Approximating Wasserstein distances with PyTorch
iamalexkorotin/NeuralOptimalTransport
PyTorch implementation of "Neural Optimal Transport" (ICLR 2023 Spotlight)