Rename component "name" argument to "instance_name" and remove "component_name".

PiperOrigin-RevId: 266254963

Author: tfx-copybara

RELEASE.md

@@ -115,8 +115,9 @@
     equivalent overriding functionality is now available by specifying optional
     keyword arguments (see each component class definition for details).
 *   The optional arguments "executor" and "unique_name" of component classes
-    have been uniformly renamed to "executor_spec" and "name", respectively.
-    The "driver" optional argument of component classes is no longer available:
+    have been uniformly renamed to "executor_spec" and "instance_name",
+    respectively.
+*   The "driver" optional argument of component classes is no longer available:
     users who need to override the driver for a component should subclass the
     component and override the DRIVER_CLASS field.
 *   The `example_gen.component.ExampleGen` class has been refactored into the

tfx/components/base/base_component.py

@@ -55,14 +55,11 @@ class BaseComponent(with_metaclass(abc.ABCMeta, object)):
       this component (optional, defaults to base_driver.BaseDriver).
   """
 
-  # Subclasses may optionally override this component name property, which
-  # would otherwise default to the component class name. This is used for
-  # tracking component instances in pipelines and display in the UI.
-  COMPONENT_NAME = None
   # Subclasses must override this property (by specifying a types.ComponentSpec
   # class, e.g. "SPEC_CLASS = MyComponentSpec").
   SPEC_CLASS = _abstract_property()
   # Subclasses must also override the executor spec.
+  #
   # Note: EXECUTOR_CLASS has been replaced with EXECUTOR_SPEC. A custom
   # component's existing executor class definition "EXECUTOR_CLASS = MyExecutor"
   # should be replaced with "EXECUTOR_SPEC = ExecutorClassSpec(MyExecutor).
@@ -75,21 +72,16 @@ def __init__(
       self,
       spec: types.ComponentSpec,
       custom_executor_spec: Optional[executor_spec.ExecutorSpec] = None,
-      name: Optional[Text] = None,
-      component_name: Optional[Text] = None):
+      instance_name: Optional[Text] = None):
     """Initialize a component.
 
     Args:
       spec: types.ComponentSpec object for this component instance.
       custom_executor_spec: Optional custom executor spec overriding the default
         executor specified in the component attribute.
-      name: Optional unique identifying name for this instance of the component
-        in the pipeline. Required if two instances of the same component is used
-        in the pipeline.
-      component_name: Optional component name override for this instance of the
-        component. This will override the class-level COMPONENT_NAME attribute
-        and the name of the class (which would otherwise be used as the
-        component name).
+      instance_name: Optional unique identifying name for this instance of the
+        component in the pipeline. Required if two instances of the same
+        component is used in the pipeline.
     """
     self.spec = spec
     if custom_executor_spec:
@@ -99,10 +91,8 @@ def __init__(
              'ExecutorSpec') % (custom_executor_spec, self.__class__))
     self.executor_spec = (custom_executor_spec or self.__class__.EXECUTOR_SPEC)
     self.driver_class = self.__class__.DRIVER_CLASS
-    self.component_name = (
-        component_name or self.__class__.COMPONENT_NAME or
-        self.__class__.__name__)
-    self.name = name
+    # TODO(b/139540680): consider making instance_name private.
+    self.instance_name = instance_name
     self._upstream_nodes = set()
     self._downstream_nodes = set()
     self._validate_component_class()
@@ -141,10 +131,10 @@ def _validate_spec(self, spec):
           (self.__class__, self.__class__.SPEC_CLASS, spec))
 
   def __repr__(self):
-    return ('%s(spec: %s, executor_spec: %s, driver_class: %s, name: %s, '
-            'inputs: %s, outputs: %s)') % (
+    return ('%s(spec: %s, executor_spec: %s, driver_class: %s, '
+            'component_id: %s, inputs: %s, outputs: %s)') % (
                 self.__class__.__name__, self.spec, self.executor_spec,
-                self.driver_class, self.name, self.inputs, self.outputs)
+                self.driver_class, self.component_id, self.inputs, self.outputs)
 
   @property
   def component_type(self) -> Text:
@@ -166,23 +156,24 @@ def exec_properties(self) -> Dict[Text, Any]:
   # we will have two component level keys:
   # - component_type: the path of the python executor or the image uri of the
   #   executor.
-  # - component_id: <component_name>.<unique_name>
+  # - component_id: <component_class_name>.<unique_name>
   @property
   def component_id(self):
-    """Component id.
+    """Component id, unique across all component instances in a pipeline.
 
