def testDenseFeatures(self):
features = {
"image_a": [[[[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]],
[[[0.7, 0.7, 0.7], [0.1, 0.2, 0.3]]]],
"image_b": [[[[0.1, 0.2, 0.1], [0.2, 0.1, 0.2]]],
[[[0.1, 0.2, 0.3], [0.3, 0.2, 0.1]]]],
}
feature_columns = [
hub.image_embedding_column("image_a", self.spec),
hub.image_embedding_column("image_b", self.spec),
]
if not feature_column_v2.is_feature_column_v2(feature_columns):
self.skipTest(
"Resources not implemented in the state manager of feature "
"column v2.")
with tf.Graph().as_default():
# We want to test with dense_features_v2.DenseFeatures. This symbol was
# added in https://github.com/tensorflow/tensorflow/commit/64586f18724f737393071125a91b19adf013cf8a.
feature_layer = tf.compat.v2.keras.layers.DenseFeatures(
feature_columns)
feature_layer_out = feature_layer(features)
with tf_v1.train.MonitoredSession() as sess:
output = sess.run(feature_layer_out)
self.assertAllClose(output, [[0.5, 0.7, 0.9, 0.3, 0.3, 0.3],
[0.8, 0.9, 1.0, 0.4, 0.4, 0.4]])
def testConfig(self):
module_path = os.path.join(self.get_temp_dir(), "module")
export_module_spec(self.spec, module_path)
image_column = hub.image_embedding_column("image", module_path)
config = image_column.get_config()
cloned_image_column = hub.feature_column._ImageEmbeddingColumn.from_config(
config)
self.assertEqual(cloned_image_column.module_spec_path,
image_column.module_spec_path)
with self.assertRaisesRegexp(NotImplementedError, "Can only generate"):
image_column = hub.image_embedding_column("image", self.spec)
config = image_column.get_config()
def testWorksWithCannedEstimator(self):
image_column = hub.image_embedding_column("image", self.spec)
other_column = tf_v1.feature_column.numeric_column("number")
feature_columns = [image_column, other_column]
estimator = tf_v1.estimator.DNNClassifier(
hidden_units=[10],
feature_columns=feature_columns,
model_dir=self.get_temp_dir())
# This only tests that estimator apis are working with the feature
# column without throwing exceptions.
features = {
"image":
np.array([[[[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]],
[[[0.7, 0.7, 0.7], [0.1, 0.2, 0.3]]]],
dtype=np.float32),
"number":
np.array([[20], [1]]),
}
labels = np.array([[1], [0]])
if hasattr(tf.compat, "v1"):
numpy_input_fn = tf.compat.v1.estimator.inputs.numpy_input_fn
else:
numpy_input_fn = tf_v1.estimator.inputs.numpy_input_fn
input_fn = numpy_input_fn(features, labels, shuffle=True)
estimator.train(input_fn, max_steps=1)
estimator.evaluate(input_fn, steps=1)
estimator.predict(input_fn)
def testMakeParseExampleSpec(self):
image_column = hub.image_embedding_column("image", self.spec)
parsing_spec = tf_v1.feature_column.make_parse_example_spec(
[image_column])
self.assertEqual(
parsing_spec,
{"image": tf_v1.FixedLenFeature([1, 2, 3], dtype=tf.float32)})
def testDenseFeatures_shareAcrossApplication(self):
features = {
"image": [[[[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]],
[[[0.7, 0.7, 0.7], [0.1, 0.2, 0.3]]]],
}
feature_columns = [
hub.image_embedding_column("image", self.spec),
]
if not feature_column_v2.is_feature_column_v2(feature_columns):
self.skipTest(
"Resources not implemented in the state manager of feature "
"column v2.")
