Merge branch 'google:master' into image-embedder-python

WORKSPACE

@@ -155,6 +155,7 @@ http_archive(
     name = "com_google_audio_tools",
     strip_prefix = "multichannel-audio-tools-master",
     urls = ["https://github.com/google/multichannel-audio-tools/archive/master.zip"],
+    repo_mapping = {"@com_github_glog_glog" : "@com_github_glog_glog_no_gflags"},
 )
 
 http_archive(

docs/BUILD

@@ -12,3 +12,21 @@ py_binary(
         "//third_party/py/tensorflow_docs/api_generator:public_api",
     ],
 )
+
+py_binary(
+    name = "build_java_api_docs",
+    srcs = ["build_java_api_docs.py"],
+    data = [
+        "//third_party/java/doclava/current:doclava.jar",
+        "//third_party/java/jsilver:jsilver_jar",
+    ],
+    env = {
+        "DOCLAVA_JAR": "$(location //third_party/java/doclava/current:doclava.jar)",
+        "JSILVER_JAR": "$(location //third_party/java/jsilver:jsilver_jar)",
+    },
+    deps = [
+        "//third_party/py/absl:app",
+        "//third_party/py/absl/flags",
+        "//third_party/py/tensorflow_docs/api_generator/gen_java",
+    ],
+)

docs/build_java_api_docs.py

@@ -0,0 +1,58 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+"""Generate Java reference docs for MediaPipe."""
+import pathlib
+
+from absl import app
+from absl import flags
+
+from tensorflow_docs.api_generator import gen_java
+
+_JAVA_ROOT = flags.DEFINE_string('java_src', None,
+                                 'Override the Java source path.',
+                                 required=False)
+
+_OUT_DIR = flags.DEFINE_string('output_dir', '/tmp/mp_java/',
+                               'Write docs here.')
+
+_SITE_PATH = flags.DEFINE_string('site_path', '/mediapipe/api_docs/java',
+                                 'Path prefix in the _toc.yaml')
+
+_ = flags.DEFINE_string('code_url_prefix', None,
+                        '[UNUSED] The url prefix for links to code.')
+
+_ = flags.DEFINE_bool(
+    'search_hints', True,
+    '[UNUSED] Include metadata search hints in the generated files')
+
+
+def main(_) -> None:
+  if not (java_root := _JAVA_ROOT.value):
+    # Default to using a relative path to find the Java source.
+    mp_root = pathlib.Path(__file__)
+    while (mp_root := mp_root.parent).name != 'mediapipe':
+      # Find the nearest `mediapipe` dir.
+      pass
+    java_root = mp_root / 'tasks/java'
+
+  gen_java.gen_java_docs(
+      package='com.google.mediapipe',
+      source_path=pathlib.Path(java_root),
+      output_dir=pathlib.Path(_OUT_DIR.value),
+      site_path=pathlib.Path(_SITE_PATH.value))
+
+
+if __name__ == '__main__':
+  app.run(main)

docs/build_py_api_docs.py

@@ -1,4 +1,4 @@
-# Copyright 2022 The TensorFlow Authors. All Rights Reserved.
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

mediapipe/__init__.py

@@ -1,14 +1,13 @@
-"""Copyright 2019 - 2020 The MediaPipe Authors.
+# Copyright 2019 - 2022 The MediaPipe Authors.
-
+#
-Licensed under the Apache License, Version 2.0 (the "License");
+# Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
+# you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
+# You may obtain a copy of the License at
-
+#
-     http://www.apache.org/licenses/LICENSE-2.0
+#      http://www.apache.org/licenses/LICENSE-2.0
-
+#
-Unless required by applicable law or agreed to in writing, software
+# Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
+# distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
+# See the License for the specific language governing permissions and
-limitations under the License.
+# limitations under the License.
-"""

