Instructions to use OpenGVLab/pvt_v2_b2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use OpenGVLab/pvt_v2_b2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-classification", model="OpenGVLab/pvt_v2_b2") pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png")# pip install -U transformers accelerate # Load model directly from transformers import AutoImageProcessor, AutoModelForImageClassification processor = AutoImageProcessor.from_pretrained("OpenGVLab/pvt_v2_b2") model = AutoModelForImageClassification.from_pretrained("OpenGVLab/pvt_v2_b2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Fixed model name/path in config
Verified signature5395c47 -
Added first encoder layer to "out_indices" and "out_features"
Verified signatureThis will lead to better results when using PVTv2 as a backbone with default settings in a model like Deformable DETR.a2781b1 -
Fixed "out_indices" and "out_features" in config
Verified signatureThese fields were previously being stored incorrectly as "_out_features" and "_out_indices" with the underscore prefix. This was an error, and they should be stored without the prefix as "out_indices" and "out_features" for proper loading as a backbone model.0e9c438
-
Update README.md
Verified signaturee787234 -
Upload folder using huggingface_hub
Verified signaturefb44b0d -
initial commit
Verified signatureb691362