def test_construct_with_materialization_disabled(self):
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn='my_preprocessing_fn',
materialize=False)
self._verify_outputs(transform, materialize=False)
def testEnableCache(self):
module_file = '/path/to/preprocessing.py'
transform_1 = component.Transform(
examples=self.examples,
schema=self.schema,
module_file=module_file,
)
self.assertEqual(None, transform_1.enable_cache)
transform_2 = component.Transform(
examples=self.examples,
schema=self.schema,
module_file=module_file,
enable_cache=True,
)
self.assertEqual(True, transform_2.enable_cache)
class TransformRunner(TfxComponentRunner):
def __init__(self, args):
component = transform_component.Transform(
input_data=channel.Channel('ExamplesPath'),
schema=channel.Channel('SchemaPath'),
module_file=args.module_file)
def test_construct_with_cache_disabled(self):
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn='my_preprocessing_fn',
disable_analyzer_cache=True)
self._verify_outputs(transform, disable_analyzer_cache=True)
def test_construct_duplicate_user_module(self):
with self.assertRaises(ValueError):
_ = component.Transform(
input_data=self.input_data,
schema=self.schema,
module_file='/path/to/preprocessing.py',
preprocessing_fn='path.to.my_preprocessing_fn',
)
def testConstructDuplicateUserModule(self):
with self.assertRaises(ValueError):
_ = component.Transform(
examples=self.examples,
schema=self.schema,
module_file='/path/to/preprocessing.py',
preprocessing_fn='path.to.my_preprocessing_fn',
)
def test_construct_with_cache_disabled_but_input_cache(self):
with self.assertRaises(ValueError):
_ = component.Transform(examples=self.examples,
schema=self.schema,
preprocessing_fn='my_preprocessing_fn',
disable_analyzer_cache=True,
analyzer_cache=channel_utils.as_channel(
[standard_artifacts.TransformCache()]))
def test_construct_from_module_file(self):
module_file = '/path/to/preprocessing.py'
transform = component.Transform(
examples=self.examples,
schema=self.schema,
module_file=module_file,
)
self._verify_outputs(transform)
self.assertEqual(module_file, transform.exec_properties['module_file'])
def test_construct_with_materialization_disabled_but_output_examples(self):
with self.assertRaises(ValueError):
_ = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn='my_preprocessing_fn',
materialize=False,
transformed_examples=channel_utils.as_channel(
[standard_artifacts.Examples()]))
def testConstructFromModuleFile(self):
module_file = '/path/to/preprocessing.py'
transform = component.Transform(
input_data=self.input_data,
schema=self.schema,
module_file=module_file,
)
self._verify_outputs(transform)
self.assertEqual(module_file, transform.spec.exec_properties['module_file'])
def test_construct_with_parameter(self):
module_file = data_types.RuntimeParameter(name='module-file', ptype=Text)
transform = component.Transform(
examples=self.examples,
schema=self.schema,
module_file=module_file,
)
self._verify_outputs(transform)
self.assertJsonEqual(
str(module_file), str(transform.exec_properties['module_file']))
def testConstructFromPreprocessingFn(self):
preprocessing_fn = 'path.to.my_preprocessing_fn'
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn=preprocessing_fn,
)
self._verify_outputs(transform)
self.assertEqual(preprocessing_fn,
transform.spec.exec_properties['preprocessing_fn'])
def test_construct_from_preprocessing_fn(self):
preprocessing_fn = 'path.to.my_preprocessing_fn'
transform = component.Transform(
input_data=self.input_data,
schema=self.schema,
preprocessing_fn=preprocessing_fn,
)
self._verify_outputs(transform)
self.assertEqual(preprocessing_fn,
transform.spec.exec_properties['preprocessing_fn'])
def __init__(self, input_data: str, schema: str, module_file: str):
component = transform_component.Transform(
input_data=channel.Channel('ExamplesPath'),
schema=channel.Channel('SchemaPath'),
module_file=module_file)
super().__init__(component, {
"input_data": input_data,
"schema": schema
})
def test_construct_from_preprocessing_fn(self):
preprocessing_fn = 'path.to.my_preprocessing_fn'
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn=preprocessing_fn,
)
self._verify_outputs(transform)
self.assertEqual(
preprocessing_fn, transform.exec_properties[
standard_component_specs.PREPROCESSING_FN_KEY])
def test_construct_with_force_tf_compat_v1_override(self):
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn='my_preprocessing_fn',
force_tf_compat_v1=True,
)
self._verify_outputs(transform)
self.assertEqual(
