NightPrince/Arabic-Poetry-FineTuned

Arabic Poetry Fine-Tuned Model

This model is a fine-tuned version of the GPT-2 model, specifically trained on Arabic poetry. It is designed to generate Arabic poetry and can be used for creative writing, educational purposes, or research in natural language processing.

Try the Model

You can try the model directly in this interactive demo:

Model Details

• Model Type: GPT-2
• Language: Arabic
• License: Apache-2.0
• Author: NightPrince

Setup

You can run the smashed model with these steps:

0. Check requirements from the original repo NightPrince/Arabic-Poetry-FineTuned installed. In particular, check python, cuda, and transformers versions.
1. Make sure that you have installed quantization related packages.

   pip install transformers accelerate bitsandbytes>0.37.0
   

2. Load & run the model.

   from transformers import AutoModelForCausalLM, AutoTokenizer
   
   
   model = AutoModelForCausalLM.from_pretrained("PrunaAI/NightPrince-Arabic-Poetry-FineTuned-bnb-8bit-smashed", trust_remote_code=True, device_map='auto')
   tokenizer = AutoTokenizer.from_pretrained("NightPrince/Arabic-Poetry-FineTuned")
   
   input_ids = tokenizer("حدثني عن الحب في زمن الحرب,", return_tensors='pt').to(model.device)["input_ids"]
   
   outputs = model.generate(input_ids, max_new_tokens=216)
   tokenizer.decode(outputs[0])
   

Configurations

The configuration info are in smash_config.json.

Credits & License

The license of the smashed model follows the license of the original model. Please check the license of the original model NightPrince/Arabic-Poetry-FineTuned before using this model which provided the base model. The license of the pruna-engine is here on Pypi.

Intended Use

This model is intended for generating Arabic poetry. It can be used in applications such as:

• Creative writing tools
• Educational resources for learning Arabic poetry
• Research in natural language processing and generation

Training Data

The model was fine-tuned on a dataset of Arabic poetry. The dataset includes works from various poets and covers a range of styles and themes.

Training Procedure

• Framework: PyTorch
• Hardware: Trained on a GPU
• Epochs: 5
• Batch Size: [8]
• Learning Rate: [private]

Evaluation

The model was evaluated based on its ability to generate coherent and stylistically appropriate poetry. The training loss achieved was approximately 2.67, indicating a good level of learning.

Limitations and Biases

As with any language model, this model may generate biased or inappropriate content. Users should be aware of these limitations and use the model responsibly.

Acknowledgements

This model was developed by NightPrince and is hosted on Hugging Face. Special thanks to the creators of the original GPT-2 model and the Hugging Face team for their support.

Contact

For questions or feedback, please contact NightPrince via Hugging Face.