What is trAIn?

trAIn is an innovative mobile app designed to revolutionize personal fitness by leveraging the power of AI to create tailored workout routines. This AI-powered workout planner analyzes user data, such as fitness goals, current physical condition, and past workout performance, to suggest personalized exercise plans that adapt over time. Whether a user is a beginner looking to start a fitness journey or an experienced athlete seeking to enhance their routine, trAIn delivers customized suggestions that cater to individual needs. The app's intelligent algorithms ensure that users are consistently challenged while minimizing the risk of injury, making it an ideal fitness companion for achieving long-term success.

Set Up Docker/PostgreSQL

Step 1.) Install Docker

• Download the appropriate version of docker desktop app onto your computer (for windows make sure its the right version chip arcitechture arm64 vs x86 for your device)

• Once downloaded, open the desktop app

Step 2.) Modify "/bin/setup_postgre.sh"

• Update the password variable to your desired value

• Once completed modify the .env file and to your updated password

Step 3.) Run ./bin/setup_postgre.sh

• This will install the appropriate version of the postgre docker container onto your computer

• The following should be the ending of the expected output if it worked

CONTAINER ID   IMAGE             COMMAND                  CREATED        STATUS                  PORTS                                              NAMES
6df75c658f04   postgres:17.0     "docker-entrypoint.s…"   1 second ago   Up Less than a second   0.0.0.0:5432->5432/tcp                             postgresql

• IMPORTANT: After you run, change password back to tempPassword in "/bin/setup_postgre.sh" before committing to repo

Set Up Python

Step 1.) Download Python3.12 or later

• This will be required for all of our api endpoints, including the initial setup for the database

Step 2.) Create a python virtual environment in the root project directory

• Run python -m venv env

Step 3.) Download the required libraries

• Run pip install -r requirements.txt

Step 4.) Create a .env file that contains the following content and modify tempPassword and my_new_databse to whatever values you'd like (password must match password in bin/setup_postgre.sh)

DB_HOST=localhost
DB_PORT=5432
DB_USER=postgres
DB_PASSWORD=tempPassword
DB_NAME=my_new_database

** NOTE MAKE SURE YOU'RE GIT IGNORE CONTAINS .env OR YOUR PASSWORD WILL LEAK

Step 5.) Run postgre_make_db.py

• execute python3 postgre_make_db.py, and if the output says Database 'my_new_database' created sucessfully congrats it worked!

Step 5.) Run data_to_db.py

• Populates database with workouts

Step 6.) Run docker exec -it my_postgres /bin/bash

• Enters you into a bash terminal of the docker container

Step 7.) Log into your postgres DB with psql -U postgres

Step 8.) Run the following commands to find the tables in your database

$ \c <name of your database>
$ \dt

• Theses commands will first select database and then list all tables in the database

Project File Hierarchy

bin/*

• Shell scripts to setup, run, and restart postgreSQL

db-setup/*

• SQL schema, python files setting up our database and tables
• postgre_make_db.py: creates our overarching database to hold all tables
• postgre_make_tb.py: creates user database
• data_to_db.py: converts json workout data to a table in our postgreSQL database
• free-exercise-db.json: the workout data used as context for the model
• workouts_schema.sql: the schema for the table created in data_to_db.py

web_api/*

• Python files for handling backend messaging
• api_endpoints.py: offers several API endpoints about sending user data, sending chat message, and login capabilities
• rag.py: offers all rag functions to generate workouts, replace workouts, or general chat capabilities
• helper_functions: general functions to help with other functions needed for login and rag
• examples/*.json: Example responses for different chat types for api_endpoints.py

sample_app/*

• Frontend application built using flutter