Neural network 3D visualization framework, build interactive and intuitive model in browsers, support pre-trained deep learning models from TensorFlow, Keras, TensorFlow.js
@CharlesLiuyx Hi, I met with some problems when I was trying to visualize the LeNet model in the examples. I used the tensorspacejs_converter to preprocess the model first, and it seemed that the preprocessing was completed successfully, here is the output of the preprocessing in the terminal:
2022-04-24 14:57:46.833002: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
Saving enc_model...
Saving converted tfjs model...
Removing enc_model file...
Mission Complete!!!
And the following is the javascript problem I met when I was trying to use the converted model.json in the HTML
Uncaught (in promise) TypeError: Cannot read properties of undefined (reading 'setShape')
at R.configureLayerShape (Sequential.js:333:25)
at R.initTSPModel (Sequential.js:155:9)
at AbstractModel.js:312:10
I also tried to use the model.json and mnist.weights.bin downloaded from the tensorspace-website project, and it worked. So I assume that there might be something wrong with my preprocessing. Could you please help me with it?
@CharlesLiuyx Hi, I met with some problems when I was trying to visualize the LeNet model in the examples. I used the tensorspacejs_converter to preprocess the model first, and it seemed that the preprocessing was completed successfully, here is the output of the preprocessing in the terminal:
And the following is the javascript problem I met when I was trying to use the converted model.json in the HTML
I also tried to use the model.json and mnist.weights.bin downloaded from the tensorspace-website project, and it worked. So I assume that there might be something wrong with my preprocessing. Could you please help me with it?