mediapipe/calculators/audio/BUILD

@@ -277,6 +277,7 @@ cc_test(
         "//mediapipe/framework/port:gtest_main",
         "//mediapipe/framework/port:status",
         "//mediapipe/util:time_series_test_util",
+        "@com_google_audio_tools//audio/dsp/mfcc",
         "@eigen_archive//:eigen3",
     ],
 )
@@ -352,6 +353,8 @@ cc_test(
         "//mediapipe/framework/port:status",
         "//mediapipe/framework/tool:validate_type",
         "//mediapipe/util:time_series_test_util",
+        "@com_google_audio_tools//audio/dsp:resampler",
+        "@com_google_audio_tools//audio/dsp:resampler_q",
         "@com_google_audio_tools//audio/dsp:signal_vector_util",
         "@eigen_archive//:eigen3",
     ],

mediapipe/calculators/core/bypass_calculator.cc

@@ -130,7 +130,7 @@ class BypassCalculator : public Node {
       pass_out.insert(entry.second);
       auto& packet = cc->Inputs().Get(entry.first).Value();
       if (packet.Timestamp() == cc->InputTimestamp()) {
-        cc->Outputs().Get(entry.first).AddPacket(packet);
+        cc->Outputs().Get(entry.second).AddPacket(packet);
       }
     }
     Timestamp bound = cc->InputTimestamp().NextAllowedInStream();

mediapipe/calculators/core/bypass_calculator_test.cc

@@ -42,10 +42,10 @@ constexpr char kTestGraphConfig1[] = R"pb(
   node {
     calculator: "BypassCalculator"
     input_stream: "PASS:appearances"
-    input_stream: "TRUNCATE:0:video_frame"
+    input_stream: "IGNORE:0:video_frame"
-    input_stream: "TRUNCATE:1:feature_config"
+    input_stream: "IGNORE:1:feature_config"
     output_stream: "PASS:passthrough_appearances"
-    output_stream: "TRUNCATE:passthrough_federated_gaze_output"
+    output_stream: "IGNORE:passthrough_federated_gaze_output"
     node_options: {
       [type.googleapis.com/mediapipe.BypassCalculatorOptions] {
         pass_input_stream: "PASS"

mediapipe/calculators/tensor/BUILD

@@ -156,6 +156,7 @@ cc_test(
         "//mediapipe/framework/port:parse_text_proto",
         "@com_google_absl//absl/status",
         "@com_google_absl//absl/strings",
+        "@com_google_audio_tools//audio/dsp:window_functions",
     ],
 )
 

mediapipe/gpu/BUILD

@@ -750,19 +750,12 @@ objc_library(
     ],
 )
 
-proto_library(
+mediapipe_proto_library(
     name = "scale_mode_proto",
     srcs = ["scale_mode.proto"],
     visibility = ["//visibility:public"],
 )
 
-mediapipe_cc_proto_library(
-    name = "scale_mode_cc_proto",
-    srcs = ["scale_mode.proto"],
-    visibility = ["//visibility:public"],
-    deps = [":scale_mode_proto"],
-)
-
 cc_library(
     name = "gl_quad_renderer",
     srcs = ["gl_quad_renderer.cc"],

mediapipe/model_maker/models/gesture_recognizer/BUILD

@@ -0,0 +1,51 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+load(
+    "//mediapipe/framework/tool:mediapipe_files.bzl",
+    "mediapipe_files",
+)
+
+licenses(["notice"])
+
+package(
+    default_visibility = ["//mediapipe/model_maker/python/vision/gesture_recognizer:__subpackages__"],
+)
+
+mediapipe_files(
+    srcs = [
+        "canned_gesture_classifier.tflite",
+        "gesture_embedder.tflite",
+        "gesture_embedder/keras_metadata.pb",
+        "gesture_embedder/saved_model.pb",
+        "gesture_embedder/variables/variables.data-00000-of-00001",
+        "gesture_embedder/variables/variables.index",
+        "hand_landmark_full.tflite",
+        "palm_detection_full.tflite",
+    ],
+)
+
+filegroup(
+    name = "models",
+    srcs = [
+        "canned_gesture_classifier.tflite",
+        "gesture_embedder.tflite",
+        "gesture_embedder/keras_metadata.pb",
+        "gesture_embedder/saved_model.pb",
+        "gesture_embedder/variables/variables.data-00000-of-00001",
+        "gesture_embedder/variables/variables.index",
+        "hand_landmark_full.tflite",
+        "palm_detection_full.tflite",
+    ],
+)

