Instructions to use intexcp/donut with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use intexcp/donut with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("image-to-text", model="intexcp/donut")# Load model directly from transformers import AutoTokenizer, AutoModelForMultimodalLM tokenizer = AutoTokenizer.from_pretrained("intexcp/donut") model = AutoModelForMultimodalLM.from_pretrained("intexcp/donut", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "crop_size": null, | |
| "data_format": "channels_first", | |
| "default_to_square": true, | |
| "device": null, | |
| "disable_grouping": null, | |
| "do_align_long_axis": false, | |
| "do_center_crop": null, | |
| "do_convert_rgb": null, | |
| "do_normalize": true, | |
| "do_pad": true, | |
| "do_rescale": true, | |
| "do_resize": true, | |
| "do_thumbnail": true, | |
| "image_mean": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "image_processor_type": "DonutImageProcessorFast", | |
| "image_std": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "input_data_format": null, | |
| "processor_class": "DonutProcessor", | |
| "resample": 2, | |
| "rescale_factor": 0.00392156862745098, | |
| "return_tensors": null, | |
| "size": { | |
| "height": 1600, | |
| "width": 1200 | |
| } | |
| } | |