Instructions to use byczong/donut-ft-terms-struct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use byczong/donut-ft-terms-struct with Transformers:

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

pipe = pipeline("image-text-to-text", model="byczong/donut-ft-terms-struct")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("byczong/donut-ft-terms-struct")
model = AutoModelForMultimodalLM.from_pretrained("byczong/donut-ft-terms-struct", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use byczong/donut-ft-terms-struct with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/byczong/donut-ft-terms-struct

SGLang

How to use byczong/donut-ft-terms-struct 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 "byczong/donut-ft-terms-struct" \
    --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": "byczong/donut-ft-terms-struct",
		"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 "byczong/donut-ft-terms-struct" \
        --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": "byczong/donut-ft-terms-struct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use byczong/donut-ft-terms-struct with Docker Model Runner:
```
docker model run hf.co/byczong/donut-ft-terms-struct
```

Model Card

Donut fine-tuned for full document structuring (parsing) on pl-insurance-terms-struct dataset.

Trained for 10 epochs with max_seq_len=7168.

Field-level f1 score: 0.57
TED-based accuracy: 0.67

Note: This model and its tokenizer were not (pre-) trained for Polish.

Downloads last month: 26

Safetensors

Model size

0.2B params

Tensor type

I64

F32

Model tree for byczong/donut-ft-terms-struct

Base model

naver-clova-ix/donut-base

Finetuned

(488)

this model

byczong
/

donut-ft-terms-struct

Model Card

Model tree for byczong/donut-ft-terms-struct

Dataset used to train byczong/donut-ft-terms-struct