Hey_bit

Intelligent habit-forming application that helps you change your daily behaviour based on science, using Ai tools.

Hey_bit is an application for habit formation. It uses large language models and a text corpus to create a personalized training plan using RAG (retrieval augmented generation).

First, a user logs in to their Google Calendar and gives the application access to their calendar (we recommend creating a separate calendar in Google for this purpose). Next, the user selects from a list of topics and habits and specifies the level they are already at. The app then creates the first 7 events, which are always scheduled for 9 a.m., so that the user can move the event to a free time slot during the day. The user can provide feedback for each appointment.

After a week, the app collects the user's feedback and then creates a personalized second weekly schedule. The longer a user uses the app, the more personalized the Habit Trainer becomes.

Use the application:

Download the files.

Create an .env file with your GROQ access key e.g. GROQ_API_KEY=gsk_****

The Google Calendar API is currently in developer mode, which means that we still need Google's security check to register more than one email address. 
However, you can enable the service yourself on Google Cloud and store your credentials in 'data/credentials/calendar_credentials.json.' 
You can find good instructions here: https://developers.google.com/calendar/api/quickstart/python

1. run 01_Initialize_database.ipynb.
2. run 01_Initialize_RAG_from_pdf.ipynb or 01_Initialize_RAG_from_txt.ipynb if you want to load your own documents (textbooks) as a source of knowledge for your habit trainer.
3. run streamlit run main_start.py to register a user and choose your habit and level. A first training week is created
4. run python main_update.py after one week to collect user feedback and generate the next week of training
5. if you want to chat with your documents and ask more questions of habit formation, run streamlit run main_chat.py.

Environment

Use the requirements file to create a new environment for this task.

pyenv local 3.11.3
python -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt