All the code needed to build the perfect toast machine, an AI-powered device that makes the perfect toast using odor.
A full walkthrough for building the device can be found here: How to Build an AI-powered Toaster
The Edge Impulse project used to train the machine learning model can be found here: perfect-toast-machine. You'll want to clone the project, go to Deployment, download the Arduino .zip library, and install it as a library in the Arduino IDE in order to run the perfect-toast-machine Arduino code.
If you want to eliminate some of the sensors, you can perform feature selection as shown in this tutorial.
- datasets/ - Data that I captured building the project. You are welcome to use it, but I can't promise it will work in your environment.
- images/ - Images used for this README.
- perfect-toast-machine/ - Arduino code used for deployment. You'll need to install the .zip Arduino library from the Edge Impulse project first.
- perfect-toast-machine-limited-sensors/ - Arduino code used for deployment. You'll need to install the .zip Arduino library from the Edge Impulse project first. Use this instead of perfect-toast-machine if you want to perform inference using just the multichannel gas sensor board and the BME680 board.
- tests/ - Internal tests that I used while making this project. You should not need to use them (unless you want to see how I test embedded machine learning code).
- toast-odor-data-collection/ - Run this code on your Arduino to collect raw data from the gas sensors. Used in tandem with serial-data-collect-csv.py to log raw data to CSV files to your computer.
- feature_selection_demo_iris.ipynb - Demonstration of feature selection using the iris dataset. Not needed for this project but could be instructive.
- ptm_dataset_curation.ipynb - Jupyter Notebook script (meant to be run on Google Colab) used to curate and standardize the dataset. Run this first after collecting your dataset before you upload it to Edge Impulse for training.
- ptm_dataset_curation.ipynb - Jupyter Notebook (for Google Colab) used to perform feature importance ranking and selection along with standardization of the dataset. Run this instead of ptm_dataset_curation.ipynb if you want to do feature selection.
- README.md - You're looking at it.
- serial-data-collect-csv.py - Python (v3+) script used to collect raw data from the toast-odor-data-collection Arduino program and log to CSV files. See this repo for an example on how to use the script.
All code, unless otherwise noted, is licensed under the Zero-Clause BSD (0BSD) license.
Permission to use, copy, modify, and/or distribute this software for any purpose with or without fee is hereby granted.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.