mediapipe/model_maker/python/text/core/BUILD

@@ -0,0 +1,35 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Placeholder for internal Python strict library and test compatibility macro.
+
+package(
+    default_visibility = ["//mediapipe:__subpackages__"],
+)
+
+licenses(["notice"])
+
+py_library(
+    name = "bert_model_options",
+    srcs = ["bert_model_options.py"],
+)
+
+py_library(
+    name = "bert_model_spec",
+    srcs = ["bert_model_spec.py"],
+    deps = [
+        ":bert_model_options",
+        "//mediapipe/model_maker/python/core:hyperparameters",
+    ],
+)

mediapipe/model_maker/python/text/core/__init__.py

@@ -0,0 +1,13 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.

mediapipe/model_maker/python/text/core/bert_model_options.py

@@ -0,0 +1,33 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Configurable model options for a BERT model."""
+
+import dataclasses
+
+
+@dataclasses.dataclass
+class BertModelOptions:
+  """Configurable model options for a BERT model.
+
+  See https://arxiv.org/abs/1810.04805 (BERT: Pre-training of Deep Bidirectional
+  Transformers for Language Understanding) for more details.
+
+    Attributes:
+      seq_len: Length of the sequence to feed into the model.
+      do_fine_tuning: If true, then the BERT model is not frozen for training.
+      dropout_rate: The rate for dropout.
+  """
+  seq_len: int = 128
+  do_fine_tuning: bool = True
+  dropout_rate: float = 0.1

mediapipe/model_maker/python/text/core/bert_model_spec.py

@@ -0,0 +1,58 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Specification for a BERT model."""
+
+import dataclasses
+from typing import Dict
+
+from mediapipe.model_maker.python.core import hyperparameters as hp
+from mediapipe.model_maker.python.text.core import bert_model_options
+
+_DEFAULT_TFLITE_INPUT_NAME = {
+    'ids': 'serving_default_input_word_ids:0',
+    'mask': 'serving_default_input_mask:0',
+    'segment_ids': 'serving_default_input_type_ids:0'
+}
+
+
+@dataclasses.dataclass
+class BertModelSpec:
+  """Specification for a BERT model.
+
+  See https://arxiv.org/abs/1810.04805 (BERT: Pre-training of Deep Bidirectional
+  Transformers for Language Understanding) for more details.
+
+    Attributes:
+      hparams: Hyperparameters used for training.
+      model_options: Configurable options for a BERT model.
+      do_lower_case: boolean, whether to lower case the input text. Should be
+        True / False for uncased / cased models respectively, where the models
+        are specified by the `uri`.
+      tflite_input_name: Dict, input names for the TFLite model.
+      uri: URI for the BERT module.
+      name: The name of the object.
+  """
+
+  hparams: hp.BaseHParams = hp.BaseHParams(
+      epochs=3,
+      batch_size=32,
+      learning_rate=3e-5,
+      distribution_strategy='mirrored')
+  model_options: bert_model_options.BertModelOptions = (
+      bert_model_options.BertModelOptions())
+  do_lower_case: bool = True
+  tflite_input_name: Dict[str, str] = dataclasses.field(
+      default_factory=lambda: _DEFAULT_TFLITE_INPUT_NAME)
+  uri: str = 'https://tfhub.dev/tensorflow/bert_en_uncased_L-12_H-768_A-12/1'
+  name: str = 'Bert'

mediapipe/model_maker/python/text/text_classifier/BUILD

@@ -0,0 +1,146 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Placeholder for internal Python strict library and test compatibility macro.
+# Placeholder for internal Python strict test compatibility macro.
+
+package(
+    default_visibility = ["//mediapipe:__subpackages__"],
+)
+
+licenses(["notice"])
+
+py_library(
+    name = "model_options",
+    srcs = ["model_options.py"],
+    deps = ["//mediapipe/model_maker/python/text/core:bert_model_options"],
+)
+
+py_library(
+    name = "model_spec",
+    srcs = ["model_spec.py"],
+    deps = [
+        ":model_options",
+        "//mediapipe/model_maker/python/core:hyperparameters",
+        "//mediapipe/model_maker/python/text/core:bert_model_spec",
+    ],
+)
+
+py_test(
+    name = "model_spec_test",
+    srcs = ["model_spec_test.py"],
+    deps = [
+        ":model_options",
+        ":model_spec",
+        "//mediapipe/model_maker/python/core:hyperparameters",
+    ],
+)
+
+py_library(
+    name = "dataset",
+    srcs = ["dataset.py"],
+    deps = ["//mediapipe/model_maker/python/core/data:classification_dataset"],
+)
+
+py_test(
+    name = "dataset_test",
+    srcs = ["dataset_test.py"],
+    deps = [":dataset"],
+)
+
+py_library(
+    name = "preprocessor",
+    srcs = ["preprocessor.py"],
+    deps = [":dataset"],
+)
+
+py_test(
+    name = "preprocessor_test",
+    srcs = ["preprocessor_test.py"],
+    tags = ["requires-net:external"],
+    deps = [
+        ":dataset",
+        ":model_spec",
+        ":preprocessor",
+    ],
+)
+
+py_library(
+    name = "text_classifier_options",
+    srcs = ["text_classifier_options.py"],
+    deps = [
+        ":model_options",
+        ":model_spec",
+        "//mediapipe/model_maker/python/core:hyperparameters",
+    ],
+)
+
+py_library(
+    name = "text_classifier",
+    srcs = ["text_classifier.py"],
+    deps = [
+        ":dataset",
+        ":model_options",
+        ":model_spec",
+        ":preprocessor",
+        ":text_classifier_options",
+        "//mediapipe/model_maker/python/core:hyperparameters",
+        "//mediapipe/model_maker/python/core/data:dataset",
+        "//mediapipe/model_maker/python/core/tasks:classifier",
+        "//mediapipe/model_maker/python/core/utils:model_util",
+        "//mediapipe/model_maker/python/core/utils:quantization",
+        "//mediapipe/tasks/python/metadata/metadata_writers:metadata_writer",
+        "//mediapipe/tasks/python/metadata/metadata_writers:text_classifier",
+    ],
+)
+
+py_test(
+    name = "text_classifier_test",
+    size = "large",
+    srcs = ["text_classifier_test.py"],
+    data = [
+        "//mediapipe/model_maker/python/text/text_classifier/testdata",
+    ],
+    tags = ["requires-net:external"],
+    deps = [
+        ":dataset",
+        ":model_options",
+        ":model_spec",
+        ":text_classifier",
+        ":text_classifier_options",
+        "//mediapipe/model_maker/python/core:hyperparameters",
+        "//mediapipe/tasks/python/test:test_utils",
+    ],
+)
+
+py_library(
+    name = "text_classifier_demo_lib",
+    srcs = ["text_classifier_demo.py"],
+    deps = [
+        ":dataset",
+        ":model_spec",
+        ":text_classifier",
+        ":text_classifier_options",
+        "//mediapipe/model_maker/python/core:hyperparameters",
+        "//mediapipe/model_maker/python/core/utils:quantization",
+    ],
+)
+
+py_binary(
+    name = "text_classifier_demo",
+    srcs = ["text_classifier_demo.py"],
+    deps = [
+        ":text_classifier_demo_lib",
+    ],
+)

mediapipe/model_maker/python/text/text_classifier/__init__.py

@@ -0,0 +1,13 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.

