Instructions to use scrollprize/timesformer_GP_scroll1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use scrollprize/timesformer_GP_scroll1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="scrollprize/timesformer_GP_scroll1", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("scrollprize/timesformer_GP_scroll1", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download timesformer_config.py from scrollprize/timesformer_GP_scroll1: direct link, hf CLI and curl.
- Browser
- Download file 600 Bytes
-
https://huggingface.co/scrollprize/timesformer_GP_scroll1/resolve/main/timesformer_config.py
- Command line
-
hf download hf://scrollprize/timesformer_GP_scroll1/timesformer_config.py
-
curl -L -o timesformer_config.py https://huggingface.co/scrollprize/timesformer_GP_scroll1/resolve/main/timesformer_config.py
600 Bytes
| from transformers import PretrainedConfig | |
| class TimesformerScrollprizeConfig(PretrainedConfig): | |
| # model_type = "timesformerscrollprize" | |
| def __init__( | |
| self, | |
| window_size=64, | |
| depth=8, | |
| n_heads=6, | |
| patch_size=16, | |
| num_frames=26, | |
| num_classes=16, | |
| dim=512, | |
| **kwargs, | |
| ): | |
| self.window_size=window_size | |
| self.depth=depth | |
| self.n_heads=n_heads | |
| self.patch_size=patch_size | |
| self.num_frames=num_frames | |
| self.num_classes=num_classes | |
| self.dim=dim | |
| super().__init__(**kwargs) |