google-research/big_transfer
A repository providing pre-trained Big Transfer (BiT) models for visual representation learning and multi-framework fine-tuning code.
Not currently ranked — collecting fresh signals.
star history
This repository releases pre-trained models from the Big Transfer (BiT) paper, trained on ILSVRC-2012 and ImageNet-21k datasets. It provides code to fine-tune these models across TensorFlow 2, PyTorch, and Jax/Flax frameworks. The models enable transfer learning for various computer vision tasks through high-quality visual representations learned via large-scale pre-training on ImageNet data.
Frequently asked
- What is google-research/big_transfer?
- A repository providing pre-trained Big Transfer (BiT) models for visual representation learning and multi-framework fine-tuning code.
- Is big_transfer open source?
- Yes — google-research/big_transfer is open source, released under the Apache-2.0 license.
- What language is big_transfer written in?
- google-research/big_transfer is primarily written in Python.
- How popular is big_transfer?
- google-research/big_transfer has 1.5k stars on GitHub.
- Where can I find big_transfer?
- google-research/big_transfer is on GitHub at https://github.com/google-research/big_transfer.