True,
bool(transform.spec.exec_properties[
standard_component_specs.FORCE_TF_COMPAT_V1_KEY]))
def test_construct_with_stats_disabled(self):
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn='my_preprocessing_fn',
disable_statistics=True,
)
self._verify_outputs(transform, disable_statistics=True)
self.assertEqual(
True,
bool(transform.spec.exec_properties[
standard_component_specs.DISABLE_STATISTICS_KEY]))
def __init__(self, input_data: dsl.PipelineParam,
schema: dsl.PipelineParam, module_file: dsl.PipelineParam):
component = transform_component.Transform(
input_data=channel.Channel('ExamplesPath'),
schema=channel.Channel('SchemaPath'),
module_file='')
super().__init__(
component, {
"input_data": input_data,
"schema": schema,
"module_file": module_file,
})
def test_construct_from_preprocessing_fn_with_custom_config(self):
preprocessing_fn = 'path.to.my_preprocessing_fn'
custom_config = {'param': 1}
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn=preprocessing_fn,
custom_config=custom_config,
)
self._verify_outputs(transform)
self.assertEqual(preprocessing_fn,
transform.spec.exec_properties['preprocessing_fn'])
self.assertEqual(json.dumps(custom_config),
transform.spec.exec_properties['custom_config'])
def test_construct_with_splits_config(self):
splits_config = transform_pb2.SplitsConfig(
analyze=['train'], transform=['eval'])
module_file = '/path/to/preprocessing.py'
transform = component.Transform(
examples=self.examples,
schema=self.schema,
module_file=module_file,
splits_config=splits_config,
)
self._verify_outputs(transform)
self.assertEqual(
json_format.MessageToJson(
splits_config, sort_keys=True, preserving_proto_field_name=True),
transform.exec_properties['splits_config'])
def test_construct_with_splits_config(self):
splits_config = transform_pb2.SplitsConfig(analyze=['train'],
transform=['eval'])
module_file = '/path/to/preprocessing.py'
transform = component.Transform(
examples=self.examples,
schema=self.schema,
module_file=module_file,
splits_config=splits_config,
)
self._verify_outputs(transform)
self.assertEqual(
proto_utils.proto_to_json(splits_config),
transform.exec_properties[
standard_component_specs.SPLITS_CONFIG_KEY])
def test_construct_from_preprocessing_fn_with_custom_config(self):
preprocessing_fn = 'path.to.my_preprocessing_fn'
custom_config = {'param': 1}
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn=preprocessing_fn,
custom_config=custom_config,
)
self._verify_outputs(transform)
self.assertEqual(
preprocessing_fn, transform.spec.exec_properties[
standard_component_specs.PREPROCESSING_FN_KEY])
self.assertEqual(
json.dumps(custom_config), transform.spec.exec_properties[
standard_component_specs.CUSTOM_CONFIG_KEY])
def test_construct_from_preprocessing_fn_with_stats_options_updater_fn(
self):
preprocessing_fn = 'path.to.my_preprocessing_fn'
stats_options_updater_fn = 'path.to.my.stats_options_updater_fn'
transform = component.Transform(
examples=self.examples,
schema=self.schema,
preprocessing_fn=preprocessing_fn,
stats_options_updater_fn=stats_options_updater_fn)
self._verify_outputs(transform)
self.assertEqual(
preprocessing_fn, transform.exec_properties[
standard_component_specs.PREPROCESSING_FN_KEY])
self.assertEqual(
stats_options_updater_fn, transform.exec_properties[
standard_component_specs.STATS_OPTIONS_UPDATER_FN_KEY])
def test_construct(self):
source_data_dir = os.path.join(os.path.dirname(__file__), 'testdata',
'taxi')
preprocessing_fn_file = os.path.join(source_data_dir, 'module',
'preprocess.py')
transform = component.Transform(
input_data=channel.as_channel([
types.TfxType(type_name='ExamplesPath', split='train'),
types.TfxType(type_name='ExamplesPath', split='eval'),
]),
schema=channel.as_channel([types.TfxType(type_name='SchemaPath')]),
module_file=preprocessing_fn_file,
)
self.assertEqual('TransformPath',
transform.outputs.transform_output.type_name)
self.assertEqual('ExamplesPath',
transform.outputs.transformed_examples.type_name)
def test_construct_missing_user_module(self):
with self.assertRaises(ValueError):
_ = component.Transform(
input_data=self.input_data,
schema=self.schema,
)
def testConstructMissingUserModule(self):
with self.assertRaises(ValueError):
_ = component.Transform(
examples=self.examples,
schema=self.schema,
)
def testConstructMissingUserModule(self):
with self.assertRaises(ValueError):
_ = component.Transform(
input_data=self.input_data,
schema=self.schema,
)
def _create_pipeline():
"""Implements the chicago east pipeline with TFX."""
