This repo contains the content related to the talk:
Construyendo soluciones con Generative AI para la transformación digital
Content
New Features
- Added model arguments control
I wanted to showcase the true value of GenAI applications, and the best way was to create a chatbot application for the entire conference.
Build an application to support event's participants/attendees.
Increase participants/attendees satisfaction (CSAT = customer satisfaction).
- GUI (graphical interface).
- Conference Knowledge Base.
- Personalization (unique experience for each user).
Maximize participants/attendees experience.
- Langchain
- Streamlit
- Weaviate
- OpenAI
Collected all text information for each session in the calendar.
In regard to vector stores, I selected weaviate since they are open source and provide a great free tier.
To create an account visit: Weaviate - Homepage
I selected OpenAI as the LLM service provider, since they offer a very good price/quality ratio.
- gpt-3.5-turbo
- text-embedding-ada-002
For reference, around 750 words predicted by gpt-3.5-turbo
costs 0.002 USD.
Create the virtual environment with:
python -m venv .venv
Then, activate it with:
source .venv/bin/activate
Finally, install dependencies with:
pip install -r requirements.txt
To run the streamlit application locally, use:
streamlit run app.py
This will automatically open a window in your default web browser.
Pre-commit is already part of this project dependencies. If you would like to installed it as standalone run:
pip install pre-commit
To activate pre-commit run the following commands:
- Install Git hooks:
pre-commit install
- Update current hooks:
pre-commit autoupdate
To test your installation of pre-commit run:
pre-commit run --all-files