trains the beer classifier

train.py

import turicreate as tc
import os

current_dir = os.path.dirname(__file__)
data = tc.SFrame(os.path.join(current_dir, 'data/processed/beers.sframe'))

train, test = data.random_split(0.8)

model = tc.image_classifier.create(train, target='label', max_iterations=100)

# evaluate on test data
print(model.evaluate(test))

# persist the model
model.export_coreml(os.path.join(
    current_dir, 'model/BeerClassifier.mlmodel'))

model.save(os.path.join(current_dir, 'model/v1.model'))