StoryBuddy

An AI for Writing Stories

Writing Story Outlines

Begin by crafting the structure of your stories:

• Intro: Set the stage for your narrative.
• Body: Develop the plot through several sections.
• Conclusion: Bring your story to a satisfying end.

Extracting Keywords

• Pull out keywords from your story outlines.

Sourcing Books

• Search public domain websites for books matching your keywords.

Tagging Books

Use tags for data cleaning:

• <d></d>: Delete section.
• <a></a>: Author's name.
• <t></t>: Title of the work.
• <g></g>: Genre.
• <s></s>: Source of the book.
• <p></p>: Publication year.
• <c></c>: Content of the book.

Preparing the Text Files

• Place the plain text book files in the ../books directory.
• Ensure the files are in .txt format.

Generating K-Fold Datasets

• Execute _booktodataset_new_kfold.py script located in the scripts folder to create the k-fold training datasets.

Training the Model

• Once the dataset is prepared, start training the model.
• Run _run_build_model.py from the scripts folder. This script will reinitiate training if any interruptions occur.

Post-Training Steps

• Move the trained model to the services folder.

Launching the API

• Start the backend service by running python3 main.py.

Starting the Web Application

• Navigate to the StoryBuddyWeb folder.
• Launch the application with python3 app.py.

Publishing to story3.com

To add your story to story3.com:

• Prepare your story and twists in story.json, adhering to the provided structure.
• Rename the file to match the title of your story.
• Execute the python3 staticuploadstory.py script from the services directory.
• Follow the prompts to specify the path to your JSON file.
• Observe the console for errors during upload.
• Once uploaded, visit story3.com to officially publish your story.
• Share your story on various social media channels.