|
--- |
|
base_model: |
|
- openai-community/gpt2 |
|
datasets: |
|
- merve/poetry |
|
license: mit |
|
--- |
|
# Poem Generation Project |
|
|
|
Generate short, contextually relevant poems based on input themes or keywords (e.g., nature, love, seasons) using a fine-tuned GPT-2 model. This project uses the [Poetry Dataset by Merve](https://huggingface.co/datasets/merve/poetry) and the [GPT-2 model](https://huggingface.co/openai-community/gpt2) from Hugging Face. |
|
|
|
## Project Overview |
|
|
|
This project is designed to generate creative and stylistic short poems based on user-input themes or keywords. By fine-tuning a GPT-2 model on a curated poetry dataset, the model learns poetic structures and language styles, allowing it to generate verses relevant to a given theme. |
|
|
|
## Dataset |
|
|
|
- **Dataset**: [Poetry Dataset by Merve](https://huggingface.co/datasets/merve/poetry) |
|
- This dataset contains various poems, which are used to fine-tune the GPT-2 model for poem generation. |
|
|
|
## Model |
|
|
|
- **Pre-trained Model**: [GPT-2 by OpenAI](https://huggingface.co/openai-community/gpt2) |
|
- **Fine-Tuned Model**: This GPT-2 model has been fine-tuned specifically for poetry generation on the provided dataset. |
|
|
