A command-line tool for using RAG model on Local documents
Project description
FeynmanAI
FeynmanAI is a powerful command-line tool that leverages the Retrieval-Augmented Generation (RAG) model to interact with local documents. Whether you want to query your documents for specific information or quiz yourself on the contents, FeynmanAi offers an intuitive interface to get answers and enhance your understanding.
Features
- Interactive Query Loop: Ask questions and get answers based on your documents.
- Quiz Mode: Test your knowledge with random questions from your documents.
- Multi-format Support: Works with PDF, TXT, PPT, and DOCX files.
- Additional Options: Clear the screen after each query or enable text-to-speech for answers.
Installation
You can install FeynmanAi using pip:
pip install FeynmanAI
Usage
The FeynmanAI
package provides a command-line interface (CLI) that you can use to interact with the RAG model and your Local documents.
Interactive Query Loop
To start the interactive query loop, run the following command:
You can use either -d
or --document
flag:
FeynmanAI -d /path/to/your/document.pdf
Example :
The package includes sample documents to get you started :
FeynmanAI -d Sample_documents/Distributed.pdf
This command will load the specified document into the model's knowledge base and initiate the interactive query loop. In this mode, you can ask questions and receive answers derived from the loaded documents. Available commands include:
Enter your question (or 'quit' to exit, 'clear' to clear screen):
: Type your question, and the tool will retrieve an answer from the RAG model.'quit'
: Exit the interactive query loop.'clear'
: Clear the screen and continue the interactive session.
Quiz Mode
You can also enter quiz mode by using the -qz
or --quiz
flag:
In quiz mode, the tool will randomly select a topic from your documents and ask you a question related to that topic. You can type your answer, and the tool will provide feedback.
To exit quiz mode, type 'quit'
.
Example
FeynmanAI -d Sample_documents/Distributed.pdf -qz
Other Options
The FeynmanAI
package also provides the following additional options:
-cls
or--clear
: Clear the screen after each query in the interactive mode.-rd
or--read
: Read the answer out loud (requires text-to-speech capabilities on your system).
Troubleshooting
If you encounter any issues, please check the error messages and make sure you have the required dependencies installed. If the problem persists, feel free to open an issue on the project's GitHub repository.
Contributing
If you would like to contribute to the FeynmanAi project, please follow these steps:
- Fork the repository.
- Create a new branch for your feature or bug fix.
- Make your changes and commit them.
- Push your changes to your fork.
- Submit a pull request.
Please make sure to write tests for any new functionality you add, and ensure that all existing tests pass before submitting your pull request.
License
FeynmanAi is released under the MIT License.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for FeynmanAI-0.1.4-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 005d15cf9048e646d82df4fbbf87b7826eb8ea7adc48bae00fbb29993d64be16 |
|
MD5 | 91ad87d000a4b984953bebfd18cb055f |
|
BLAKE2b-256 | 21833f5db320d88cacad4e159c8eea03e3a25f8a66ec135211176b775492ce96 |