WD 1.4 SwinV2 Tagger V2

Supports ratings, characters and general tags.

Trained using https://github.com/SmilingWolf/SW-CV-ModelZoo.
TPUs used for training kindly provided by the TRC program.

Dataset

Last image id: 5944504
Trained on Danbooru images with IDs modulo 0000-0899.
Validated on images with IDs modulo 0950-0999.
Images with less than 10 general tags were filtered out.
Tags with less than 600 images were filtered out.

Validation results

v2.0: P=R: threshold = 0.3771, F1 = 0.6854

What's new

Model v2.1/Dataset v2:
Re-exported to work around an ONNXRuntime v1.17.1 bug.
Bumped the minimum ONNXRuntime version to >= 1.17.0.
Now timm compatible! Load it up and give it a spin using the canonical one-liner!
Exported to msgpack for compatibility with the JAX-CV codebase.
The batch dimension of the ONNX model is not fixed to 1 anymore. Now you can go crazy with batch inference.
No change to the trained weights themselves. There might be small prediction discrepancies across frameworks due to implementation details.

Model v2.0/Dataset v2:
Initial release.

Runtime deps

ONNX model requires onnxruntime >= 1.17.0

Final words

Subject to change and updates.
Downstream users are encouraged to use tagged releases rather than relying on the head of the repo.