Getting started with myHealth!

Requirements to run myHealth!

Poetry must be installed to run myHealth!

• Instructions on installing poetry

To run this project

1. Install poetry. Link to instructions.
2. Intialize virtual environment.

poetry shell

3. Install dependencies.

poetry install

4. Run

python myhealth/myHealth.py path/to/image.jpeg

• If you don't have an image and want to try out a demo, you may run the following command:

python myhealth/myHealth.py path/to/image.jpeg

If you really want to use a makefile.

• You will still need to install poetry.
• After you install poetry, you may install the dependencies with the following cmd: make install
• Run a demo with the following cmd: make run_demo

---

Project Structure:

data

• The data directory contains 2 NoSQL databases containing nutritional data.
• One contains a great deal of nutritional information on a few foods.
• The other is a database that contains only a little nutritional data on around 4,800 foods.
• These databases are TinyDB databases.
  • Link for more information on TinyDB

model

• The model directory contains:
  • Small CNN model used to classify images of food.
  • Images of the performance of the model.
  • A few images of foods to use for demos.

myhealth

• The myhealth directory contains the source code.

tests

• The tests directory contains unit tests.

poetry.lock

• This is a file containing the dependencies and their versions that are capatible and stable.

pyproject.toml

• This file contains the metadata for myHealth.

Using Poetry

Install dependencies

poetry install

Add dependencies

poetry add [package-name]

Activate Poetry Virtual Environment

poetry shell

Exit Poetry's Virtual Environment

exit

Run cmds

poetry run [cmd]

• Like black or pytest.

To build this project

poetry build

To publish this project

poetry publish

---

Typer

Typer is the tool we will use to create the CLI (command line interface).

• Getting started with Typer
  • A more in depth introduction
• Hints and docstrings
• Testing Typer apps
• Packaging Typer to publish

---

Model

Mini Model

• Used for development and demo purposes.
• Followed this tutorial (among a few others)
  • Used this tutorial in particular to save and load the model.

Performance

---

Data Sources

• AI Crowd Food Recognition 2022
• Food 101
• UEC FOOD 256