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:
1from transformers import AutoModelForCausalLM, AutoTokenizer
234model = AutoModelForCausalLM.from_pretrained("PrunaAI/NightPrince-Arabic-Poetry-FineTuned-bnb-8bit-smashed", trust_remote_code=True, device_map='auto')5tokenizer = AutoTokenizer.from_pretrained("NightPrince/Arabic-Poetry-FineTuned")67input_ids = tokenizer("حدثني عن الحب في زمن الحرب,", return_tensors='pt').to(model.device)["input_ids"]89outputs = model.generate(input_ids, max_new_tokens=216)10tokenizer.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.