with tf.Graph().as_default():
feature_layer = feature_column_v2.DenseFeatures(feature_columns)
global_vars_before = tf.global_variables()
feature_layer_out_1 = feature_layer(features)
global_vars_middle = tf.global_variables()
feature_layer_out_2 = feature_layer(features)
global_vars_after = tf.global_variables()
self.assertLen(global_vars_before, 0)
self.assertLen(global_vars_middle, 1)
self.assertLen(global_vars_after, 1)
with tf_v1.train.MonitoredSession() as sess:
output_1 = sess.run(feature_layer_out_1)
output_2 = sess.run(feature_layer_out_2)
self.assertAllClose(output_1,
[[0.5, 0.7, 0.9], [0.8, 0.9, 1.0]])
self.assertAllEqual(output_1, output_2)
def testDenseFeatures_shareAcrossApplication(self):
features = {
"image": [[[[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]],
[[[0.7, 0.7, 0.7], [0.1, 0.2, 0.3]]]],
}
feature_columns = [
hub.image_embedding_column("image",
self.randomly_initialized_spec),
]
if not feature_column_v2.is_feature_column_v2(feature_columns):
self.skipTest(
"Resources not implemented in the state manager of feature "
"column v2.")
with tf.Graph().as_default():
# We want to test with dense_features_v2.DenseFeatures. This symbol was
# added in https://github.com/tensorflow/tensorflow/commit/64586f18724f737393071125a91b19adf013cf8a.
feature_layer = tf.compat.v2.keras.layers.DenseFeatures(
feature_columns)
feature_layer_out_1 = feature_layer(features)
feature_layer_out_2 = feature_layer(features)
with tf.compat.v1.train.MonitoredSession() as sess:
output_1 = sess.run(feature_layer_out_1)
output_2 = sess.run(feature_layer_out_2)
self.assertAllClose(output_1, output_2)
def testInputLayer(self):
features = {
"image_a": [[[[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]],
[[[0.7, 0.7, 0.7], [0.1, 0.2, 0.3]]]],
"image_b": [[[[0.1, 0.2, 0.1], [0.2, 0.1, 0.2]]],
[[[0.1, 0.2, 0.3], [0.3, 0.2, 0.1]]]],
}
feature_columns = [
hub.image_embedding_column("image_a", self.spec),
hub.image_embedding_column("image_b", self.spec),
]
with tf.Graph().as_default():
input_layer = tf.feature_column.input_layer(features, feature_columns)
with tf.train.MonitoredSession() as sess:
output = sess.run(input_layer)
self.assertAllClose(output, [[0.5, 0.7, 0.9, 0.3, 0.3, 0.3],
[0.8, 0.9, 1.0, 0.4, 0.4, 0.4]])
def testDenseFeatures(self):
features = {
"image_a": [[[[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]],
[[[0.7, 0.7, 0.7], [0.1, 0.2, 0.3]]]],
"image_b": [[[[0.1, 0.2, 0.1], [0.2, 0.1, 0.2]]],
[[[0.1, 0.2, 0.3], [0.3, 0.2, 0.1]]]],
}
feature_columns = [
hub.image_embedding_column("image_a", self.spec),
hub.image_embedding_column("image_b", self.spec),
]
with tf.Graph().as_default():
feature_layer = feature_column_lib.DenseFeatures(feature_columns)
feature_layer_out = feature_layer(features)
with tf_v1.train.MonitoredSession() as sess:
output = sess.run(feature_layer_out)
self.assertAllClose(output, [[0.5, 0.7, 0.9, 0.3, 0.3, 0.3],
[0.8, 0.9, 1.0, 0.4, 0.4, 0.4]])
def testImageSizeManuallySpecified(self):
spec = hub.create_module_spec(create_image_module_fn([None, None]))
image_column = hub.image_embedding_column("image",
spec,
image_size=[229, 229])
parsing_spec = tf_v1.feature_column.make_parse_example_spec(
[image_column])
self.assertEqual(
parsing_spec,
{"image": tf_v1.FixedLenFeature([229, 229, 3], dtype=tf.float32)})
def testDenseFeatures(self):
features = {
"image_a": [[[[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]],
[[[0.7, 0.7, 0.7], [0.1, 0.2, 0.3]]]],
"image_b": [[[[0.1, 0.2, 0.1], [0.2, 0.1, 0.2]]],
[[[0.1, 0.2, 0.3], [0.3, 0.2, 0.1]]]],
}
feature_columns = [
hub.image_embedding_column("image_a", self.spec),
hub.image_embedding_column("image_b", self.spec),
]
if not feature_column_v2.is_feature_column_v2(feature_columns):
self.skipTest("Resources not implemented in the state manager of feature "
"column v2.")