     If unique name is available, component_id will be:
-      <component_name>.<unique_name>
+      <component_class_name>.<instance_name>
     otherwise, component_id will be:
-      <component_name>
+      <component_class_name>
 
     Returns:
       component id.
     """
-    if self.name:
-      return '{}.{}'.format(self.component_name, self.name)
+    component_class_name = self.__class__.__name__
+    if self.instance_name:
+      return '{}.{}'.format(component_class_name, self.instance_name)
     else:
-      return self.component_name
+      return component_class_name
 
   @property
   def upstream_nodes(self):

tfx/components/base/base_component_test.py

@@ -43,7 +43,6 @@ class _BasicComponentSpec(types.ComponentSpec):
 
 class _BasicComponent(base_component.BaseComponent):
 
-  COMPONENT_NAME = 'MyBasicComponent'
   SPEC_CLASS = _BasicComponentSpec
   EXECUTOR_SPEC = executor_spec.ExecutorClassSpec(base_executor.BaseExecutor)
 
@@ -63,7 +62,7 @@ class ComponentTest(tf.test.TestCase):
   def testComponentBasic(self):
     input_channel = types.Channel(type_name='InputType')
     component = _BasicComponent(folds=10, input=input_channel)
-    self.assertEqual(component.component_name, 'MyBasicComponent')
+    self.assertEqual(component.component_id, '_BasicComponent')
     self.assertIs(input_channel, component.inputs.input)
     self.assertIsInstance(component.outputs.output, types.Channel)
     self.assertEqual(component.outputs.output.type_name, 'OutputType')

tfx/components/evaluator/component.py

@@ -75,7 +75,7 @@ def __init__(
       feature_slicing_spec: Optional[evaluator_pb2.FeatureSlicingSpec] = None,
       output: Optional[types.Channel] = None,
       model: Optional[types.Channel] = None,
-      name: Optional[Text] = None):
+      instance_name: Optional[Text] = None):
     """Construct an Evaluator component.
 
     Args:
@@ -89,12 +89,12 @@ def __init__(
         instance that describes how Evaluator should slice the data.
       output: Channel of `ModelEvalPath` to store the evaluation results.
       model: Future replacement of the `model_exports` argument.
-      name: Name assigned to this specific instance of Evaluator. Required
-        only if multiple Evaluator components are declared in the same pipeline.
+      instance_name: Optional name assigned to this specific instance of
+        Evaluator. Required only if multiple Evaluator components are declared
+        in the same pipeline.
 
       Either `model_exports` or `model` must be present in the input arguments.
     """
-
     model_exports = model_exports or model
     output = output or types.Channel(
         type=standard_artifacts.ModelEvaluation,
@@ -105,4 +105,4 @@ def __init__(
         feature_slicing_spec=(feature_slicing_spec or
                               evaluator_pb2.FeatureSlicingSpec()),
         output=output)
-    super(Evaluator, self).__init__(spec=spec, name=name)
+    super(Evaluator, self).__init__(spec=spec, instance_name=instance_name)

tfx/components/example_gen/big_query_example_gen/component.py

@@ -41,7 +41,7 @@ def __init__(self,
                input_config: Optional[example_gen_pb2.Input] = None,
                output_config: Optional[example_gen_pb2.Output] = None,
                example_artifacts: Optional[types.Channel] = None,
-               name: Optional[Text] = None):
+               instance_name: Optional[Text] = None):
     """Constructs a BigQueryExampleGen component.
 
     Args:
@@ -55,8 +55,8 @@ def __init__(self,
         size 2:1.
       example_artifacts: Optional channel of 'ExamplesPath' for output train and
         eval examples.
-      name: Optional unique name. Necessary if multiple BigQueryExampleGen
-        components are declared in the same pipeline.
+      instance_name: Optional unique instance name. Necessary if multiple
+        BigQueryExampleGen components are declared in the same pipeline.
 
