Image-Text-to-Text
PEFT
Safetensors
English
lora
qlora
4-bit precision
adapter
medical
radiology
image-captioning
image-to-text
llava-onevision
qwen2
vision-language
imageclef
rocov2
conversational
Instructions to use HoqueMahmudul/llava-onevision-7b-qlora-radiology-image-caption with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use HoqueMahmudul/llava-onevision-7b-qlora-radiology-image-caption with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("llava-hf/llava-onevision-qwen2-7b-ov-hf") model = PeftModel.from_pretrained(base_model, "HoqueMahmudul/llava-onevision-7b-qlora-radiology-image-caption") - Notebooks
- Google Colab
- Kaggle
Download tokenizer_config.json from HoqueMahmudul/llava-onevision-7b-qlora-radiology-image-caption: direct link, hf CLI and curl.
- Browser
- Download file 572 Bytes
-
https://huggingface.co/HoqueMahmudul/llava-onevision-7b-qlora-radiology-image-caption/resolve/main/tokenizer_config.json
- Command line
-
hf download hf://HoqueMahmudul/llava-onevision-7b-qlora-radiology-image-caption/tokenizer_config.json
-
curl -L -o tokenizer_config.json https://huggingface.co/HoqueMahmudul/llava-onevision-7b-qlora-radiology-image-caption/resolve/main/tokenizer_config.json
572 Bytes
| { | |
| "add_prefix_space": false, | |
| "backend": "tokenizers", | |
| "bos_token": null, | |
| "clean_up_tokenization_spaces": false, | |
| "eos_token": "<|im_end|>", | |
| "errors": "replace", | |
| "extra_special_tokens": [ | |
| "<|im_start|>", | |
| "<|im_end|>" | |
| ], | |
| "is_local": false, | |
| "max_length": null, | |
| "model_max_length": 131072, | |
| "pad_to_multiple_of": null, | |
| "pad_token": "<|endoftext|>", | |
| "pad_token_type_id": 0, | |
| "padding_side": "right", | |
| "processor_class": "LlavaOnevisionProcessor", | |
| "split_special_tokens": false, | |
| "tokenizer_class": "Qwen2Tokenizer", | |
| "unk_token": null | |
| } | |