Instructions to use yrrhall/arabic-nano-gpt-v0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yrrhall/arabic-nano-gpt-v0 with Transformers:

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

pipe = pipeline("text-generation", model="yrrhall/arabic-nano-gpt-v0")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("yrrhall/arabic-nano-gpt-v0")
model = AutoModelForCausalLM.from_pretrained("yrrhall/arabic-nano-gpt-v0", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yrrhall/arabic-nano-gpt-v0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yrrhall/arabic-nano-gpt-v0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yrrhall/arabic-nano-gpt-v0",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/yrrhall/arabic-nano-gpt-v0

SGLang

How to use yrrhall/arabic-nano-gpt-v0 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 "yrrhall/arabic-nano-gpt-v0" \
    --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": "yrrhall/arabic-nano-gpt-v0",
		"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 "yrrhall/arabic-nano-gpt-v0" \
        --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": "yrrhall/arabic-nano-gpt-v0",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use yrrhall/arabic-nano-gpt-v0 with Docker Model Runner:
```
docker model run hf.co/yrrhall/arabic-nano-gpt-v0
```

arabic-nano-gpt

This model is a fine-tuned version of openai-community/gpt2 on the arabic wikimedia/wikipedia dataset.

Repository on GitHub: e-hossam96/arabic-nano-gpt

The model achieves the following results on the held-out test set:

Loss: 3.28796

How to Use

import torch
from transformers import pipeline

model_ckpt = "e-hossam96/arabic-nano-gpt-v0"
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")


lm = pipeline(task="text-generation", model=model_ckpt, device=device)

prompt = """المحرك النفاث هو محرك ينفث الموائع (الماء أو الهواء) بسرعة فائقة \
لينتج قوة دافعة اعتمادا على مبدأ قانون نيوتن الثالث للحركة. \
هذا التعريف الواسع للمحركات النفاثة يتضمن أيضا"""

output = lm(prompt, max_new_tokens=128)

print(output[0]["generated_text"])

Model description

Embedding Size: 256
Attention Heads: 4
Attention Layers: 4

Training and evaluation data

The entire wikipedia dataset was split into three splits based on the 90-5-5 ratios.

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.001
train_batch_size: 64
eval_batch_size: 64
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 256
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.01
num_epochs: 24

Training Loss

Validation Loss

Framework versions

Transformers 4.45.2
Pytorch 2.5.0
Datasets 3.0.1
Tokenizers 0.20.1

Downloads last month: 260

Safetensors

Model size

5.52M params

Tensor type

F32

Model tree for yrrhall/arabic-nano-gpt-v0

Base model

openai-community/gpt2

Finetuned

(2251)

this model

Duplicated from e-hossam96/arabic-nano-gpt-v0

yrrhall
/

arabic-nano-gpt-v0