Video Classification
Transformers
Safetensors
vjepa21
feature-extraction
video
vjepa
vjepa2
v-jepa-2.1
self-supervised
world-model
custom_code
Instructions to use apiantonio/vjepa2.1-vit-base-384 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use apiantonio/vjepa2.1-vit-base-384 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("video-classification", model="apiantonio/vjepa2.1-vit-base-384", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("apiantonio/vjepa2.1-vit-base-384", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Fix transformers 4.x/5.x compat, implement output_hidden_states/attentions and out_layers, fix hierarchical predictor input, add video processor
caed165 verified | { | |
| "auto_map": { | |
| "AutoVideoProcessor": "video_processing_vjepa21.VJEPA21VideoProcessor" | |
| }, | |
| "crop_size": { | |
| "height": 384, | |
| "width": 384 | |
| }, | |
| "do_center_crop": true, | |
| "do_normalize": true, | |
| "do_rescale": true, | |
| "do_resize": true, | |
| "image_mean": [ | |
| 0.485, | |
| 0.456, | |
| 0.406 | |
| ], | |
| "image_std": [ | |
| 0.229, | |
| 0.224, | |
| 0.225 | |
| ], | |
| "resample": 2, | |
| "rescale_factor": 0.00392156862745098, | |
| "return_metadata": false, | |
| "size": { | |
| "shortest_edge": 384 | |
| }, | |
| "video_processor_type": "VJEPA21VideoProcessor" | |
| } |