How to load SlicingOpLambda and TFOpLambda Layer in tfjs?

General Discussion
, , ,
#1

I’m trying to train a custom model for pose detection using MoveNet. For that, I need to normalise the landmarks or key points of the training data. The Keras and tf functions which are being used in python for normalizing are working fine but when I am loading that model in tfjs, the model is giving errors stating SlicingOpLambda and TFOpLambda Layer is undefined. If someone can help with this, please do.

#2

Check Strided slice implementation leads to negative tensor dimension · Issue #4853 · tensorflow/tfjs · GitHub

#4

Hi Aseem,

I’m facing the same issue and I wonder if you found a solution to this problem.

#5

I was talking to one of our SWEs on the team and they suggested for you to try converting your layers model to a graph model first with the TensorFlow.js converter.

Check this documentation here:

In this situation you could use --input_format keras and then set the --output_format tfjs_graph_model as frozen model should be suitable for inference in browser as you do not need to retain in the browser.

Let me know if that helps!