ml4d: machine learning for design

ml4d: machine learning for design is a design platform module conducted by @nkymut at the Division of Industrial Design SDE NUS during the 2017/2018 semester. Students have design&built interactive hardware prototypes to explore how the latest machine learning technologies can serve as a design tool to improve our of life.

The concept of the platform was inspired by The Wekinator and Machine Learning for Musicians and Artists by Dr. Rebecca Fiebrink and ml4a by Gene Kogan.

This repository contains a collection of sample codes used during the platform. The codes are written in Python using cross-platform modules to enable seamless development between both Windows/Mac and Raspberry Pi. This feature allows rapid-prototyping of an interactive system using machine learning techniques.

Requirements

---

1. Python3
2. RaspberryPi 2 or newer (Raspbian Stretch)
3. OpenCV
4. pyglet
5. pyFirmata

Installation

---

1. Install Python 3.6

• [Win/Mac] Anaconda https://anaconda.org/download/
• [Raspberry Pi] sudo apt-get install python3.6

2. Install pyglet, python-osc, opencv-python, pyserial, pyFirmata modules.

[Win/Mac] start Anaconda Prompt and

pip install pyglet python-osc opencv-python pyserial pyfirmata

[Raspberry Pi]

sudo pip3 install pyglet python-osc opencv-python pyserial pyfirmata

3. Download the repository.

[Win/Mac]

git clone https://github.com/nkymut/ml4d.git
cd ml4d/week1_pyglet
python 01helloWorld.py

[Raspberry Pi]

git clone https://github.com/nkymut/ml4d.git
cd ml4d/week1_pyglet
python3 01helloWorld.py

ToDo:

---

• More detailed instructions
• Face Recognition with dlib https://github.com/davisking/dlib
• Object Recognition with pytorch-yolo-v3 https://github.com/ayooshkathuria/pytorch-yolo-v3

References

---

ml4a: http://ml4a.github.io/ Kadenze Machine Learning for Musicians and Artists: https://www.kadenze.com/courses/machine-learning-for-musicians-and-artists Wekinator: http://www.wekinator.org/ pyglet: https://bitbucket.org/pyglet/pyglet/wiki/Home python-osc: https://github.com/attwad/python-osc pyfirmata: https://github.com/tino/pyFirmata

Image Credits

---

These files are licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license. (rock.jpg, paper.jpg, scissors.jpg): Author: Sertion