mediapipe/model_maker/python/text/text_classifier/dataset.py

@@ -0,0 +1,88 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Text classifier dataset library."""
+
+import csv
+import dataclasses
+import random
+
+from typing import Optional, Sequence
+import tensorflow as tf
+
+from mediapipe.model_maker.python.core.data import classification_dataset
+
+
+@dataclasses.dataclass
+class CSVParameters:
+  """Parameters used when reading a CSV file.
+
+  Attributes:
+    text_column: Column name for the input text.
+    label_column: Column name for the labels.
+    fieldnames: Sequence of keys for the CSV columns. If None, the first row of
+      the CSV file is used as the keys.
+    delimiter: Character that separates fields.
+    quotechar: Character used to quote fields that contain special characters
+      like the `delimiter`.
+  """
+  text_column: str
+  label_column: str
+  fieldnames: Optional[Sequence[str]] = None
+  delimiter: str = ","
+  quotechar: str = '"'
+
+
+class Dataset(classification_dataset.ClassificationDataset):
+  """Dataset library for text classifier."""
+
+  @classmethod
+  def from_csv(cls,
+               filename: str,
+               csv_params: CSVParameters,
+               shuffle: bool = True) -> "Dataset":
+    """Loads text with labels from a CSV file.
+
+    Args:
+      filename: Name of the CSV file.
+      csv_params: Parameters used for reading the CSV file.
+      shuffle: If True, randomly shuffle the data.
+
+    Returns:
+      Dataset containing (text, label) pairs and other related info.
+    """
+    with tf.io.gfile.GFile(filename, "r") as f:
+      reader = csv.DictReader(
+          f,
+          fieldnames=csv_params.fieldnames,
+          delimiter=csv_params.delimiter,
+          quotechar=csv_params.quotechar)
+
+    lines = list(reader)
+    if shuffle:
+      random.shuffle(lines)
+
+    label_names = sorted(set([line[csv_params.label_column] for line in lines]))
+    index_by_label = {label: index for index, label in enumerate(label_names)}
+
+    texts = [line[csv_params.text_column] for line in lines]
+    text_ds = tf.data.Dataset.from_tensor_slices(tf.cast(texts, tf.string))
+    label_indices = [
+        index_by_label[line[csv_params.label_column]] for line in lines
+    ]
+    label_index_ds = tf.data.Dataset.from_tensor_slices(
+        tf.cast(label_indices, tf.int64))
+    text_label_ds = tf.data.Dataset.zip((text_ds, label_index_ds))
+
+    return Dataset(
+        dataset=text_label_ds, size=len(texts), label_names=label_names)

mediapipe/model_maker/python/text/text_classifier/dataset_test.py

@@ -0,0 +1,75 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import csv
+import os
+
+import tensorflow as tf
+
+from mediapipe.model_maker.python.text.text_classifier import dataset
+
+
+class DatasetTest(tf.test.TestCase):
+
+  def _get_csv_file(self):
+    labels_and_text = (('neutral', 'indifferent'), ('pos', 'extremely great'),
+                       ('neg', 'totally awful'), ('pos', 'super good'),
+                       ('neg', 'really bad'))
+    csv_file = os.path.join(self.get_temp_dir(), 'data.csv')
+    if os.path.exists(csv_file):
+      return csv_file
+    fieldnames = ['text', 'label']
+    with open(csv_file, 'w') as f:
+      writer = csv.DictWriter(f, fieldnames=fieldnames)
+      writer.writeheader()
+      for label, text in labels_and_text:
+        writer.writerow({'text': text, 'label': label})
+    return csv_file
+
+  def test_from_csv(self):
+    csv_file = self._get_csv_file()
+    csv_params = dataset.CSVParameters(text_column='text', label_column='label')
+    data = dataset.Dataset.from_csv(filename=csv_file, csv_params=csv_params)
+    self.assertLen(data, 5)
+    self.assertEqual(data.num_classes, 3)
+    self.assertEqual(data.label_names, ['neg', 'neutral', 'pos'])
+    data_values = set([(text.numpy()[0], label.numpy()[0])
+                       for text, label in data.gen_tf_dataset()])
+    expected_data_values = set([(b'indifferent', 1), (b'extremely great', 2),
+                                (b'totally awful', 0), (b'super good', 2),
+                                (b'really bad', 0)])
+    self.assertEqual(data_values, expected_data_values)
+
+  def test_split(self):
+    ds = tf.data.Dataset.from_tensor_slices(['good', 'bad', 'neutral', 'odd'])
+    data = dataset.Dataset(ds, 4, ['pos', 'neg'])
+    train_data, test_data = data.split(0.5)
+    expected_train_data = [b'good', b'bad']
+    expected_test_data = [b'neutral', b'odd']
+
+    self.assertLen(train_data, 2)
+    train_data_values = [elem.numpy() for elem in train_data._dataset]
+    self.assertEqual(train_data_values, expected_train_data)
+    self.assertEqual(train_data.num_classes, 2)
+    self.assertEqual(train_data.label_names, ['pos', 'neg'])
+
+    self.assertLen(test_data, 2)
+    test_data_values = [elem.numpy() for elem in test_data._dataset]
+    self.assertEqual(test_data_values, expected_test_data)
+    self.assertEqual(test_data.num_classes, 2)
+    self.assertEqual(test_data.label_names, ['pos', 'neg'])
+
+
+if __name__ == '__main__':
+  tf.test.main()

