Instructions to use abduazizovanozima7/uzbek-trocr-final-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use abduazizovanozima7/uzbek-trocr-final-v2 with Transformers:

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

pipe = pipeline("image-text-to-text", model="abduazizovanozima7/uzbek-trocr-final-v2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("abduazizovanozima7/uzbek-trocr-final-v2")
model = AutoModelForMultimodalLM.from_pretrained("abduazizovanozima7/uzbek-trocr-final-v2")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use abduazizovanozima7/uzbek-trocr-final-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "abduazizovanozima7/uzbek-trocr-final-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "abduazizovanozima7/uzbek-trocr-final-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/abduazizovanozima7/uzbek-trocr-final-v2

SGLang

How to use abduazizovanozima7/uzbek-trocr-final-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 "abduazizovanozima7/uzbek-trocr-final-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": "abduazizovanozima7/uzbek-trocr-final-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 "abduazizovanozima7/uzbek-trocr-final-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": "abduazizovanozima7/uzbek-trocr-final-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use abduazizovanozima7/uzbek-trocr-final-v2 with Docker Model Runner:
```
docker model run hf.co/abduazizovanozima7/uzbek-trocr-final-v2
```

uzbek-trocr-final-v2

This model is a fine-tuned version of microsoft/trocr-base-handwritten on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 3.0923
Cer: 0.7650

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: 3e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
4.3617	1.7857	100	4.1494	0.8327
3.4918	3.5714	200	3.5616	0.7861
2.9422	5.3571	300	3.3129	0.7798
2.5864	7.1429	400	3.1919	0.7742
2.3168	8.9286	500	3.1471	0.7649
2.0836	10.7143	600	3.0981	0.7618
1.8460	12.5	700	3.0959	0.7625
1.5391	14.2857	800	3.0745	0.7669
1.5876	16.0714	900	3.0640	0.7609
1.4246	17.8571	1000	3.0842	0.7617
1.3221	19.6429	1100	3.0923	0.7650

Framework versions

Transformers 5.0.0
Pytorch 2.9.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

Downloads last month: 4

Safetensors

Model size

0.3B params

Tensor type

F32

Inference Providers NEW

Image-Text-to-Text

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

Model tree for abduazizovanozima7/uzbek-trocr-final-v2

Base model

microsoft/trocr-base-handwritten

Finetuned

(36)

this model