Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/pvscreations/sample-demo

https://github.com/hpssjellis/my-examples-of-webAI
https://github.com/pvscreations/sample-demo

Last synced: about 11 hours ago
JSON representation

https://github.com/hpssjellis/my-examples-of-webAI

Awesome Lists containing this project

README

        

# my-examples-of-webAI
WebAI as SPA's (single page applications) which are great to feed into LLM's like ChatGPT to make changes add CSS styling or check for security issues.

By Jeremy Ellis LinkedIn use at your own risk. Downloads ML Models which may encure data charges depending on your cell plan.

The github for this page is at github.com/hpssjellis/my-examples-of-webAI



  1. a-posenet.html A single file application (SPA) of the impressive webcam posenet


  2. b-xor.html An excellent chance for beginners to optimize a Keras model online. Make it faster and more accurate


  3. c-handpose.html Another cool example, this time using the hand.


  4. d-knn-classify.html A cool live train and run vision model for 3 labels. Best to use your face, look left, open mouth etc. If you don't like the results just train it more times.


  5. e-TinyMLjs.html An example of training a machine learning model using either your cell phone accelerometer or your desktop mouse motion in an SPA. This code is really to train an Arduino style microcontroller combining the physical power of TinyML with the simplicity and security of WebAI. This is my area of study ( Jeremy Ellis)


  6. f-sounds.html Fairly raw sounds. Train "Right", "Left" and background with minimal sounds. Can change the labels to anything. Watch the slider react to the sounds.
  7. Now try this prompt in your favorite LLM such as chatGPT. If it doesn't work check out the following code from chatGPT z-mobilenet-chatgpt.html or from copilot. They are very simlar z-mobilenet-copilot.html




| LLM Prompt |
|-----------------|
|Write a single file html application that uses tensorflowjs to load a mobilenet vision model and activates the webcam every 1000 ms to classify the image. Give the option for the rear or front camera.This code should be as simple as possible using standard javascript async and await, no jQuery use getElementById and also no EventListeners just normal buttons with onClick. Any CSS should be minimal and used in tags with inline style. Can you do that?|


If you want to view this page as a webpage click this link

hpssjellis.github.io/my-examples-of-webAI/ or use the QR code

![image](https://github.com/user-attachments/assets/51130e54-6215-4273-acf3-38cfeef35f52)

For a much more intense page made by Jeremy Ellis around 2018-2019 about TensorflowJS is at this link beginner-tensorflowjs-examples-in-javascript