Neural network 3D visualization framework, build interactive and intuitive model in browsers, support pre-trained deep learning models from TensorFlow, Keras, TensorFlow.js
Previously, when loading a tensorflow frozen model in TensorSpace, it is required to provide two parameters: modelUrl and weightUrl. As TensorSpace uses tfjs's new loading API, weightUrl is no longer required, so new tensorflow load API will look like this temporary:
Previously, when loading a tensorflow frozen model in TensorSpace, it is required to provide two parameters:
modelUrl
andweightUrl
. As TensorSpace uses tfjs's new loading API,weightUrl
is no longer required, so new tensorflow load API will look like this temporary:Next step is to change
modelUrl
tourl
, maketensorflow loader
conform withkeras loader
andtfjs loader
.TODO: