Buckets:
| """Load a single model and print all missing and unexpected weights. | |
| Swap AutoModel for AutoBackbone / Tipsv2VisionModel / Tipsv2TextModel and | |
| change REPO as needed. | |
| """ | |
| from transformers import AutoModel, AutoBackbone, Tipsv2VisionModel, Tipsv2TextModel, AutoModelForDepthEstimation, AutoModelForNormalEstimation, AutoModelForSemanticSegmentation, Tipsv2DptModel, Tipsv2DptForDepthEstimation, Tipsv2DptForNormalEstimation, Tipsv2DptForSemanticSegmentation | |
| REPO = "google/tipsv2-b14-dpt" | |
| REVISION = "refs/pr/1" | |
| model, loading_info = AutoModel.from_pretrained( | |
| REPO, | |
| trust_remote_code=True, | |
| output_loading_info=True, | |
| ) | |
| print(f"Loaded {REPO} -> {type(model).__name__}") | |
Xet Storage Details
- Size:
- 685 Bytes
- Xet hash:
- 893232f28a8614699c1016c298f73cae755641659368237e1de81807237dbe6b
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.