Closed Beckschen closed 1 month ago
Add the ViTamin model, which is trained on public DataComp-1B using OpenCLIP framework and obtains 82.9% zero-shot ImageNet-1K accuracy with 436M parameters.
This work has been accepted to CVPR 2024 (https://arxiv.org/abs/2404.02132).
Close this PR since @rwightman merged the ViTamin model_configs in https://github.com/mlfoundations/open_clip/pull/886/commits/f100adc91a1ebb0845a9108668b4d1af0bcf6f64
Thanks, @rwightman
Add the ViTamin model, which is trained on public DataComp-1B using OpenCLIP framework and obtains 82.9% zero-shot ImageNet-1K accuracy with 436M parameters.
This work has been accepted to CVPR 2024 (https://arxiv.org/abs/2404.02132).