README.md
[//]: #@corifeus-header
[](https://www.facebook.com/corifeus.software) [](https://paypal.me/patrikx3) [](https://www.patrikx3.com/en/front/contact) [](https://travis-ci.com/patrikx3/play-ai-ui)
[](https://uptimerobot.patrikx3.com/)
---
# 🤖 P3X Experimental AI / machine learning / tensors based UI playground v2019.2.8-1
🙏 This is an open-source project. Star this repository, if you like it, or even donate to maintain the servers and the development. Thank you so much!
Possible, this mini server, rarely, is down, please hang on for 15-30 minutes and the server will be back up.
All my domains ([patrikx3.com](https://patrikx3.com) and [corifeus.com](https://corifeus.com)) could have minor errors, since I am developing in my free time. However, it is usually stable.
**Bugs are evident™ - MATRIX️**
### Node Version Requirement
```
>=10.13.0
```
### Built on Node
```
v11.9.0
```
The ```async``` and ```await``` keywords are required.
Install NodeJs:
https://nodejs.org/en/download/package-manager/
# Description
[//]: #@corifeus-header:end
This App only works the latest browser - eg. there is no Babel and I am using ES2017 features. Besides, it not optimized for Android or Tablets, it is created for a minimum 1027x768 **'big'** browser
## Deployed playground
https://ai.play.patrikx3.com/
## Uses
* https://ml5js.org
* https://js.tensorflow.org/
* https://github.com/justadudewhohacks/face-api.js
* https://github.com/justadudewhohacks/face-api.js/issues/174
### Classification
https://groups.google.com/a/tensorflow.org/forum/?utm_medium=email&utm_source=footer#!msg/tfjs/Sc-r53YMyGc/IYL6Q_8zDQAJ
#### Blog post
https://itnext.io/face-api-js-javascript-api-for-face-recognition-in-the-browser-with-tensorflow-js-bcc2a6c4cf07
##### Github repo
https://github.com/justadudewhohacks/face-api.js
#### Retrain
https://github.com/adwellj/node-tfjs-retrain
### Functions
Webcam is required, as well a computer that works with WebGl.
* Recognize
* Recognize at most 3 objects in a video with a probability
* You can choose 2 objects and name them
* Capture an image from the webcam and assign it to one of the named object
* Make sure that every name object have at least 15 captures
* Then you can train the neural network
* Then when you can guess which named object is shown on the webcam right now
## How to run on your own localhost
```bash
git clone https://github.com/patrikx3/play-ai-ui
cd play-ai-ui
npm install
npm run dev
# open http://localhost:8080
```
## The programs
[Here are the programs](https://github.com/patrikx3/play-ai-ui/tree/master/src/angular/page)
[//]: #@corifeus-footer
---
[**P3X-PLAY-AI-UI**](https://pages.corifeus.com/play-ai-ui) Build v2019.2.8-1
[](https://www.facebook.com/corifeus.software) [](https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=QZVM4V6HVZJW6) [](https://www.patrikx3.com/en/front/contact)
## P3X Sponsors
[IntelliJ - The most intelligent Java IDE](https://www.jetbrains.com/?from=patrikx3)
[](https://www.jetbrains.com/?from=patrikx3) [](https://www.nosqlbooster.com/)
[The Smartest IDE for MongoDB](https://www.nosqlbooster.com)
[//]: #@corifeus-footer:end