Neural network 3D visualization framework, build interactive and intuitive model in browsers, support pre-trained deep learning models from TensorFlow, Keras, TensorFlow.js
I tried to use Tensorspace on my own models trained with Keras. In this case the layers in Tensorspace depict the "Input" of the Keras layer. However, in your examples (LeNet, etc...) the Tensorspace layers depict the output of the model's layers. Can you precise which data the Tensorspace layers should depict?
If it is the output of the model's layers; what could be the origin of my Tensorspace showing the input data of the model's layers?
Dear,
I tried to use Tensorspace on my own models trained with Keras. In this case the layers in Tensorspace depict the "Input" of the Keras layer. However, in your examples (LeNet, etc...) the Tensorspace layers depict the output of the model's layers. Can you precise which data the Tensorspace layers should depict?
If it is the output of the model's layers; what could be the origin of my Tensorspace showing the input data of the model's layers?
Thanks for your help, Lionel