Instructions to use ytu-ce-cosmos/turkish-gpt2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ytu-ce-cosmos/turkish-gpt2 with Transformers:

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

pipe = pipeline("text-generation", model="ytu-ce-cosmos/turkish-gpt2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ytu-ce-cosmos/turkish-gpt2")
model = AutoModelForCausalLM.from_pretrained("ytu-ce-cosmos/turkish-gpt2")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ytu-ce-cosmos/turkish-gpt2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ytu-ce-cosmos/turkish-gpt2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ytu-ce-cosmos/turkish-gpt2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ytu-ce-cosmos/turkish-gpt2

SGLang

How to use ytu-ce-cosmos/turkish-gpt2 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 "ytu-ce-cosmos/turkish-gpt2" \
    --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": "ytu-ce-cosmos/turkish-gpt2",
		"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 "ytu-ce-cosmos/turkish-gpt2" \
        --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": "ytu-ce-cosmos/turkish-gpt2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ytu-ce-cosmos/turkish-gpt2 with Docker Model Runner:
```
docker model run hf.co/ytu-ce-cosmos/turkish-gpt2
```

turkish-gpt2

This is a Turkish GPT-2 model. GPT-2 is designed for text generation tasks, providing the ability to continue a given text snippet in a coherent and contextually relevant manner. Due to the diverse nature of the training data, which includes websites, books, and other text sources, this model can exhibit biases. Users should be aware of these biases and use the model responsibly.

Example Usage

from transformers import AutoTokenizer, GPT2LMHeadModel
from transformers import pipeline

model = GPT2LMHeadModel.from_pretrained("ytu-ce-cosmos/turkish-gpt2")
tokenizer = AutoTokenizer.from_pretrained("ytu-ce-cosmos/turkish-gpt2")

text_generator = pipeline('text-generation', model=model, tokenizer=tokenizer)
r = text_generator("Teknolojinin gelişimi hayatımızı önemli ölçüde etkiledi. ", max_length=100)
[{'generated_text': 'Teknolojinin gelişimi hayatımızı önemli ölçüde etkiledi. "Dijitalleşme" ile birlikte hayatımızın belirli bir parçası daha rahata ermeye başladı.'}]

Relevant information can be found in the paper.

Acknowledgments

Research supported with Cloud TPUs from Google's TensorFlow Research Cloud (TFRC). Thanks for providing access to the TFRC ❤️
Thanks to the generous support from the Hugging Face team, it is possible to download models from their S3 storage 🤗

Citation

@article{kesgin2024introducing,
  title={Introducing cosmosGPT: Monolingual Training for Turkish Language Models},
  author={Kesgin, H Toprak and Yuce, M Kaan and Dogan, Eren and Uzun, M Egemen and Uz, Atahan and Seyrek, H Emre and Zeer, Ahmed and Amasyali, M Fatih},
  journal={arXiv preprint arXiv:2404.17336},
  year={2024}
}

Contact

COSMOS AI Research Group, Yildiz Technical University Computer Engineering Department
https://cosmos.yildiz.edu.tr/
cosmos@yildiz.edu.tr

Downloads last month: 3,462

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for ytu-ce-cosmos/turkish-gpt2

Adapters

1 model

Finetunes

2 models

Dataset used to train ytu-ce-cosmos/turkish-gpt2

Spaces using ytu-ce-cosmos/turkish-gpt2 3

Paper for ytu-ce-cosmos/turkish-gpt2

Introducing cosmosGPT: Monolingual Training for Turkish Language Models

Paper • 2404.17336 • Published Apr 26, 2024 • 5