Image-Text-to-Text
PEFT
Safetensors
English
vehme
handwritten-mathematics
mathematical-reasoning
automated-grading
error-detection
error-localization
qwen2.5-vl
vision-language-model
grpo
Instructions to use phuongnt/VEHME-FERMAT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use phuongnt/VEHME-FERMAT with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("/home3/s20245636/lab/OCR/VEHM/instr_tuning/output/600steps/red") model = PeftModel.from_pretrained(base_model, "phuongnt/VEHME-FERMAT") - Notebooks
- Google Colab
- Kaggle
Download additional_config.json from phuongnt/VEHME-FERMAT: direct link, hf CLI and curl.
- Browser
- Download file 73 Bytes
-
https://huggingface.co/phuongnt/VEHME-FERMAT/resolve/main/additional_config.json
- Command line
-
hf download hf://phuongnt/VEHME-FERMAT/additional_config.json
-
curl -L -o additional_config.json https://huggingface.co/phuongnt/VEHME-FERMAT/resolve/main/additional_config.json
73 Bytes
| {"lora_dtype": "bfloat16", "lorap_lr_ratio": null, "lorap_emb_lr": 1e-06} |