facebookresearch/moco-v3
PyTorch implementation of MoCo v3 for self-supervised training of ResNet and ViT models on ImageNet.
Not currently ranked — collecting fresh signals.
star history
This repository re-implements the MoCo v3 self-supervised contrastive learning algorithm in PyTorch for GPU training. It supports pre-training ResNet and Vision Transformer architectures on ImageNet-1k using various batch sizes, followed by linear evaluation or end-to-end fine-tuning. The implementation reproduces the results from the original TensorFlow/TPU-based paper.
Frequently asked
- What is facebookresearch/moco-v3?
- PyTorch implementation of MoCo v3 for self-supervised training of ResNet and ViT models on ImageNet.
- Is moco-v3 open source?
- Yes — facebookresearch/moco-v3 is an open-source project tracked on heatdrop.
- What language is moco-v3 written in?
- facebookresearch/moco-v3 is primarily written in Python.
- How popular is moco-v3?
- facebookresearch/moco-v3 has 1.3k stars on GitHub.
- Where can I find moco-v3?
- facebookresearch/moco-v3 is on GitHub at https://github.com/facebookresearch/moco-v3.