def test_wrong_window_size(self):
estimator = ARRegressor(periodicities=10,
num_features=1,
input_window_size=10,
output_window_size=6)
def _bad_window_size_input_fn():
return ({
TrainEvalFeatures.TIMES: [[1]],
TrainEvalFeatures.VALUES: [[[1.]]]
}, None)
def _good_data():
return ({
TrainEvalFeatures.TIMES:
np.arange(16)[None, :],
TrainEvalFeatures.VALUES:
array_ops.reshape(np.arange(16), [1, 16, 1])
}, None)
with self.assertRaisesRegexp(ValueError, "set window_size=16"):
estimator.train(input_fn=_bad_window_size_input_fn, steps=1)
# Get a checkpoint for evaluation
estimator.train(input_fn=_good_data, steps=1)
with self.assertRaisesRegexp(ValueError,
"requires a window of at least"):
estimator.evaluate(input_fn=_bad_window_size_input_fn, steps=1)
def test_wrong_window_size(self):
estimator = ARRegressor(
periodicities=10, num_features=1,
input_window_size=10, output_window_size=6)
def _bad_window_size_input_fn():
return ({TrainEvalFeatures.TIMES: [[1]],
TrainEvalFeatures.VALUES: [[[1.]]]},
None)
def _good_data():
return ({TrainEvalFeatures.TIMES: np.arange(16)[None, :],
TrainEvalFeatures.VALUES: array_ops.reshape(
np.arange(16), [1, 16, 1])},
None)
with self.assertRaisesRegexp(ValueError, "set window_size=16"):
estimator.train(input_fn=_bad_window_size_input_fn, steps=1)
# Get a checkpoint for evaluation
estimator.train(input_fn=_good_data, steps=1)
with self.assertRaisesRegexp(ValueError, "requires a window of at least"):
estimator.evaluate(input_fn=_bad_window_size_input_fn, steps=1)
def train_helper(self, input_window_size, loss,
max_loss=None, train_steps=200,
anomaly_prob=0.01,
anomaly_distribution=None,
multiple_periods=False):
np.random.seed(3)
data_noise_stddev = 0.2
if max_loss is None:
if loss == ARModel.NORMAL_LIKELIHOOD_LOSS:
max_loss = 1.0
else:
max_loss = 0.05 / (data_noise_stddev ** 2)
train_data, test_data = self.create_data(
noise_stddev=data_noise_stddev,
anomaly_prob=anomaly_prob,
multiple_periods=multiple_periods)
output_window_size = 10
window_size = input_window_size + output_window_size
class _RunConfig(estimator_lib.RunConfig):
@property
def tf_random_seed(self):
return 3
estimator = ARRegressor(
periodicities=self.period,
anomaly_prior_probability=0.01 if anomaly_distribution else None,
anomaly_distribution=anomaly_distribution,
num_features=2,
output_window_size=output_window_size,
num_time_buckets=20,
input_window_size=input_window_size,
hidden_layer_sizes=[16],
loss=loss,
config=_RunConfig())
train_input_fn = input_pipeline.RandomWindowInputFn(
time_series_reader=input_pipeline.NumpyReader(train_data),
window_size=window_size,
batch_size=64,
num_threads=1,
shuffle_seed=2)
test_input_fn = test_utils.AllWindowInputFn(
time_series_reader=input_pipeline.NumpyReader(test_data),
window_size=window_size)
# Test training
estimator.train(
input_fn=train_input_fn,
steps=train_steps)
test_evaluation = estimator.evaluate(input_fn=test_input_fn, steps=1)
test_loss = test_evaluation["loss"]
logging.info("Final test loss: %f", test_loss)
self.assertLess(test_loss, max_loss)
if loss == ARModel.SQUARED_LOSS:
# Test that the evaluation loss is reported without input scaling.
self.assertAllClose(
test_loss,
np.mean((test_evaluation["mean"] - test_evaluation["observed"]) ** 2))
# Test predict
train_data_times = train_data[TrainEvalFeatures.TIMES]
train_data_values = train_data[TrainEvalFeatures.VALUES]
test_data_times = test_data[TrainEvalFeatures.TIMES]
test_data_values = test_data[TrainEvalFeatures.VALUES]
predict_times = np.expand_dims(np.concatenate(
[train_data_times[input_window_size:], test_data_times]), 0)
predict_true_values = np.expand_dims(np.concatenate(
[train_data_values[input_window_size:], test_data_values]), 0)
state_times = np.expand_dims(train_data_times[:input_window_size], 0)
state_values = np.expand_dims(
train_data_values[:input_window_size, :], 0)
state_exogenous = state_times[:, :, None][:, :, :0]
def prediction_input_fn():
return ({
PredictionFeatures.TIMES: training.limit_epochs(
predict_times, num_epochs=1),
PredictionFeatures.STATE_TUPLE: (state_times,
state_values,
state_exogenous)
}, {})
(predictions,) = tuple(estimator.predict(input_fn=prediction_input_fn))
predicted_mean = predictions["mean"][:, 0]
true_values = predict_true_values[0, :, 0]
if loss == ARModel.NORMAL_LIKELIHOOD_LOSS:
variances = predictions["covariance"][:, 0]
standard_deviations = np.sqrt(variances)
# Note that we may get tighter bounds with more training steps.
errors = np.abs(predicted_mean - true_values) > 4 * standard_deviations
fraction_errors = np.mean(errors)
logging.info("Fraction errors: %f", fraction_errors)
def train_helper(self, input_window_size, loss,
max_loss=None, train_steps=200,
anomaly_prob=0.01,
anomaly_distribution=None,
multiple_periods=False):
np.random.seed(3)
data_noise_stddev = 0.2
if max_loss is None:
if loss == ARModel.NORMAL_LIKELIHOOD_LOSS:
max_loss = 1.0
else:
max_loss = 0.05 / (data_noise_stddev ** 2)
train_data, test_data = self.create_data(
noise_stddev=data_noise_stddev,
anomaly_prob=anomaly_prob,
multiple_periods=multiple_periods)
output_window_size = 10
window_size = input_window_size + output_window_size
class _RunConfig(estimator_lib.RunConfig):
@property
def tf_random_seed(self):
return 3
estimator = ARRegressor(
periodicities=self.period,
anomaly_prior_probability=0.01 if anomaly_distribution else None,
anomaly_distribution=anomaly_distribution,
num_features=2,
output_window_size=output_window_size,
num_time_buckets=20,
input_window_size=input_window_size,
hidden_layer_sizes=[16],
loss=loss,
config=_RunConfig())
train_input_fn = input_pipeline.RandomWindowInputFn(
time_series_reader=input_pipeline.NumpyReader(train_data),
window_size=window_size,
batch_size=64,
num_threads=1,
shuffle_seed=2)
test_input_fn = test_utils.AllWindowInputFn(
time_series_reader=input_pipeline.NumpyReader(test_data),
window_size=window_size)
# Test training
estimator.train(
input_fn=train_input_fn,
steps=train_steps)
test_evaluation = estimator.evaluate(input_fn=test_input_fn, steps=1)
test_loss = test_evaluation["loss"]
logging.info("Final test loss: %f", test_loss)
self.assertLess(test_loss, max_loss)
if loss == ARModel.SQUARED_LOSS:
# Test that the evaluation loss is reported without input scaling.
self.assertAllClose(
test_loss,
np.mean((test_evaluation["mean"] - test_evaluation["observed"]) ** 2))
# Test predict
train_data_times = train_data[TrainEvalFeatures.TIMES]
train_data_values = train_data[TrainEvalFeatures.VALUES]
test_data_times = test_data[TrainEvalFeatures.TIMES]
test_data_values = test_data[TrainEvalFeatures.VALUES]
predict_times = np.expand_dims(np.concatenate(
[train_data_times[input_window_size:], test_data_times]), 0)
predict_true_values = np.expand_dims(np.concatenate(
[train_data_values[input_window_size:], test_data_values]), 0)
state_times = np.expand_dims(train_data_times[:input_window_size], 0)
state_values = np.expand_dims(
train_data_values[:input_window_size, :], 0)
def prediction_input_fn():
return ({
PredictionFeatures.TIMES: training.limit_epochs(
predict_times, num_epochs=1),
PredictionFeatures.STATE_TUPLE: (state_times, state_values)
}, {})
(predictions,) = tuple(estimator.predict(input_fn=prediction_input_fn))
predicted_mean = predictions["mean"][:, 0]
true_values = predict_true_values[0, :, 0]
if loss == ARModel.NORMAL_LIKELIHOOD_LOSS:
variances = predictions["covariance"][:, 0]
standard_deviations = np.sqrt(variances)
# Note that we may get tighter bounds with more training steps.
errors = np.abs(predicted_mean - true_values) > 4 * standard_deviations
fraction_errors = np.mean(errors)
logging.info("Fraction errors: %f", fraction_errors)