     Raises:
       RuntimeError: Only one of query and input_config should be set.
@@ -68,4 +68,4 @@ def __init__(self,
         input_config=input_config,
         output_config=output_config,
         example_artifacts=example_artifacts,
-        name=name)
+        instance_name=instance_name)

tfx/components/example_gen/component.py

@@ -57,9 +57,8 @@ def __init__(self,
                input_config: example_gen_pb2.Input,
                output_config: Optional[example_gen_pb2.Output] = None,
                custom_config: Optional[example_gen_pb2.CustomConfig] = None,
-               component_name: Optional[Text] = 'ExampleGen',
                example_artifacts: Optional[types.Channel] = None,
-               name: Optional[Text] = None):
+               instance_name: Optional[Text] = None):
     """Construct an QueryBasedExampleGen component.
 
     Args:
@@ -73,13 +72,10 @@ def __init__(self,
       custom_config: An
         [example_gen_pb2.CustomConfig](https://github.com/tensorflow/tfx/blob/master/tfx/proto/example_gen.proto)
         instance, providing custom configuration for ExampleGen.
-      component_name: Name of the component.  This should be unique per
-        component class. Default to 'ExampleGen' and may be overwritten by
-        subclasses.
       example_artifacts: Channel of 'ExamplesPath' for output train and
         eval examples.
-      name: Unique name. Required only if multiple ExampleGen components are
-        declared in the same pipeline.
+      instance_name: Optional unique instance name. Required only if multiple
+        ExampleGen components are declared in the same pipeline.
     """
     # Configure outputs.
     output_config = output_config or utils.make_default_output_config(
@@ -94,7 +90,8 @@ def __init__(self,
         output_config=output_config,
         custom_config=custom_config,
         examples=example_artifacts)
-    super(_QueryBasedExampleGen, self).__init__(spec=spec, name=name)
+    super(_QueryBasedExampleGen, self).__init__(
+        spec=spec, instance_name=instance_name)
 
 
 class FileBasedExampleGen(base_component.BaseComponent):
@@ -127,11 +124,10 @@ def __init__(
       input_config: Optional[example_gen_pb2.Input] = None,
       output_config: Optional[example_gen_pb2.Output] = None,
       custom_config: Optional[example_gen_pb2.CustomConfig] = None,
-      component_name: Optional[Text] = 'ExampleGen',
       example_artifacts: Optional[types.Channel] = None,
       custom_executor_spec: Optional[executor_spec.ExecutorSpec] = None,
       input: Optional[types.Channel] = None,  # pylint: disable=redefined-builtin
-      name: Optional[Text] = None):
+      instance_name: Optional[Text] = None):
     """Construct a FileBasedExampleGen component.
 
     Args:
@@ -147,16 +143,13 @@ def __init__(
         'eval' with size 2:1.
       custom_config: An optional example_gen_pb2.CustomConfig instance,
         providing custom configuration for executor.
-      component_name: Name of the component, should be unique per component
-        class. Do not use -- will be deprecated in a future release.
       example_artifacts: Channel of 'ExamplesPath' for output train and
         eval examples.
       custom_executor_spec: Optional custom executor spec overriding the default
         executor spec specified in the component attribute.
       input: Future replacement of the 'input_base' argument.
-      name: Name assigned to this specific instance of FileBasedExampleGen.
-        Required only if multiple FileBasedExampleGen components are declared in
-        the same pipeline.
+      instance_name: Optional unique instance name. Required only if multiple
+        ExampleGen components are declared in the same pipeline.
 
       Either `input_base` or `input` must be present in the input arguments.
     """
@@ -177,4 +170,6 @@ def __init__(
         custom_config=custom_config,
         examples=example_artifacts)
     super(FileBasedExampleGen, self).__init__(
-        spec=spec, custom_executor_spec=custom_executor_spec, name=name)
+        spec=spec,
+        custom_executor_spec=custom_executor_spec,
+        instance_name=instance_name)

tfx/components/example_gen/component_test.py

@@ -44,12 +44,12 @@ def __init__(self,
                input_config,
                output_config=None,
                example_artifacts=None,
-               name=None):
+               instance_name=None):
     super(TestQueryBasedExampleGenComponent, self).__init__(
         input_config=input_config,
         output_config=output_config,
         example_artifacts=example_artifacts,
-        name=name)
+        instance_name=instance_name)
 
 
 class TestFileBasedExampleGenComponent(component.FileBasedExampleGen):
@@ -61,13 +61,13 @@ def __init__(self,
                input_config=None,
                output_config=None,
                example_artifacts=None,
-               name=None):
+               instance_name=None):
     super(TestFileBasedExampleGenComponent, self).__init__(
         input_base=input_base,
         input_config=input_config,
         output_config=output_config,
         example_artifacts=example_artifacts,
-        name=name)
+        instance_name=instance_name)
 
 
 class ComponentTest(tf.test.TestCase):

tfx/components/example_gen/csv_example_gen/component.py

@@ -41,7 +41,7 @@ def __init__(self,
                output_config: Optional[example_gen_pb2.Output] = None,
                example_artifacts: Optional[types.Channel] = None,
                input: Optional[types.Channel] = None,  # pylint: disable=redefined-builtin
-               name: Optional[Text] = None):
+               instance_name: Optional[Text] = None):
     """Construct a CsvExampleGen component.
 
