miszkur/SelfSupervisedLearning
This repo reproduces results of "Understanding Self-Supervised Learning Dynamics without Contrastive Pairs" paper (https://arxiv.org/pdf/2102.06810.pdf)
This project helps deep learning researchers and practitioners understand and reproduce cutting-edge self-supervised learning methods like BYOL, SimSiam, DirectPred, and DirectCopy. It takes image datasets like CIFAR-10 or STL-10 and outputs trained models that can extract meaningful features without labeled data, along with visualizations of how the model learns. Researchers in computer vision and self-supervised learning would use this to validate research or explore model dynamics.
No commits in the last 6 months.
Use this if you are a deep learning researcher or practitioner interested in reproducing and exploring the dynamics of self-supervised learning models, particularly without contrastive pairs.
Not ideal if you are looking for a plug-and-play solution for general image classification without needing to understand the underlying self-supervised learning mechanisms.
Stars
9
Forks
1
Language
Jupyter Notebook
License
—
Category
Last pushed
Apr 30, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/miszkur/SelfSupervisedLearning"
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