Keras.js – Run trained Keras models in your browser (opens in new tab)

On these demos, I'm getting several seconds on the imagenet inception v3 recognition on an i7 macbook pro (nvidia gpu), on both gpu and cpu modes.

I've built tensorflow for android, running inceptionv3 trained on imagenet and it's much faster, running just on mobile CPU pretty much realtime, around 5fps. On a desktop CPU/GPU it's obviously even faster

You mean with tensorflow or theano as the backend? They have all kinds of optimizations that isn't possible to replicate here yet. There is certainly room for optimization! Also, 1000-layer resnets should theoretically be possible, but probably isn't that practical. Lots of exciting work happening in searching for more efficient architectures.

1000 layer networks aren't used in practice. 50 layers is enough for state-of-the art performance.

It sounds like inference (prediction) performance is ok (probably < 1 second for images with ResNet50).

I doubt training performance would be very fun.

Thanks! For WebGL, credit goes to https://github.com/waylonflinn/weblas. I only really use GEMM, but it works quite well. In keras.js, convolution is implemented with the oft-used im2col transformation to turn it into a matrix multiply followed by reshape. Convolution kernels directly GLSL could potentially provide speed gains I'm sure, but I can't even imagine writing it for tensors of arbitrary shape.

They mentioned they used the weblas library/module.

See code for example https://github.com/transcranial/keras-js/blob/master/src/Ten... or code on weblas github repo.

They're using pre trained models though, like

https://github.com/heuritech/convnets-keras

I don't think they expect people to train them in the browser, just run the pretrained ones for image recognition or something

I agree. These are very marginal computation costs.

This might have the most value as a Node transport though.

Thanks. Nothing fancy with the network architecture diagrams. The layers are just div elements with the layer name as the id. There's a definition of inbound/outbound connections by layer names, extracted from the keras json config, which is used to draw SVG paths.