wassname/keras-js

mirror of https://github.com/wassname/keras-js.git synced 2026-09-22 13:10:34 +08:00

T

Leon Chen 741a5887a1 update mnist convnet demo

2016-09-19 15:57:17 -04:00

update mnist convnet demo

2016-09-19 15:57:17 -04:00

remove gpu.js

2016-08-26 23:53:13 -04:00

implement Embedding layer, with tests

2016-09-13 09:51:48 -04:00

modify Model class instance arguments

2016-09-19 15:56:06 -04:00

modify enable gpu flag

2016-09-19 11:00:21 -04:00

.gitignore

update mnist convnet demo

2016-09-19 15:57:17 -04:00

encoder.py

update encoder.py

2016-09-17 23:40:34 -04:00

index.html

implement Embedding layer, with tests

2016-09-13 09:51:48 -04:00

LICENSE

initial commit

2016-08-19 23:02:21 -04:00

package.json

use postcss with autoprefixer for managing demos CSS

2016-09-19 11:36:23 -04:00

README.md

start implementation of activations

2016-08-21 21:47:40 -04:00

webpack.dev.config.js

enable async and update webpack config

2016-09-16 15:55:28 -04:00

webpack.prod.config.js

enable async and update webpack config

2016-09-16 15:55:28 -04:00

README.md

Keras.js

Run trained Keras models in your browser, GPU-powered using WebGL.

Why?

You have trained a cool and useful Keras model. Instead of setting up backend servers to process jobs and send the results to your users, you want your model to run as close to your user as possible.

Can I do training?

No. Why would you even want to?

License

MIT