Instructions to use sam749/donut-base-finetuned-sroie-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sam749/donut-base-finetuned-sroie-v2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="sam749/donut-base-finetuned-sroie-v2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("sam749/donut-base-finetuned-sroie-v2")
model = AutoModelForMultimodalLM.from_pretrained("sam749/donut-base-finetuned-sroie-v2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use sam749/donut-base-finetuned-sroie-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sam749/donut-base-finetuned-sroie-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sam749/donut-base-finetuned-sroie-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/sam749/donut-base-finetuned-sroie-v2

SGLang

How to use sam749/donut-base-finetuned-sroie-v2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "sam749/donut-base-finetuned-sroie-v2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sam749/donut-base-finetuned-sroie-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "sam749/donut-base-finetuned-sroie-v2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sam749/donut-base-finetuned-sroie-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use sam749/donut-base-finetuned-sroie-v2 with Docker Model Runner:
```
docker model run hf.co/sam749/donut-base-finetuned-sroie-v2
```

donut-base-finetuned-sroie-v2

This model is a fine-tuned version of naver-clova-ix/donut-base on an sam749/SROIE-donut dataset.

Use

from transformers import DonutProcessor, VisionEncoderDecoderModel

device = "cuda" if torch.cuda.is_available() else "cpu"
dtype = torch.float16 if torch.cuda.is_available() else torch.float32

processor = DonutProcessor.from_pretrained("sam749/donut-base-finetuned-sroie-v2")
model = VisionEncoderDecoderModel.from_pretrained("sam749/donut-base-finetuned-sroie-v2", dtype=dtype)
model.to(device)

def generate(image):
    # prepare encoder inputs
    pixel_values = processor(image, return_tensors="pt").pixel_values
    
    # generate answer
    outputs = model.generate(
        pixel_values.to(device),
        use_cache=True,
        num_beams=1,
        bad_words_ids=[[processor.tokenizer.unk_token_id]],
        return_dict_in_generate=True,
    )
    
    # postprocess
    sequence = processor.batch_decode(outputs.sequences)[0]
    sequence = sequence.replace(processor.tokenizer.eos_token, "").replace(processor.tokenizer.pad_token, "")
    sequence = re.sub(r"<.*?>", "", sequence, count=1).strip()  # remove first task start token
    
    return processor.token2json(sequence)

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 1
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 3
mixed_precision_training: Native AMP

Training results

Framework versions

Transformers 4.48.1
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0

Downloads last month: 40

Safetensors

Model size

0.2B params

Tensor type

I64

F16

Inference Providers NEW

Image-Text-to-Text

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for sam749/donut-base-finetuned-sroie-v2

Base model

naver-clova-ix/donut-base

Finetuned

(486)

this model

Dataset used to train sam749/donut-base-finetuned-sroie-v2

Space using sam749/donut-base-finetuned-sroie-v2 1

Collection including sam749/donut-base-finetuned-sroie-v2

Multimodal

Collection

3 items • Updated Apr 30