Instructions to use atkinschang/TableFormer-MLX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use atkinschang/TableFormer-MLX with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download atkinschang/TableFormer-MLX --local-dir TableFormer-MLX
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Publish MLX-converted weights of docling-project/docling-models at fc0f2d45e2218ea24bce5045f58a389aed16dc23
28bb517 Download fast/preprocessor_config.json from atkinschang/TableFormer-MLX: direct link, hf CLI and curl.
- Browser
- Download file 553 Bytes
-
https://huggingface.co/atkinschang/TableFormer-MLX/resolve/main/fast/preprocessor_config.json
- Command line
-
hf download hf://atkinschang/TableFormer-MLX/fast/preprocessor_config.json
-
curl -L -o preprocessor_config.json https://huggingface.co/atkinschang/TableFormer-MLX/resolve/main/fast/preprocessor_config.json
553 Bytes
| { | |
| "antialias": false, | |
| "do_convert_rgb": true, | |
| "do_normalize": true, | |
| "do_rescale": true, | |
| "do_resize": true, | |
| "image_mean": [ | |
| 0.94247851, | |
| 0.94254675, | |
| 0.94292611 | |
| ], | |
| "image_std": [ | |
| 0.17910956, | |
| 0.17940403, | |
| 0.17931663 | |
| ], | |
| "interpolation": "bilinear", | |
| "normalize_before_resize": true, | |
| "output_dtype": "float32", | |
| "output_layout": "CWH", | |
| "page_height": 1024, | |
| "page_interpolation": "area", | |
| "resample": "bilinear", | |
| "rescale_factor": 0.00392156862745098, | |
| "size": { | |
| "height": 448, | |
| "width": 448 | |
| } | |
| } | |