Quantization problem

LenBonJJ · June 13, 2021, 7:54am

https://forum.edgeimpulse.com/uploads/default/original/2X/e/ea3f9633c89474f2c8e382ac3ad787896fe4cad0.png

may I know how you get the concrete_func?
I am trying to quantized and used in openmv but failed
I will appreciate it if you could give me some advice

LenBonJJ · June 13, 2021, 7:56am

aurel · June 14, 2021, 7:50am

Hi @LenBonJJ,

You can retrieve the quantized version from the studio -> Dashboard:

Aurelien

janjongboom · June 22, 2021, 6:39am

Answered in your other thread as well, but the function is here in TensorFlow:

github.com

tensorflow/tensorflow/blob/08756627a6defd3e4131ec8af5e81009f1de1f0d/tensorflow/lite/python/lite.py#L1279


      duplicate elements.
    trackable_obj: tf.AutoTrackable object associated with `funcs`. A
      reference to this object needs to be maintained so that Variables do not
      get garbage collected since functions have a weak reference to
      Variables. This is only required when the tf.AutoTrackable object is not
      maintained by the user (e.g. `from_saved_model`).
  """
  super(TFLiteConverterV2, self).__init__(funcs, trackable_obj)
@classmethod
def from_concrete_functions(cls, funcs):
  """Creates a TFLiteConverter object from ConcreteFunctions.
  Args:
    funcs: List of TensorFlow ConcreteFunctions. The list should not contain
      duplicate elements. Currently converter can only convert a single
      ConcreteFunction. Converting multiple functions is under development.
  Returns:
    TFLiteConverter object.