Instructions to use aneforge/resnet-50 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ANEForge
How to use aneforge/resnet-50 with ANEForge:
# Run this model on the Apple Neural Engine, without CoreML. import aneforge as af net = af.load_resnet("aneforge/resnet-50") # BatchNorm folded into the preceding conv at load logits = net(pixels) # pixels: a preprocessed [1, 3, 224, 224] float32 batch -> [1, 1000] - Notebooks
- Google Colab
- Kaggle
Download preprocessor_config.json from aneforge/resnet-50: direct link, hf CLI and curl.
- Browser
- Download file 266 Bytes
-
https://huggingface.co/aneforge/resnet-50/resolve/main/preprocessor_config.json
- Command line
-
hf download hf://aneforge/resnet-50/preprocessor_config.json
-
curl -L -o preprocessor_config.json https://huggingface.co/aneforge/resnet-50/resolve/main/preprocessor_config.json
266 Bytes
| { | |
| "crop_pct": 0.875, | |
| "do_normalize": true, | |
| "do_resize": true, | |
| "feature_extractor_type": "ConvNextFeatureExtractor", | |
| "image_mean": [ | |
| 0.485, | |
| 0.456, | |
| 0.406 | |
| ], | |
| "image_std": [ | |
| 0.229, | |
| 0.224, | |
| 0.225 | |
| ], | |
| "resample": 3, | |
| "size": 224 | |
| } | |