def testZeroHiddenLayers(self):
# Build config.
feature_spec = {
"time_feature_1": {
"length": 10,
"is_time_series": True,
},
"time_feature_2": {
"length": 10,
"is_time_series": True,
},
"aux_feature_1": {
"length": 1,
"is_time_series": False,
},
}
config = configurations.base()
config["inputs"]["features"] = feature_spec
config = configdict.ConfigDict(config)
config.hparams.output_dim = 1
config.hparams.num_pre_logits_hidden_layers = 0
# Build model.
features = input_ops.build_feature_placeholders(config.inputs.features)
labels = input_ops.build_labels_placeholder()
model = astro_model.AstroModel(features, labels, config.hparams,
tf.estimator.ModeKeys.TRAIN)
model.build()
# Validate Tensor shapes.
self.assertShapeEquals((None, 21), model.pre_logits_concat)
logits_w = testing.get_variable_by_name("logits/kernel")
self.assertShapeEquals((21, 1), logits_w)
def testOneTimeSeriesFeature(self):
# Build config.
feature_spec = {
"time_feature_1": {
"length": 10,
"is_time_series": True,
}
}
config = configurations.base()
config["inputs"]["features"] = feature_spec
config = configdict.ConfigDict(config)
# Build model.
features = input_ops.build_feature_placeholders(config.inputs.features)
labels = input_ops.build_labels_placeholder()
model = astro_model.AstroModel(features, labels, config.hparams,
tf.estimator.ModeKeys.TRAIN)
model.build()
# Validate hidden layers.
self.assertItemsEqual(["time_feature_1"],
model.time_series_hidden_layers.keys())
self.assertIs(model.time_series_features["time_feature_1"],
model.time_series_hidden_layers["time_feature_1"])
self.assertEqual(len(model.aux_hidden_layers), 0)
self.assertIs(model.time_series_features["time_feature_1"],
model.pre_logits_concat)
def testInvalidModeRaisesError(self):
# Build config.
config = configdict.ConfigDict(configurations.base())
# Build model.
features = input_ops.build_feature_placeholders(config.inputs.features)
labels = input_ops.build_labels_placeholder()
with self.assertRaises(ValueError):
_ = astro_model.AstroModel(features, labels, config.hparams,
"training")
def testZeroFeaturesRaisesError(self):
# Build config.
config = configurations.base()
config["inputs"]["features"] = {}
config = configdict.ConfigDict(config)
# Build model.
features = input_ops.build_feature_placeholders(config.inputs.features)
labels = input_ops.build_labels_placeholder()
model = astro_model.AstroModel(features, labels, config.hparams,
tf.estimator.ModeKeys.TRAIN)
with self.assertRaises(ValueError):
# Raises ValueError because at least one feature is required.
model.build()
def testEvalMode(self):
# Build config.
feature_spec = {
"time_feature_1": {
"length": 10,
"is_time_series": True,
},
"time_feature_2": {
"length": 10,
"is_time_series": True,
},
"aux_feature_1": {
"length": 1,
"is_time_series": False,
},
}
config = configurations.base()
config["inputs"]["features"] = feature_spec
config = configdict.ConfigDict(config)
config.hparams.output_dim = 1
# Build model.
features = input_ops.build_feature_placeholders(config.inputs.features)
labels = input_ops.build_labels_placeholder()
model = astro_model.AstroModel(features, labels, config.hparams,
tf.estimator.ModeKeys.TRAIN)
model.build()
# Validate Tensor shapes.
self.assertShapeEquals((None, 21), model.pre_logits_concat)
self.assertShapeEquals((None, 1), model.logits)
self.assertShapeEquals((None, 1), model.predictions)
self.assertShapeEquals((None, ), model.batch_losses)
self.assertShapeEquals((), model.total_loss)
# Execute the TensorFlow graph.
scaffold = tf.train.Scaffold()
scaffold.finalize()
with self.test_session() as sess:
sess.run([scaffold.init_op, scaffold.local_init_op])
step = sess.run(model.global_step)
self.assertEqual(0, step)
# Fetch total loss.
features = testing.fake_features(feature_spec, batch_size=16)
labels = testing.fake_labels(config.hparams.output_dim,
batch_size=16)
feed_dict = input_ops.prepare_feed_dict(model, features, labels)
total_loss = sess.run(model.total_loss, feed_dict=feed_dict)
self.assertShapeEquals((), total_loss)
def local_global():
"""Base configuration for a CNN model with separate local/global views."""
config = parent_configs.base()
# Override the model features to be local_view and global_view time series.
config["inputs"]["features"] = {
"local_view": {
"length": 201,
"is_time_series": True,
},
"global_view": {
"length": 2001,
"is_time_series": True,
},
}
# Add configurations for the convolutional layers of time series features.
config["hparams"]["time_series_hidden"] = {
"local_view": {
"cnn_num_blocks": 2,
"cnn_block_size": 2,
"cnn_initial_num_filters": 16,
"cnn_block_filter_factor": 2,
"cnn_kernel_size": 5,
"convolution_padding": "same",
"pool_size": 7,
"pool_strides": 2,
},
"global_view": {
"cnn_num_blocks": 5,
"cnn_block_size": 2,
"cnn_initial_num_filters": 16,
"cnn_block_filter_factor": 2,
"cnn_kernel_size": 5,
"convolution_padding": "same",
"pool_size": 5,
"pool_strides": 2,
},
}
config["hparams"]["num_pre_logits_hidden_layers"] = 4
config["hparams"]["pre_logits_hidden_layer_size"] = 512
return config
def base():
"""Base configuration for a CNN model with a single global view."""
config = parent_configs.base()
# Add configuration for the convolutional layers of the global_view feature.
config["hparams"]["time_series_hidden"] = {
"global_view": {
"cnn_num_blocks": 5,
"cnn_block_size": 2,
"cnn_initial_num_filters": 16,
"cnn_block_filter_factor": 2,
"cnn_kernel_size": 5,
"convolution_padding": "same",
"pool_size": 5,
"pool_strides": 2,
},
}
config["hparams"]["num_pre_logits_hidden_layers"] = 4
config["hparams"]["pre_logits_hidden_layer_size"] = 1024
return config
def base():
"""Base config for a fully connected model with a single global view."""
config = parent_configs.base()
# Add configuration for the fully-connected layers of the global_view feature.
config["hparams"]["time_series_hidden"] = {
"global_view": {
"num_local_layers": 0,
"local_layer_size": 128,
# If > 0, the first layer is implemented as a wide convolutional layer
# for invariance to small translations.
"translation_delta": 0,
# Pooling type following the wide convolutional layer.
"pooling_type": "max",
# Dropout rate for the fully connected layers.
"dropout_rate": 0.0,
},
}
return config
def local_global():
"""Base config for a locally fully connected model with local/global views."""
config = parent_configs.base()
# Override the model features to be local_view and global_view time series.
config["inputs"]["features"] = {
"local_view": {
"length": 201,
"is_time_series": True,
},
"global_view": {
"length": 2001,
"name_in_proto": "light_curve",
"is_time_series": True,
"data_source": "",
},
}
# Add configurations for the fully-connected layers of time series features.
config["hparams"]["time_series_hidden"] = {
"local_view": {
"num_local_layers": 0,
"local_layer_size": 128,
"translation_delta": 0, # For wide convolution.
"pooling_type": "max", # For wide convolution.
"dropout_rate": 0.0,
},
"global_view": {
"num_local_layers": 0,
"local_layer_size": 128,
"translation_delta": 0, # For wide convolution.
"pooling_type": "max", # For wide convolution.
"dropout_rate": 0.0,
},
}
return config
