How well does a decoder-only agent like GPT-3.5 handle user queries with different levels of proficiency in a conversational chatbot?
This README provides instructions on how to set up and run the application.
- Python
- Streamlit
- OpenAI API key
You can create a virtual environment to manage dependencies:
- python -m venv venv
- source venv/bin/activate # On Windows, use
venv\Scripts\activate
Alternatively, if you need to install pip, run:
- curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
- python get-pip.py
Install the necessary packages by running:
- pip install -r requirements.txt
- Add to a .env file to be used for connecting to these platforms.
- Copy the OpenAI api key and paste in the sidebar
Note: Due to size constraints, the BERT model is not included in this repo. To use this, run the following command to train the encoder for inference:
- python train_encoder.py
To chat with the decoder-only agent:
- streamlit run decoder_agent.py
To chat with the seq2seq (encoder-decoder) agent:
- streamlit run encoder_decoder_agent.py