StudySetCreator is a Python tool that generates study sets (flashcards) from PDF files using OpenAI's language models. It processes the content of a PDF file, extracts text and images, and uses the OpenAI API to create question-answer pairs suitable for studying or revision purposes. The generated study set is saved as a CSV file, ready to be imported into flashcard applications or used directly.

• PDF Processing: Extracts text and images from PDF files.
• OpenAI Integration: Utilizes OpenAI's GPT models to generate study cards from the extracted content.
• Batch Processing: Supports processing in chunks to handle large PDF files efficiently.
• Resume Capability: Can resume processing from where it left off in case of interruptions.
• Language Support: Generates study sets in the specified language.
• Customization: Allows customization of various parameters like model selection, output file name, chunk size, etc.

• Python: Version 3.7 or higher.
• OpenAI API Key: Required to access OpenAI's language models.

1. Clone the Repository

   git clone https://github.com/jaylann/StudySetCreator.git
   cd StudySetCreator

2. Create a Virtual Environment (optional but recommended)

   python3 -m venv venv
   source venv/bin/activate # On Windows use `venv\Scripts\activate`

3. Install Dependencies

   pip install -r requirements.txt

The application requires an OpenAI API key to function. This key should be stored in a .env file in the project's root directory.

1. Create the .env File

   There is a .env.template file provided in the project. Copy this template to create your .env file:

   cp .env.template .env

2. Edit the .env File

   Open the .env file in a text editor and add your OpenAI API key:

   OPENAI_API_KEY=your-openai-api-key-here

   Replace your-openai-api-key-here with your actual OpenAI API key.

Run the main.py script with the required arguments to generate a study set from a PDF file.

python main.py [options]

• --model: OpenAI model to use (default: gpt-4o-mini).
• --output: Output CSV file name (default: study_set.csv).
• --input: Input PDF file to process (required).
• --in_dir: Input directory containing PDF files to process.
• --out_dir: Output directory to save the study sets.
• --chunk_size: Number of pages to process at once (default: 10).
• --use_batch: Use OpenAI Batch API for processing.
• --text_only: Extract text only, ignore images.
• --language: Language for the study set (default: english).
• --no_resume: Whether to resume processing from the last checkpoint. WARNING: If set and a progress file exists, it will be overwritten.

1. Basic Usage

   Generate a study set from document.pdf using the default settings.

   python main.py --input document.pdf --output document.csv

2. Specify Output File and Model

   Generate a study set from lecture_notes.pdf, using the gpt-4o model, and save the output to flashcards.csv.

   python main.py --model gpt-4o --output flashcards.csv --input lecture_notes.pdf

3. Process Only Text Content

   Generate a study set ignoring images in the PDF.

   python main.py --text_only --input textbook.pdf --output textbook.csv

4. Use Batch Processing

   Use OpenAI's Batch API to process the PDF (suitable for large PDFs. Reduces cost by ~50% but may take longer).

   python main.py --use_batch --input large_document.pdf --output large_document.csv

5. Specify Language

   Generate a study set in Spanish.

   python main.py --language spanish --input notas_de_clase.pdf --output notas_de_clase.csv

The system prompt used by the OpenAI API can be customized to change how study cards are generated.

• Prompt File: ./storage/prompt.txt

  Edit this file to modify the prompt. The placeholder [LANGUAGE] in the prompt will be replaced with the language specified via the --language argument.

The JSON schema defines the expected structure of the API responses.

• Schema File: ./storage/schema.json

  Edit this file to change the schema if you need the responses in a different format.

The application uses logging to provide information about its operation.

• Log Output: The application outputs logs to the console. You can modify the logging configuration in src/utils/logging.py if you need to change log levels or output formats.

• Resume Processing: If the processing is interrupted, the application can resume from where it left off using the progress saved in progress.json.
• Progress File: The file progress.json is used to keep track of progress. It can be deleted to start processing from the beginning.
• Batch Processing Errors: If a batch job fails or is still in progress, an error message will be logged.

All required Python packages are listed in requirements.txt. Install them using:

pip install -r requirements.txt

• API Usage: Be mindful of your OpenAI API usage and billing.
• Supported Models: Ensure that the model you specify (e.g., gpt-4) is available to your OpenAI account.

Contributions are welcome! Please open an issue or submit a pull request for any bugs or feature requests.

This project is licensed under the MIT License.

Read more about this on my blog

Made with ❤️ by Justin Lanfermann