JARVIS is a Python program that generates responses to user input using OpenAI's natural language processing platform, along with Spacy and NLTK libraries. This code could serve as the foundation for building a chatbot or conversational AI.
To run JARVIS, you will need:
- An OpenAI API key
- Python 3.x Latest
- The following Python libraries: OpenAI, spacy, and nltk
- Clone the repository to your local machine.
- Replace
""with your OpenAI API key in the lineopenai.api_key = "". - Run the program in a terminal with the command
python jarvis.py. - JARVIS will prompt the user for input with the message "Hello, I'm Jarvis. How can I assist you today?".
- Enter your message and press Enter.
- JARVIS will generate a response based on your input and display it in the console.
To stop the program, press Ctrl+C.
You can customize JARVIS to fit your needs by modifying the model_engine and model_prompt variables. The model_engine variable specifies the OpenAI model to use for generating responses. The model_prompt variable contains the message that JARVIS will display to the user before each response. You can change this message to fit your use case.
Contributions are welcome to JARVIS! If you find a bug or want to suggest an improvement, please open an issue or pull request.
The latest update has made the following changes to the code:
- Define a single model prompt for generating responses.
- Use a fixed prompt for all user inputs.
- Generate only one response for each user input.
- Use a temperature of 0.5 for response generation.
- Remove the use of a stop sequence for response generation.
- Remove some unnecessary pre-processing steps for the prompt.
- Set the maximum token limit to 1024 for response generation.
- Set a timeout limit of 10 seconds for response generation.
- Do not remove non-ASCII characters from the generated response.