AI Model Integration
This application leverages the power of Google’s Gemini 1.5 Flash and Anthropic’s Claude 3 Opus language models to provide intelligent features such as:
Plot Suggestions: The selected AI model analyzes your uploaded dataset and suggests appropriate visualizations. * Gemini: See
utils.get_plot_suggestion_from_gemini(). * Claude: Seeutils.get_plot_suggestion_from_claude().Graph Interpretation: The selected AI model can provide textual interpretations of generated plots, helping you understand the insights revealed by your data. * Gemini: See
utils.generate_graph_interpretation_gemini(). * Claude: Seeutils.generate_graph_interpretation_claude().Interactive Chat: You can interact with the selected AI model to ask questions about your graphs and data. * Gemini: See
utils.handle_graph_communication_gemini(). * Claude: Seeutils.handle_graph_communication_claude().
The AI models are used within the following functions:
In utils.py: *
utils.get_plot_suggestion_from_gemini()*utils.generate_graph_interpretation_gemini()*utils.handle_graph_communication_gemini()*utils.get_plot_suggestion_from_claude()*utils.generate_graph_interpretation_claude()*utils.handle_graph_communication_claude()In app.py: *
app.generate_plots()*app.get_interpretation()*app.graph_chat()
API Key Configuration
To use the AI-powered features, you’ll need to obtain an API key from either Google (for Gemini) or Anthropic (for Claude) and provide it to the application. You will need to select which model you are using. Instructions for obtaining and configuring the API key will be provided in the [Getting Started] section (you need to add details about the API Key configuration to the getting_started.rst file.)
Model Selection
The user can select either Gemini or Claude as their preferred AI model for generating visualizations, interpretations, and interactive chat. The application will use the selected model for all subsequent AI-related operations. This choice is made through the user interface and passed to the backend functions.