     Args:
@@ -56,13 +56,13 @@ def __init__(self,
       example_artifacts: Optional channel of 'ExamplesPath' for output train and
         eval examples.
       input: Forwards compatibility alias for the 'input_base' argument.
-      name: Optional unique name. Necessary if multiple CsvExampleGen components
-        are declared in the same pipeline.
+      instance_name: Optional unique instance name. Necessary if multiple
+        CsvExampleGen components are declared in the same pipeline.
     """
     super(CsvExampleGen, self).__init__(
         input_base=input_base,
         input_config=input_config,
         output_config=output_config,
         example_artifacts=example_artifacts,
         input=input,
-        name=name)
+        instance_name=instance_name)

tfx/components/example_gen/custom_executors/avro_component_test.py

@@ -62,7 +62,7 @@ def testRun(self, mock_publisher):
         input_base=external_input(self.avro_dir_path),
         input_config=self.input_config,
         output_config=self.output_config,
-        name='AvroExampleGenComponent')
+        instance_name='AvroExampleGen')
 
     output_data_dir = os.path.join(
         os.environ.get('TEST_UNDECLARED_OUTPUTS_DIR', self.get_temp_dir()),
@@ -92,7 +92,7 @@ def testRun(self, mock_publisher):
     mock_publisher.return_value.publish_execution.assert_called_once()
 
     # Get output paths.
-    component_id = '.'.join([example_gen.component_name, example_gen.name])
+    component_id = example_gen.component_id
     output_path = os.path.join(pipeline_root, component_id, 'examples/1')
     train_examples = standard_artifacts.Examples(split='train')
     train_examples.uri = os.path.join(output_path, 'train')

tfx/components/example_gen/custom_executors/parquet_component_test.py

@@ -63,7 +63,7 @@ def testRun(self, mock_publisher):
         input_base=external_input(self.parquet_dir_path),
         input_config=self.input_config,
         output_config=self.output_config,
-        name='parquetExampleGenComponent')
+        instance_name='ParquetExampleGen')
 
     output_data_dir = os.path.join(
         os.environ.get('TEST_UNDECLARED_OUTPUTS_DIR', self.get_temp_dir()),
@@ -93,7 +93,7 @@ def testRun(self, mock_publisher):
     mock_publisher.return_value.publish_execution.assert_called_once()
 
     # Get output paths.
-    component_id = '.'.join([example_gen.component_name, example_gen.name])
+    component_id = example_gen.component_id
     output_path = os.path.join(pipeline_root, component_id, 'examples/1')
     train_examples = standard_artifacts.Examples(split='train')
     train_examples.uri = os.path.join(output_path, 'train')

tfx/components/example_gen/import_example_gen/component.py

@@ -43,7 +43,7 @@ def __init__(self,
                output_config: Optional[example_gen_pb2.Output] = None,
                example_artifacts: Optional[types.Channel] = None,
                input: Optional[types.Channel] = None,  # pylint: disable=redefined-builtin
-               name: Optional[Text] = None):
+               instance_name: Optional[Text] = None):
     """Construct an ImportExampleGen component.
 
     Args:
@@ -59,13 +59,13 @@ def __init__(self,
       example_artifacts: Optional channel of 'ExamplesPath' for output train and
         eval examples.
       input: Forwards compatibility alias for the 'input_base' argument.
-      name: Optional unique name. Necessary if multiple ImportExampleGen
-        components are declared in the same pipeline.
+      instance_name: Optional unique instance name. Necessary if multiple
+        ImportExampleGen components are declared in the same pipeline.
     """
     super(ImportExampleGen, self).__init__(
         input_base=input_base,
         input_config=input_config,
         output_config=output_config,
         example_artifacts=example_artifacts,
         input=input,
-        name=name)
+        instance_name=instance_name)