Meghna Natraj

@yahuuu yes you can by using TF 1.15. The code might looks like this ([source](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/g3doc/r1/convert/python_api.md)) ``` import numpy as np import tensorflow as tf # Generate tf.keras model. model =...

@yahuuu Quantization is most effective for larger models with more convolutions. It also does not guarantee that the model will be faster. This model might be one of the few...

The models might be quite outdated (from TF1). Please use models from TF Hub eg: https://tfhub.dev/s?deployment-format=lite&module-type=image-classification&q=quantized

The changes look good, but could you elaborate "if you want to test it on master, there are some error about "tensorflow_profiler_logger", maybe someone should fix it."