Welcome to TaleForge, an immersive platform that harnesses the power of advanced language models to create spellbinding Urdu stories. Unleash your creativity and embark on an extraordinary journey of literary exploration, as TaleForge empowers you to shape captivating narratives that transcend boundaries and captivate readers. Step into the realm of Urdu storytelling and witness the magic unfold with every tale you weave.
TaleForge is an extraordinary machine learning endeavor that celebrates the art of Urdu storytelling. By leveraging cutting-edge language models, such as GPT-2, GPT-ForAll, GPT-Neo, BERT, or GPT-J, TaleForge emboldens you to create stories that transport readers to enchanting worlds, where imagination knows no bounds. This repository contains the complete framework for generating Urdu stories, from data collection to model training and fine-tuning.
- Immersive Data Collection: Embark on a journey of inspiration as TaleForge meticulously curates a diverse dataset of cues and Urdu stories from esteemed literary sources. Immerse yourself in a tapestry of cultural richness and authenticity, where each story springs from the heart of Urdu literature, breathing life into your creative vision.
- Fine-tuned Language Models: Forge your own path to storytelling mastery by selecting the perfect language model for your creative aspirations. Choose from GPT-2, GPT-ForAll, GPT-Neo, BERT, or GPT-J, each refined to embrace the nuances and beauty of the Urdu language. Let your chosen language model be the master craftsman that sculpts your stories into literary marvels.
- Thoughtful Training and Fine-tuning: Witness the evolution of storytelling as TaleForge's language models undergo meticulous training and fine-tuning. Through iterative refinement, our models grasp the intricacies of Urdu language and culture, honing their ability to generate stories that resonate deeply with readers. Forge narratives that reflect the essence of Urdu literature, shaped by the hands of a master storyteller.
- Culturally Sensational Story Generation: At TaleForge, we cherish and honor the cultural tapestry woven within Urdu storytelling. Each tale crafted by our models celebrates the heritage and values embedded in Urdu literature, ensuring that every narrative exudes cultural sensitivity and authenticity. Forge stories that pay homage to the rich traditions and aesthetic beauty of Urdu storytelling.
- Seamless Generation Experience: Unleash the magic of storytelling with TaleForge's seamless and intuitive story generation process. With a simple cue or prompt, your imagination ignites the forge, and the model weaves together words, emotions, and imagery into unforgettable tales. Prepare to embark on a transformative journey where every story becomes a masterpiece that mesmerizes and enthralls readers.
The TaleForge project directory is meticulously designed to ensure a smooth and immersive experience. Here's an overview:
├── LICENSE
├── Makefile <- Makefile with commands like `make data` or `make train`
├── README.md <- The top-level README for developers using this project. (you are here!)
├── data
│ ├── final <- The final, canonical data sets for modeling.
│ └── raw <- The original, immutable data dump.
│
├── models <- Trained and serialized models, model predictions, or model summaries.
│
├── notebooks <- Jupyter notebooks. Each notebook is named with a number for ordering
│
├── reports <- Generated analysis as HTML, PDF, LaTeX, etc.
│
├── requirements.txt <- The requirements file for reproducing the analysis environment.
│
└── src <- Source code for use in this project.
├── data <- Scripts to download or generate data.
│ └── make_dataset.py
├── features <- Scripts to transform raw data into features for modeling.
│ └── build_features.py
├── models <- Scripts to train and fine-tune models, and generate story predictions.
│ ├── predict_model.py
│ └── train_model.py
└── visualization <- Scripts to create exploratory and results-oriented visualizations.
└── visualize.py
Unlock the power of TaleForge by following the installation instructions below:
- Clone the repository:
git clone https://github.com/syedaddan/TaleForge.git
- Navigate to the project directory:
cd TaleForge
- Create a virtual environment (optional but recommended):
python3 -m venv env
- Activate the virtual environment:
source env/bin/activate
- Install the required dependencies:
pip install -r requirements.txt
TaleForge empowers you to create captivating Urdu stories that weave magic through every word. Follow these steps to embark on your storytelling journey:
-
Preprocess the dataset:
-
Place your raw cue and story data in the
data/raw
directory. -
Run the data preprocessing script:
python src/data/make_dataset.py
-
The preprocessed data will be saved in the
data/processed
directory.
-
-
Train and fine-tune the model:
-
Customize the model configuration in the
src/models/train_model.py
script to meet your creative vision. -
Run the model training script:
python src/models/train_model.py
-
The trained and fine-tuned model will be saved in the
models
directory.
-
-
Forge mesmerizing stories:
-
Customize the generation settings in the
src/models/predict_model.py
script. -
Run the story generation script:
python src/models/predict_model.py
-
The application will display or save the generated stories based on your configuration.
-
This project is licensed under the MIT License. For more details, refer to the LICENSE file.
TaleForge invites you to unleash your storytelling prowess and forge extraordinary Urdu tales that resonate with readers. Let the transformative power of language models, combined with your boundless imagination, shape narratives that leave an indelible mark. With TaleForge, immerse yourself in a world where words transcend boundaries, emotions come alive, and stories become timeless legacies. Embark on your journey today and discover the limitless possibilities of Urdu storytelling!