Cannot convert to CoreML

[JakupGuven]

I used this https://github.com/qqwweee/keras-yolo3 to convert my model trained using Darknet to a Keras model. I'm trying to convert it to CoreML but get errors. I think it would be very good if you could specify how your virtualenv looks like when you do this @Ma-Dan

The error I get:

tf.estimator package not installed. tf.estimator package not installed. Traceback (most recent call last): File "coreml.py", line 3, in coreml_model = coremltools.converters.keras.convert('Doors.h5', input_names='input1', image_input_names='input1', output_names=['output1', 'output2', 'output3'], image_scale=1/255.) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/coremltools/converters/keras/_keras_converter.py", line 760, in convert custom_conversion_functions=custom_conversion_functions) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/coremltools/converters/keras/_keras_converter.py", line 537, in convertToSpec custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/coremltools/converters/keras/_keras_converter.py", line 168, in _convert model = _keras.models.load_model(model, custom_objects = custom_objects) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/keras/models.py", line 142, in load_model model = model_from_config(model_config, custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/keras/models.py", line 193, in model_from_config return layer_from_config(config, custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/keras/utils/layer_utils.py", line 40, in layer_from_config custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/keras/engine/topology.py", line 2582, in from_config process_layer(layer_data) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/keras/engine/topology.py", line 2560, in process_layer custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/keras/utils/layer_utils.py", line 42, in layer_from_config return layer_class.from_config(config['config']) File "/home/jakup/Utilities/coreml/local/lib/python2.7/site-packages/keras/engine/topology.py", line 1025, in from_config return cls(**config) TypeError: init() got an unexpected keyword argument 'dtype'

[JakupGuven]

I can add that I'm trying to convert a YOLOv3-tiny model

[Ma-Dan]

I used python 3.6.5 with Keras 2.1.6 tensorflow 1.5.0 coremltools 2.0 Convert with tensorflow 1.12.0 and Keras 2.2.4 will get errors like ImportError: cannot import name relu6. TypeError('Keyword argument not understood:', kwarg)

[JakupGuven]

@Ma-Dan Do you use the keras-yolo3 repo to convert from Darknet to Keras? If so how does that virtualenv look?

[JakupGuven]

Okay @Ma-Dan. I installed a fresh Python 3.6.5 from here: https://www.python.org/downloads/release/python-365/

You can see in the image I attach that I have the versions of tensorflow, keras and coremltools that you specified.

I created a virtualenv and installed dependencies with correct versions and I get this error message: Traceback (most recent call last): File "coreml.py", line 3, in coreml_model = coremltools.converters.keras.convert('Doors.h5', input_names='input1', image_input_names='input1', output_names=['output1', 'output2', 'output3'], image_scale=1/255.) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/coremltools/converters/keras/_keras_converter.py", line 752, in convert custom_conversion_functions=custom_conversion_functions) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/coremltools/converters/keras/_keras_converter.py", line 550, in convertToSpec custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/coremltools/converters/keras/_keras2_converter.py", line 197, in _convert model = _keras.models.load_model(model, custom_objects = custom_objects) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/models.py", line 270, in load_model model = model_from_config(model_config, custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/models.py", line 347, in model_from_config return layer_module.deserialize(config, custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/layers/init.py", line 55, in deserialize printable_module_name='layer') File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/utils/generic_utils.py", line 144, in deserialize_keras_object list(custom_objects.items()))) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/engine/topology.py", line 2525, in from_config process_layer(layer_data) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/engine/topology.py", line 2511, in process_layer custom_objects=custom_objects) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/layers/init.py", line 55, in deserialize printable_module_name='layer') File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/utils/generic_utils.py", line 146, in deserialize_keras_object return cls.from_config(config['config']) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/engine/topology.py", line 1271, in from_config return cls(*config) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/legacy/interfaces.py", line 91, in wrapper return func(args, kwargs) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/layers/convolutional.py", line 1806, in init super(UpSampling2D, self).init(kwargs) File "/home/jakup/Utilities/coreml/lib/python3.6/site-packages/keras/engine/topology.py", line 293, in init raise TypeError('Keyword argument not understood:', kwarg) TypeError: ('Keyword argument not understood:', 'interpolation')