Note, that size of the MobileNet weights is ~16MB.
- About TFjs
- Image classification
- WebGL procedural flowers
- MobileNet Heat Maps CAM
- MobileNet Transfer Learning
- Object detection
- ToDo: Small object (<30% of the full image) classification/detection failure
- Weight quantization
- How to use TFjs converter
Weight quantization demo on Windows 10. Quantization-Aware models. Unsupported Ops error!
To use ML applications (with GPU support) you need only browser and simple editor.
"Pure JS" TFjs links (no node.js):
Learning For Front-End Developers With Tensorflow.js.
Tips for Training your own Tensorflow.js Models in the Browser
Explore TF.js Gallery too.
But Layers models may be highly optimized by TFjs converter
Therefore we need not only editor?
(I usually need "python -m SimpleHTTPServer" or "python -m http.server 8000" too.)
Image classification applications are tested in desktop Chrome and Firefox Beta (32bit) or Dev (64bit)
+ Windows 10.
Simulations on GPU
"demidov at ipm.sci-nnov.ru"
updated 29 Jan 2020