Instructions to use not-lain/finetuned_deepseek_ocr with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use not-lain/finetuned_deepseek_ocr with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-OCR")
model = PeftModel.from_pretrained(base_model, "not-lain/finetuned_deepseek_ocr")

Transformers

How to use not-lain/finetuned_deepseek_ocr with Transformers:

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

pipe = pipeline("text-generation", model="not-lain/finetuned_deepseek_ocr", trust_remote_code=True)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("not-lain/finetuned_deepseek_ocr", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use not-lain/finetuned_deepseek_ocr with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "not-lain/finetuned_deepseek_ocr"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "not-lain/finetuned_deepseek_ocr",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/not-lain/finetuned_deepseek_ocr

SGLang

How to use not-lain/finetuned_deepseek_ocr 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 "not-lain/finetuned_deepseek_ocr" \
    --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": "not-lain/finetuned_deepseek_ocr",
		"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 "not-lain/finetuned_deepseek_ocr" \
        --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": "not-lain/finetuned_deepseek_ocr",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use not-lain/finetuned_deepseek_ocr with Docker Model Runner:
```
docker model run hf.co/not-lain/finetuned_deepseek_ocr
```

finetuned_deepseek_ocr

This model is a fine-tuned version of deepseek-ai/DeepSeek-OCR on an unknown dataset.

Model description

More information needed

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: 0.0005
train_batch_size: 4
eval_batch_size: 8
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
training_steps: 8

Training results

Framework versions

PEFT 0.17.1
Transformers 4.46.3
Pytorch 2.6.0+cu124
Datasets 4.3.0
Tokenizers 0.20.3

Downloads last month: 3

Safetensors

Model size

3B params

Tensor type

I64

BF16

Model tree for not-lain/finetuned_deepseek_ocr

Base model

deepseek-ai/DeepSeek-OCR

Adapter

(27)

this model