def test_metalearner_nearest_neighbor(self):
meta_train_data = {}
metadata_indices = [1, 2, 3]
for ix, dataset_id in enumerate(metadata_indices):
hparams = standard_artifacts.HyperParameters()
hparams.uri = os.path.join(self._input_data_dir,
f'Tuner.train_mockdata_{dataset_id}',
'best_hyperparameters')
meta_train_data[f'hparams_train_{ix}'] = [hparams]
metafeature = artifacts.MetaFeatures()
metafeature.uri = os.path.join(
self._input_data_dir,
f'MetaFeatureGen.train_mockdata_{dataset_id}', 'metafeatures')
meta_train_data[f'meta_train_features_{ix}'] = [metafeature]
input_dict = {
**meta_train_data,
}
output_dict = {
executor.OUTPUT_HYPERPARAMS: [self._hparams_out],
executor.OUTPUT_MODEL: [self._model_out],
}
exec_properties = self._exec_properties.copy()
exec_properties['algorithm'] = executor.NEAREST_NEIGHBOR
ex = executor.MetaLearnerExecutor()
ex.Do(input_dict, output_dict, exec_properties)
self._verify_hparams_outputs(executor.NEAREST_NEIGHBOR)
self._verify_model_export()
def setUp(self):
super(ComponentTest, self).setUp()
num_train = 5
self.meta_train_data = {}
for ix in range(num_train):
self.meta_train_data[f'hparams_train_{ix}'] = channel_utils.as_channel(
[standard_artifacts.HyperParameters()])
self.meta_train_data[
f'meta_train_features_{ix}'] = channel_utils.as_channel(
[artifacts.MetaFeatures()])
self.custom_config = {'some': 'thing', 'some other': 1, 'thing': 2}
def testDoWithNearestNeighbor(self):
exec_properties = self._exec_properties.copy()
exec_properties['tuner_fn'] = '%s.%s' % (
tuner_module.tuner_fn.__module__, tuner_module.tuner_fn.__name__)
exec_properties['metalearning_algorithm'] = 'nearest_neighbor'
input_dict = self._input_dict.copy()
ps_type = ps_pb2.Type(
binary_classification=ps_pb2.BinaryClassification(label='class'))
ps = ps_pb2.ProblemStatement(
owner=['nitroml'],
tasks=[ps_pb2.Task(
name='mockdata_1',
type=ps_type,
)])
exec_properties['custom_config'] = json_utils.dumps({
'problem_statement':
text_format.MessageToString(message=ps, as_utf8=True),
})
hps_artifact = artifacts.KCandidateHyperParameters()
hps_artifact.uri = os.path.join(self._testdata_dir,
'MetaLearner.nearest_neighbor',
'hparams_out')
input_dict['warmup_hyperparameters'] = [hps_artifact]
model_artifact = standard_artifacts.Model()
model_artifact.uri = os.path.join(self._testdata_dir,
'MetaLearner.nearest_neighbor',
'model')
input_dict['metamodel'] = [model_artifact]
metafeature = artifacts.MetaFeatures()
metafeature.uri = os.path.join(self._testdata_dir,
'MetaFeatureGen.test_mockdata',
'metafeatures')
input_dict['metafeature'] = [metafeature]
tuner = executor.Executor(self._context)
tuner.Do(input_dict=input_dict,
output_dict=self._output_dict,
exec_properties=exec_properties)
self._verify_output()
def setUp(self):
super(ExecutorTest, self).setUp()
source_data_dir = os.path.dirname(os.path.dirname(__file__))
input_data_dir = os.path.join(source_data_dir, 'testdata')
statistics = standard_artifacts.ExampleStatistics()
statistics.uri = os.path.join(input_data_dir,
'StatisticsGen.train_mockdata_1',
'statistics', '5')
statistics.split_names = artifact_utils.encode_split_names(
['train', 'eval'])
transformed_examples = standard_artifacts.Examples()
transformed_examples.uri = os.path.join(input_data_dir,
'Transform.train_mockdata_1',
'transformed_examples', '10')
transformed_examples.split_names = artifact_utils.encode_split_names(
['train', 'eval'])
self._input_dict = {
executor.EXAMPLES_KEY: [transformed_examples],
executor.STATISTICS_KEY: [statistics],
}
output_data_dir = os.path.join(
os.environ.get('TEST_UNDECLARED_OUTPUTS_DIR',
tempfile.mkdtemp(dir=flags.FLAGS.test_tmpdir)),
self._testMethodName)
self._metafeatures = artifacts.MetaFeatures()
self._metafeatures.uri = output_data_dir
self._output_dict = {
executor.METAFEATURES_KEY: [self._metafeatures],
}
self._exec_properties = {
'custom_config': {
'problem_statement_path': '/some/fake/path'
}
}
def __init__(self,
statistics: types.Channel = None,
transformed_examples: Optional[types.Channel] = None,
custom_config: Optional[Dict[str, Any]] = None,
instance_name: Optional[str] = None):
"""Construct a MetaFeatureGen component.
Args:
statistics: Output channel from StatisticsGen.
transformed_examples: Optional channel from tfx Transform component.
custom_config: Optional dict which contains addtional parameters.
instance_name: Optional unique instance name. Necessary if multiple
MetaFeatureGen components are declared in the same pipeline.
"""
metafeatures = types.Channel(
type=artifacts.MetaFeatures, artifacts=[artifacts.MetaFeatures()])
spec = MetaFeatureGenSpec(
metafeatures=metafeatures,
transformed_examples=transformed_examples,
statistics=statistics,
custom_config=custom_config)
super(MetaFeatureGen, self).__init__(spec=spec, instance_name=instance_name)