Image-to-Text
Transformers
Safetensors
PEFT
Thai
DeepseekOCR2
feature-extraction
ocr
thai
thai-ocr
handwriting
unsloth
lora
merged
deepseek-ocr2
custom_code
Instructions to use phoritus/deepseek-ocr2-thai with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use phoritus/deepseek-ocr2-thai 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="phoritus/deepseek-ocr2-thai", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("phoritus/deepseek-ocr2-thai", trust_remote_code=True, device_map="auto") - PEFT
How to use phoritus/deepseek-ocr2-thai with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Unsloth Desktop
Download processor_config.json from phoritus/deepseek-ocr2-thai: direct link, hf CLI and curl.
- Browser
- Download file 460 Bytes
-
https://huggingface.co/phoritus/deepseek-ocr2-thai/resolve/main/processor_config.json
- Command line
-
hf download hf://phoritus/deepseek-ocr2-thai/processor_config.json
-
curl -L -o processor_config.json https://huggingface.co/phoritus/deepseek-ocr2-thai/resolve/main/processor_config.json
460 Bytes
| { | |
| "add_special_token": false, | |
| "candidate_resolutions": [ | |
| [ | |
| 1024, | |
| 1024 | |
| ] | |
| ], | |
| "downsample_ratio": 4, | |
| "ignore_id": -100, | |
| "image_mean": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "image_std": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "image_token": "<image>", | |
| "mask_prompt": false, | |
| "normalize": true, | |
| "pad_token": "<\uff5c\u2581pad\u2581\uff5c>", | |
| "patch_size": 16, | |
| "processor_class": "DeepseekVLV2Processor", | |
| "sft_format": "deepseek" | |
| } | |