print_info("Creating pipeline")
examples = tfrecord_input(_data_root)
example_gen = ImportExampleGen(input_base=examples)
# Computes statistics over data for visualization and example validation.
# pylint: disable=line-too-long
statistics_gen = StatisticsGen(input_data=example_gen.outputs.examples) # Step 3
# pylint: enable=line-too-long
# Generates schema based on statistics files.
infer_schema = SchemaGen(stats=statistics_gen.outputs.output) # Step 3
# Performs anomaly detection based on statistics and data schema.
validate_stats = ExampleValidator( # Step 3
stats=statistics_gen.outputs.output, # Step 3
schema=infer_schema.outputs.output) # Step 3
# Performs transformations and feature engineering in training and serving.
transform = component.Transform( # Step 4
input_data=example_gen.outputs.examples, # Step 4
schema=infer_schema.outputs.output, # Step 4
module_file=_east_module_file) # Step 4
# Uses user-provided Python function that implements a model using TF-Learn.
trainer = Trainer( # Step 5
module_file=_east_module_file, # Step 5
transformed_examples=transform.outputs.transformed_examples, # Step 5
schema=infer_schema.outputs.output, # Step 5
transform_output=transform.outputs.transform_output, # Step 5
train_args=trainer_pb2.TrainArgs(num_steps=50), # Step 5
eval_args=trainer_pb2.EvalArgs(num_steps=10)) # Step 5
# Uses TFMA to compute a evaluation statistics over features of a model.
#model_analyzer = Evaluator( # Step 6
# examples=example_gen.outputs.examples, # Step 6
# model_exports=trainer.outputs.output, # Step 6
# feature_slicing_spec=evaluator_pb2.FeatureSlicingSpec(specs=[ # Step 6
# evaluator_pb2.SingleSlicingSpec( # Step 6
# column_for_slicing=['trip_start_hour']) # Step 6
# ])) # Step 6
# Performs quality validation of a candidate model (compared to a baseline).
# model_validator = ModelValidator( # Step 7
# examples=example_gen.outputs.examples, # Step 7
# model=trainer.outputs.output) # Step 7
# Checks whether the model passed the validation steps and pushes the model
# to a file destination if check passed.
# pusher = Pusher( # Step 7
# model_export=trainer.outputs.output, # Step 7
# model_blessing=model_validator.outputs.blessing, # Step 7
# push_destination=pusher_pb2.PushDestination( # Step 7
# filesystem=pusher_pb2.PushDestination.Filesystem( # Step 7
# base_directory=_serving_model_dir))) # Step 7
return [
example_gen,
statistics_gen, infer_schema, #validate_stats, # Step 3
transform, # Step 4
trainer, # Step 5
#model_analyzer, # Step 6
# model_validator, pusher # Step 7
]
def test_construct_missing_user_module(self):
with self.assertRaises(ValueError):
_ = component.Transform(
examples=self.examples,
schema=self.schema,
)