Pandino

Overview

Pandino is a powerful tool designed to analyze and visualize data using various language models. It provides a flexible API for integrating different LLMs and processing data efficiently.

Features

• Supports multiple LLM types including Groq, Deepseek, Mistral, and OpenAI.
• Provides a Flask-based API for easy integration and usage.
• Handles data processing and visualization using popular libraries like pandas and pandasai.
• Implements a secure user management system using SQLite for API key validation.
• Uses environment variables for secure storage of API keys and encryption keys.

Installation

To install Pandino, follow these steps:

1. Clone the repository:

   git clone git@github.com:tulas75/pandino.git
   cd pandino

2. Install the required dependencies:

   pip install -r requirements.txt

3. Set up environment variables: Create a .env file in the project root and add the following variables:

   GROQ_API_KEY=your_groq_api_key
   DEEPSEEK_API_KEY=your_deepseek_api_key
   DEEPINFRA_API_KEY=your_deepinfra_api_key
   MISTRAL_API_KEY=your_mistral_api_key
   OPENAI_API_KEY=your_openai_api_key
   ENCRYPTION_KEY=your_encryption_key_for_database
   

Troubleshooting

If you encounter any issues during installation or usage, please check the following:

• Ensure all environment variables are correctly set in the .env file.
• Verify that all required dependencies are installed by running pip install -r requirements.txt again.
• If you encounter issues with API keys, ensure they are valid and not expired.

Usage

User Management with SQLite or Postgres (Use database_sqlite.py or database_pg.py accordingly with your choice)

Pandino includes a secure user management system using SQLite and Postgres. The system allows you to add users and their API keys to the database.

Initializing the Database

To initialize the SQLite database, use the following command:

python database_sqlite.py init_db

Adding Users

To add a new user to the database:

python database_sqlite.py add_user <username> <api_key>

Listing Users

To list all users in the database:

python database_sqlite.py list_users

Viewing Stored API Keys

To view all stored (encrypted) API keys:

python database_sqlite.py print_keys

Running the Pandino API Service

To run the Pandino API service, use the following command:

python main.py

Accessing the API

To access the /startdatachat endpoint using curl, use the following command:

curl -X POST "http://127.0.0.1:5000/startdatachat" \
     -H "Content-Type: multipart/form-data" \
     -H "X-API-KEY: your_api_key_here" \
     -H "X-USER-NAME: your_full_user_name_here" \  
     -F "model_name=llama-3.1-70b-versatile"
     -F "llm_type=Groq"
     -F "file=your_local_csv"
     
# curl -X POST "http://127.0.0.1:5000/startdatachat" \
#      -H "Content-Type: multipart/form-data" \
#      -H "X-API-KEY: your_api_key_here" \
#      -H "X-USER-NAME: your_full_user_name_here" \
#      -d '{
#          "model_name": "llama-3.1-70b-versatile",
#          "llm_type": "Groq",
#          "file": your_csv
#      }'

To access the /enddatachat endpoint using curl, use the following command:

curl -X POST "http://127.0.0.1:5000/enddatachat" \
     -H "Content-Type: application/json" \
     -H "X-API-KEY: your_api_key_here" \
     -H "X-USER-NAME: your_full_user_name_here" \
     -d '{}'

To access the /datachat endpoint using curl, use the following command:

curl -X POST "http://127.0.0.1:5000/datachat" \
     -H "Content-Type: application/json" \
     -H "X-API-KEY: your_api_key_here" \
     -d '{
         "chat": your_request_to_pandas_here
     }'

Replace your_api_key_here with a valid API key from the database, your_full_user_name_here with a user name (it will be used to create the agent dedicated export folder), your_request_to_pandas_here with your natural language request to Pandas and adjust the model_name, llm_type, and data fields as needed.

Contributing

Contributions are welcome! Please read the contributing guidelines before getting started. To report bugs or suggest features, please open an issue on the GitHub repository.

License

This project is licensed under the GPLv3 License. See the LICENSE file for details.