The Simpsons Classifier using Python, Swift and a Demo Mobile App.

1. Installing the required tools and environment to start learning the model:

• python=3.6
• pip=19.0.3
  • tensorflow==1.13.1
  • keras==2.2.4
  • coremltools==2.1.0

2. Learning process

• Layers: 11
• Epochs: 50
• Batch size: 610
• Accuracy: 0.8489 (but it seems to be much lower)

model = Sequential()

model.add(Conv2D(filters=48, kernel_size=(3, 3), strides=(1, 1), activation='relu', input_shape=input_shape))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
model.add(Dropout(0.3))

model.add(Conv2D(filters=32, kernel_size=(5, 5), activation='relu'))
model.add(MaxPooling2D(pool_size=(3, 3)))
model.add(Dropout(0.2))

model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dense(classes_count, activation='softmax'))
model.summary()

model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
...
model.save('simpsons_model.h5')

3. Use CoremlTools.

Convert the Keras model (.h5) to the CoreML model.

coreml_model = coremltools.converters.keras.convert(model,
                                                    input_names='image',
                                                    image_input_names='image',
                                                    class_labels=labels)

coreml_model.author = 'Makarov Anton'
coreml_model.short_description = 'The Simpsons Classifier'
coreml_model.input_description['image'] = 'Image of one of the Simpsons characters'
coreml_model.output_description['output1'] = 'Character recognition probability'
coreml_model.save('simpsons_model.mlmodel')

4. Sample iOS App

• Swift
• import CoreML, Vision, AVKit (Live Camera)

private func getCoreMLModel() -> VNCoreMLModel? {
    guard let model = try? VNCoreMLModel(for: simpsons_model().model) else {
        return nil
    }
    return model
}