Converting h5 to coreMl (IOS)

Question

I'm currently working in a collaboration. My task is to convert an h5-file, which was generated by a neural network with tensorflow, to an coreML. Additionally I should implement it to my Xcode Project.

The Input is a two dimensional array of 21 Floats:

input = [[0.5, 0.4, ...]]

The output should be a Float between 0 and 1.

I've tried a lot but as far as I know the main issue is that coreML supports just the classification of a picture. I didn't find any clue how to convert an h5 to an coreML with this specific type of input and output as mentioned. Can anybody help? Thanks a lot!

Edit

This is my code. I'm confused because once I read that I just have to name the input and output instead of defining the variable as an MLMultiArray. I guess this is my main issue. But didn't catch how to define the input as an MLMultiArray.

from keras.models import load_model
import coremltools


coreml_model = coremltools.converters.keras.convert('modelv.h5',
input_names=['data'],
output_names=['output'],
                        )
coreml_model.save('PredictionModel.mlmodel')

When I run the code I'm getting this following message from the compiler.

runfile('/Path/Neuronal Network')
Traceback (most recent call last):

  File "/Path/ Neuronal Network/Converter.py", line 20, in 
    output_names='output',

  File "/path/", line 804, in convert
    use_float_arraytype=use_float_arraytype)


  File "/Path/opt/anaconda3/lib/python3.7/site-packages/coremltools/converters/keras/_keras_converter.py", line 585, in convertToSpec
    use_float_arraytype=use_float_arraytype)



  File "/Path/opt/anaconda3/lib/python3.7/site-packages/coremltools/converters/keras/_keras2_converter.py", line 328, in _convert
    graph.build()

  File "/Path/opt/anaconda3/lib/python3.7/site-packages/coremltools/converters/keras/_topology2.py", line 740, in build
    self.make_input_layers()

  File "/Path/opt/anaconda3/lib/python3.7/site-packages/coremltools/converters/keras/_topology2.py", line 169, in make_input_layers
    if isinstance(kl, InputLayer) and kl.input == ts:

  File "/Path/opt/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py", line 765, in __bool__
    self._disallow_bool_casting()

  File "/Path/opt/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py", line 534, in _disallow_bool_casting
    self._disallow_in_graph_mode("using a `tf.Tensor` as a Python `bool`")

  File "/Path/opt/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py", line 523, in _disallow_in_graph_mode
    " this function with @tf.function.".format(task))

OperatorNotAllowedInGraphError: using a `tf.Tensor` as a Python `bool` is not allowed in Graph execution. Use Eager execution or decorate this function with @tf.function.

Converting h5 to coreMl (IOS)

0 Answers