Convert Keras model - transform to image

I am trying to convert my keras model again like I did a couple of coremltools versions before. The image shape is right I guess (H,W,channel).


coreml_model_new = coremltools.converters.keras.convert(final_model,input_names='image',
                                                        output_names='probs',image_input_names='image',
                                                        class_labels='classes.txt',
                                                        predicted_feature_name='class',image_scale=scale)


I am using the same code as before and get the following error:


2500 array_shape = tuple(input_.type.multiArrayType.shape)
-> 2501 channels, height, width = array_shape
ValueError: need more than 1 value to unpack


Thanks

Replies

This is unbelievable. *** changed. I checked all dependencies.


image_input_names='image'


it is a keras model with HxWxChannel output like it should be.


Any ideas?