ShadeAlsha/LTR-weight-balancing
CVPR 2022 - official implementation for "Long-Tailed Recognition via Weight Balancing" https://arxiv.org/abs/2203.14197
When training image recognition models, you often find that the model performs poorly on categories with less training data, while doing very well on categories with a lot of data. This project helps you train more balanced models that are better at recognizing objects from 'rare' categories. It takes your existing image dataset and training setup, and outputs a more accurate, balanced image recognition model. This is useful for anyone building computer vision systems where some object categories appear less frequently than others.
128 stars. No commits in the last 6 months.
Use this if your image recognition models struggle to identify objects from categories that have fewer examples in your training data.
Not ideal if your dataset has a perfectly even distribution of images across all object categories, or if you are not working with image recognition.
Stars
128
Forks
11
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Nov 30, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ShadeAlsha/LTR-weight-balancing"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
AdaptiveMotorControlLab/CEBRA
Learnable latent embeddings for joint behavioral and neural analysis - Official implementation of CEBRA
theolepage/sslsv
Toolkit for training and evaluating Self-Supervised Learning (SSL) frameworks for Speaker...
PaddlePaddle/PASSL
PASSL包含 SimCLR，MoCo v1/v2，BYOL，CLIP，PixPro，simsiam, SwAV, BEiT，MAE 等图像自监督算法以及 Vision...
YGZWQZD/LAMDA-SSL
30 Semi-Supervised Learning Algorithms
ModSSC/ModSSC
ModSSC: A Modular Framework for Semi Supervised Classification