|
Download README.md from Ashluu/us-ijepa-v3-encoder: direct link, hf CLI and curl.
- Browser
- Download file 668 Bytes
-
https://huggingface.co/Ashluu/us-ijepa-v3-encoder/resolve/main/README.md
- Command line
-
hf download hf://Ashluu/us-ijepa-v3-encoder/README.md
-
curl -L -o README.md https://huggingface.co/Ashluu/us-ijepa-v3-encoder/resolve/main/README.md
668 Bytes
| license: other | |
| tags: [medical-imaging, ultrasound, self-supervised, i-jepa, vit] | |
| # us-ijepa-v3-encoder | |
| I-JEPA ViT-B/16 encoder for breast ultrasound, from an MSc thesis. Encoder weights only | |
| (`encoder` state dict and the training `cfg`), 343 MB, float32. Research use; not a | |
| clinical tool. | |
| - Input: 224 x 224 grayscale ultrasound, repeated to 3 channels, scaled to [0, 1]. | |
| - Output: patch tokens `[B, 196, 768]`; the thesis probes use the mean over patches. | |
| - Frozen linear probe, BUS-BRA (patient-grouped split): validation AUROC 0.814, test 0.870 | |
| on whole frames. | |
| Used by the agent repository, which loads it from `checkpoints/us_ijepa_v3_encoder.pt`. | |