def predict(file_name):
"""predict for the given file name."""
model = classifier.train()
with open(file_name, 'r') as f:
return classifier.classify(f.read(), model)
def test_train():
model = classifier.train()
p_go_given_token = model.p_go_given_token('range')
# counts look ok
assert model.nall_docs > 400
assert model.nnotgo_docs > 150
assert model.ngo_docs > 150
# probabilities within reasonable bounds
assert p_go_given_token > 0.5
assert p_go_given_token < 1.0