with tf.Graph().as_default():
feature_layer = _dense_features_module.DenseFeatures(feature_columns)
feature_layer_out = feature_layer(features)
with tf_v1.train.MonitoredSession() as sess:
output = sess.run(feature_layer_out)
self.assertAllClose(
output,
[[0.5, 0.7, 0.9, 0.3, 0.3, 0.3], [0.8, 0.9, 1.0, 0.4, 0.4, 0.4]])
def create_classifier(config, parameters):
"""Creates a DNN classifier.
Defines features and builds an 'Estimator' with them.
Args:
config: `RunConfig` object to configure the runtime of the `Estimator`.
parameters: Parameters passed to the job.
Returns:
`tf.estimator.DNNClassifier` with specified features and architecture.
"""
# Columns to be used as features.
subspecies = tf.feature_column.categorical_column_with_vocabulary_list(
'subspecies',
vocabulary_list=BEE_SUBSPECIES,
default_value=0)
subspecies = tf.feature_column.embedding_column(
subspecies, dimension=parameters.subspecies_embedding)
image = hub.image_embedding_column('image', parameters.tf_hub_module)
feature_cols = [subspecies, image]
layer = parameters.first_layer_size
lfrac = parameters.layer_reduction_fraction
nlayers = parameters.number_layers
h_units = [layer]
for _ in range(nlayers - 1):
h_units.append(math.ceil(layer * lfrac))
layer = h_units[-1]
estimator = tf.estimator.DNNClassifier(
feature_columns=feature_cols,
hidden_units=h_units,
optimizer=tf.train.AdagradOptimizer(
learning_rate=parameters.learning_rate),
dropout=parameters.dropout, config=config)
estimator = tf.contrib.estimator.add_metrics(
estimator, _estimator_metrics)
estimator = tf.contrib.estimator.forward_features(estimator, 'img_file')
return estimator
def testDenseFeatures_shareAcrossApplication(self):
features = {
"image": [[[[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]],
[[[0.7, 0.7, 0.7], [0.1, 0.2, 0.3]]]],
}
feature_columns = [
hub.image_embedding_column("image", self.randomly_initialized_spec),
]
if not feature_column_v2.is_feature_column_v2(feature_columns):
self.skipTest("Resources not implemented in the state manager of feature "
"column v2.")
with tf.Graph().as_default():
feature_layer = _dense_features_module.DenseFeatures(feature_columns)
feature_layer_out_1 = feature_layer(features)
feature_layer_out_2 = feature_layer(features)
with tf_v1.train.MonitoredSession() as sess:
output_1 = sess.run(feature_layer_out_1)
output_2 = sess.run(feature_layer_out_2)
self.assertAllClose(output_1, output_2)
def testVariableShape(self):
image_column = hub.image_embedding_column("image", self.spec)
self.assertEqual(image_column._variable_shape, [3])
def testExpectedImageSize(self):
image_column = hub.image_embedding_column("image", self.spec)
# The usage comment recommends this code pattern, so we test it here.
self.assertSequenceEqual(
hub.get_expected_image_size(image_column.module_spec), [1, 2])
def testParents(self):
image_column = hub.image_embedding_column("image", self.spec)
self.assertEqual(["image"], image_column.parents)
def testName(self):
image_column = hub.image_embedding_column(
tf.feature_column.numeric_column("image"), self.spec)
self.assertEqual("image_hub_module_embedding", image_column.name)
