Fixed some typos and revised image embedder tests

This commit is contained in:
kinaryml 2022-11-10 02:30:17 -08:00
parent 7ec0d8cf3b
commit 0e9b925726
2 changed files with 40 additions and 45 deletions

View File

@ -100,28 +100,22 @@ class ImageEmbedderTest(parameterized.TestCase):
embedder = _ImageEmbedder.create_from_options(options) embedder = _ImageEmbedder.create_from_options(options)
self.assertIsInstance(embedder, _ImageEmbedder) self.assertIsInstance(embedder, _ImageEmbedder)
def _check_cosine_similarity(self, result0, result1, quantize, def _check_embedding_value(self, result, expected_first_value):
expected_similarity): # Check embedding first value.
# Checks head_index and head_name. self.assertAlmostEqual(result.embeddings[0].embedding[0],
self.assertEqual(result0.embeddings[0].head_index, 0) expected_first_value, delta=_EPSILON)
self.assertEqual(result1.embeddings[0].head_index, 0)
self.assertEqual(result0.embeddings[0].head_name, 'feature')
self.assertEqual(result1.embeddings[0].head_name, 'feature')
# Check embedding sizes. def _check_embedding_size(self, result, quantize, expected_embedding_size):
def _check_embedding_size(result): # Check embedding size.
self.assertLen(result.embeddings, 1) self.assertLen(result.embeddings, 1)
embedding_result = result.embeddings[0] embedding_result = result.embeddings[0]
self.assertLen(embedding_result.embedding, 1024) self.assertLen(embedding_result.embedding, expected_embedding_size)
if quantize: if quantize:
self.assertEqual(embedding_result.embedding.dtype, np.uint8) self.assertEqual(embedding_result.embedding.dtype, np.uint8)
else: else:
self.assertEqual(embedding_result.embedding.dtype, float) self.assertEqual(embedding_result.embedding.dtype, float)
# Checks results sizes. def _check_cosine_similarity(self, result0, result1, expected_similarity):
_check_embedding_size(result0)
_check_embedding_size(result1)
# Checks cosine similarity. # Checks cosine similarity.
similarity = _ImageEmbedder.cosine_similarity( similarity = _ImageEmbedder.cosine_similarity(
result0.embeddings[0], result1.embeddings[0]) result0.embeddings[0], result1.embeddings[0])
@ -129,13 +123,17 @@ class ImageEmbedderTest(parameterized.TestCase):
delta=_SIMILARITY_TOLERANCE) delta=_SIMILARITY_TOLERANCE)
@parameterized.parameters( @parameterized.parameters(
(False, False, False, ModelFileType.FILE_NAME, 0.925519, -0.2101883), (False, False, False, ModelFileType.FILE_NAME,
(True, False, False, ModelFileType.FILE_NAME, 0.925519, -0.0142344), 0.925519, 1024, (-0.2101883, -0.193027)),
# (False, True, False, ModelFileType.FILE_NAME, 0.926791, 229), (True, False, False, ModelFileType.FILE_NAME,
(False, False, True, ModelFileType.FILE_CONTENT, 0.999931, -0.195062) 0.925519, 1024, (-0.0142344, -0.0131606)),
# (False, True, False, ModelFileType.FILE_NAME,
# 0.926791, 1024, (229, 231)),
(False, False, True, ModelFileType.FILE_CONTENT,
0.999931, 1024, (-0.195062, -0.193027))
) )
def test_embed(self, l2_normalize, quantize, with_roi, model_file_type, def test_embed(self, l2_normalize, quantize, with_roi, model_file_type,
expected_similarity, expected_first_value): expected_similarity, expected_size, expected_first_values):
# Creates embedder. # Creates embedder.
if model_file_type is ModelFileType.FILE_NAME: if model_file_type is ModelFileType.FILE_NAME:
base_options = _BaseOptions(model_asset_path=self.model_path) base_options = _BaseOptions(model_asset_path=self.model_path)
@ -163,12 +161,13 @@ class ImageEmbedderTest(parameterized.TestCase):
image_result = embedder.embed(self.test_image, image_processing_options) image_result = embedder.embed(self.test_image, image_processing_options)
crop_result = embedder.embed(self.test_cropped_image) crop_result = embedder.embed(self.test_cropped_image)
# Check embedding value. # Checks embeddings and cosine similarity.
self.assertAlmostEqual(image_result.embeddings[0].embedding[0], expected_result0_value, expected_result1_value = expected_first_values
expected_first_value, delta=_EPSILON) self._check_embedding_size(image_result, quantize, expected_size)
self._check_embedding_size(crop_result, quantize, expected_size)
# Checks cosine similarity. self._check_embedding_value(image_result, expected_result0_value)
self._check_cosine_similarity(image_result, crop_result, quantize, self._check_embedding_value(crop_result, expected_result1_value)
self._check_cosine_similarity(image_result, crop_result,
expected_similarity) expected_similarity)
# Closes the embedder explicitly when the embedder is not used in # Closes the embedder explicitly when the embedder is not used in
# a context. # a context.
@ -201,7 +200,7 @@ class ImageEmbedderTest(parameterized.TestCase):
crop_result = embedder.embed(self.test_cropped_image) crop_result = embedder.embed(self.test_cropped_image)
# Checks cosine similarity. # Checks cosine similarity.
self._check_cosine_similarity(image_result, crop_result, quantize, self._check_cosine_similarity(image_result, crop_result,
expected_similarity) expected_similarity)
def test_missing_result_callback(self): def test_missing_result_callback(self):
@ -283,8 +282,7 @@ class ImageEmbedderTest(parameterized.TestCase):
timestamp) timestamp)
# Checks cosine similarity. # Checks cosine similarity.
self._check_cosine_similarity( self._check_cosine_similarity(
image_result, crop_result, quantize=False, image_result, crop_result, expected_similarity=0.925519)
expected_similarity=0.925519)
def test_embed_for_video_succeeds_with_region_of_interest(self): def test_embed_for_video_succeeds_with_region_of_interest(self):
options = _ImageEmbedderOptions( options = _ImageEmbedderOptions(
@ -305,8 +303,7 @@ class ImageEmbedderTest(parameterized.TestCase):
# Checks cosine similarity. # Checks cosine similarity.
self._check_cosine_similarity( self._check_cosine_similarity(
image_result, crop_result, quantize=False, image_result, crop_result, expected_similarity=0.999931)
expected_similarity=0.999931)
def test_calling_embed_in_live_stream_mode(self): def test_calling_embed_in_live_stream_mode(self):
options = _ImageEmbedderOptions( options = _ImageEmbedderOptions(
@ -352,7 +349,7 @@ class ImageEmbedderTest(parameterized.TestCase):
def check_result(result: ImageEmbedderResult, output_image: _Image, def check_result(result: ImageEmbedderResult, output_image: _Image,
timestamp_ms: int): timestamp_ms: int):
# Checks cosine similarity. # Checks cosine similarity.
self._check_cosine_similarity(result, crop_result, quantize=False, self._check_cosine_similarity(result, crop_result,
expected_similarity=0.925519) expected_similarity=0.925519)
self.assertTrue( self.assertTrue(
np.array_equal(output_image.numpy_view(), np.array_equal(output_image.numpy_view(),
@ -384,7 +381,7 @@ class ImageEmbedderTest(parameterized.TestCase):
def check_result(result: ImageEmbedderResult, output_image: _Image, def check_result(result: ImageEmbedderResult, output_image: _Image,
timestamp_ms: int): timestamp_ms: int):
# Checks cosine similarity. # Checks cosine similarity.
self._check_cosine_similarity(result, crop_result, quantize=False, self._check_cosine_similarity(result, crop_result,
expected_similarity=0.999931) expected_similarity=0.999931)
self.assertTrue( self.assertTrue(
np.array_equal(output_image.numpy_view(), np.array_equal(output_image.numpy_view(),

View File

@ -20,7 +20,6 @@ from mediapipe.python import packet_creator
from mediapipe.python import packet_getter from mediapipe.python import packet_getter
from mediapipe.python._framework_bindings import image as image_module from mediapipe.python._framework_bindings import image as image_module
from mediapipe.python._framework_bindings import packet as packet_module from mediapipe.python._framework_bindings import packet as packet_module
from mediapipe.python._framework_bindings import task_runner as task_runner_module
from mediapipe.tasks.cc.vision.image_embedder.proto import image_embedder_graph_options_pb2 from mediapipe.tasks.cc.vision.image_embedder.proto import image_embedder_graph_options_pb2
from mediapipe.tasks.cc.components.containers.proto import embeddings_pb2 from mediapipe.tasks.cc.components.containers.proto import embeddings_pb2
from mediapipe.tasks.python.components.processors import embedder_options from mediapipe.tasks.python.components.processors import embedder_options
@ -40,7 +39,6 @@ _EmbedderOptions = embedder_options.EmbedderOptions
_RunningMode = running_mode_module.VisionTaskRunningMode _RunningMode = running_mode_module.VisionTaskRunningMode
_TaskInfo = task_info_module.TaskInfo _TaskInfo = task_info_module.TaskInfo
_ImageProcessingOptions = image_processing_options_module.ImageProcessingOptions _ImageProcessingOptions = image_processing_options_module.ImageProcessingOptions
_TaskRunner = task_runner_module.TaskRunner
_EMBEDDINGS_OUT_STREAM_NAME = 'embeddings_out' _EMBEDDINGS_OUT_STREAM_NAME = 'embeddings_out'
_EMBEDDINGS_TAG = 'EMBEDDINGS' _EMBEDDINGS_TAG = 'EMBEDDINGS'
@ -112,7 +110,7 @@ class ImageEmbedder(base_vision_task_api.BaseVisionTaskApi):
`ImageEmbedderOptions`. `ImageEmbedderOptions`.
Raises: Raises:
ValueError: If failed to create `ImageClassifier` object from the provided ValueError: If failed to create `ImageEmbedder` object from the provided
file such as invalid file path. file such as invalid file path.
RuntimeError: If other types of error occurred. RuntimeError: If other types of error occurred.
""" """
@ -185,7 +183,7 @@ class ImageEmbedder(base_vision_task_api.BaseVisionTaskApi):
image_processing_options: Options for image processing. image_processing_options: Options for image processing.
Returns: Returns:
A embedding result object that contains a list of embeddings. An embedding result object that contains a list of embeddings.
Raises: Raises:
ValueError: If any of the input arguments is invalid. ValueError: If any of the input arguments is invalid.
@ -223,7 +221,7 @@ class ImageEmbedder(base_vision_task_api.BaseVisionTaskApi):
image_processing_options: Options for image processing. image_processing_options: Options for image processing.
Returns: Returns:
A embedding result object that contains a list of embeddings. An embedding result object that contains a list of embeddings.
Raises: Raises:
ValueError: If any of the input arguments is invalid. ValueError: If any of the input arguments is invalid.
@ -265,7 +263,7 @@ class ImageEmbedder(base_vision_task_api.BaseVisionTaskApi):
per input image. per input image.
The `result_callback` provides: The `result_callback` provides:
- A embedding result object that contains a list of embeddings. - An embedding result object that contains a list of embeddings.
- The input image that the image embedder runs on. - The input image that the image embedder runs on.
- The input timestamp in milliseconds. - The input timestamp in milliseconds.