Instructions to use timm/vit_base_patch16_siglip_256.v2_webli with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- timm
How to use timm/vit_base_patch16_siglip_256.v2_webli with timm:
import timm model = timm.create_model("hf_hub:timm/vit_base_patch16_siglip_256.v2_webli", pretrained=True) - Transformers
How to use timm/vit_base_patch16_siglip_256.v2_webli with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-feature-extraction", model="timm/vit_base_patch16_siglip_256.v2_webli", device_map="auto")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("timm/vit_base_patch16_siglip_256.v2_webli", dtype="auto", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Difference in crop/resize mode between timm config and open_clip config
#1
by dhoa - opened
In timm’s config.json, I see "crop_mode": "center", but in open_clip I see "resize_mode": "squash" (which matches Google’s SigLIP config).
Maybe we need to update the crop/resize mode in timm?