CoreML: exception Espresso exception: "Invalid state": Null output blobs

Question

I run a keras model inside my iOS application using CoreML and I have this problem I can't completely understand but I think it's not related to the app but instead to the Keras model. I'm using 48x48 images, is it supported this size in CoreML?

[Espresso::handle_ex_plan] exception=Espresso exception: "Invalid state": Null output blobs [Exception from Layer: 5: sequential/conv2d_3/BiasAdd]

2020-12-15 01:06:31.245711+0100 TSD[41213:1753543] [coreml] Error computing NN outputs -1

2020-12-15 01:06:31.245849+0100 TSD[41213:1753543] [coreml] Failure in -executePlan:error:. Error computing NN outputs.

def cnn_model():
model = Sequential()
model.add(Conv2D(32, (3, 3), padding='same',
                 activation='relu'))
model.add(Conv2D(32, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.2))

model.add(Conv2D(64, (3, 3), padding='same', activation='relu'))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.2))

model.add(Conv2D(128, (3, 3), padding='same', activation='relu'))
model.add(Conv2D(128, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.2))

model.add(Flatten())
model.add(keras.layers.InputLayer(input_shape=(X.shape[1],)))
model.add(Dense(512, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(CLASSNAME_SIZE, activation='softmax'))
return model

EDIT 1

This is the problematic layer according to the error description (conv2d_3/biasadd)

Answer 1

I would remove this line,

model.add(keras.layers.InputLayer(input_shape=(X.shape[1])))

from your Keras model and try again. Anything else seems fine, but that line might be messing things up during the model conversion to Core ML.