Retrain image detection with MobileNet for use in TFjs, failure in tensorflowjs_converter

[serv-inc]

It is unclear whether this is a documentation or a programming issue. If misfiled, please say so and suggest whether a new issue should be raised, or what else can be done.

System information

• TensorFlow version:
  ('v1.13.1-0-g6612da8951', '1.13.1') (see AttributeError: 'module' object has no attribute 'KerasLayer' on Colab script when downloaded to local, non-cuda laptop #27538)
  ('v1.12.0-9492-g2c319fb415', '2.0.0-alpha0') (see ValueError: Input 0 of layer dense is incompatible with the layer: its rank is undefined, but the layer requires a defined rank on Colab script downloaded to local machine #27539) and
  tf-nightly-2.0-preview (see https://stackoverflow.com/questions/55682557/)

• Doc Link:
  https://www.tensorflow.org/hub/tutorials/image_retraining and
  https://www.tensorflow.org/tutorials/images/hub_with_keras

Describe the documentation issue
Both documents mention but do not describe conversion to mobile models. Yet, conversion using tensorflowjs_converter failed in every attempt. See https://stackoverflow.com/questions/55849309/retrain-image-detection-with-mobilenet, which mentions two attempts by me as well as questions by others with the same problem.

We welcome contributions by users. Will you be able to update submit a PR (use the doc style guide) to fix the doc Issue?

As soon as it is working, I would gladly fix and/or augment the docs. I have not gotten tensorflowjs_converter to work on a retrained MobileNet model, despite the docs stating that it should work, e.g.:

This saved model can loaded for inference later, or converted to TFLite or TFjs.

[ymodak]

This issue is more suitable on tensorflow js repo. Please post it on tensorflow js. Thanks!

[serv-inc]

@ymodak: It's now at tensorflow/tfjs#1576. Thanks!