mediapipe/model_maker/python/text/text_classifier/model_options.py

@@ -0,0 +1,45 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Configurable model options for text classifier models."""
+
+import dataclasses
+from typing import Union
+
+from mediapipe.model_maker.python.text.core import bert_model_options
+
+# BERT text classifier options inherited from BertModelOptions.
+BertClassifierOptions = bert_model_options.BertModelOptions
+
+
+@dataclasses.dataclass
+class AverageWordEmbeddingClassifierOptions:
+  """Configurable model options for an Average Word Embedding classifier.
+
+  Attributes:
+    seq_len: Length of the sequence to feed into the model.
+    wordvec_dim: Dimension of the word embedding.
+    do_lower_case: Whether to convert all uppercase characters to lowercase
+      during preprocessing.
+    vocab_size: Number of words to generate the vocabulary from data.
+    dropout_rate: The rate for dropout.
+  """
+  seq_len: int = 256
+  wordvec_dim: int = 16
+  do_lower_case: bool = True
+  vocab_size: int = 10000
+  dropout_rate: float = 0.2
+
+
+TextClassifierModelOptions = Union[AverageWordEmbeddingClassifierOptions,
+                                   BertClassifierOptions]

mediapipe/model_maker/python/text/text_classifier/model_spec.py

@@ -0,0 +1,70 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Specifications for text classifier models."""
+
+import dataclasses
+import enum
+import functools
+
+from mediapipe.model_maker.python.core import hyperparameters as hp
+from mediapipe.model_maker.python.text.core import bert_model_spec
+from mediapipe.model_maker.python.text.text_classifier import model_options as mo
+
+# BERT-based text classifier spec inherited from BertModelSpec
+BertClassifierSpec = bert_model_spec.BertModelSpec
+
+
+@dataclasses.dataclass
+class AverageWordEmbeddingClassifierSpec:
+  """Specification for an average word embedding classifier model.
+
+  Attributes:
+    hparams: Configurable hyperparameters for training.
+    model_options: Configurable options for the average word embedding model.
+    name: The name of the object.
+  """
+
+  # `learning_rate` is unused for the average word embedding model
+  hparams: hp.BaseHParams = hp.BaseHParams(
+      epochs=10, batch_size=32, learning_rate=0)
+  model_options: mo.AverageWordEmbeddingClassifierOptions = (
+      mo.AverageWordEmbeddingClassifierOptions())
+  name: str = 'AverageWordEmbedding'
+
+
+average_word_embedding_classifier_spec = functools.partial(
+    AverageWordEmbeddingClassifierSpec)
+
+mobilebert_classifier_spec = functools.partial(
+    BertClassifierSpec,
+    hparams=hp.BaseHParams(
+        epochs=3,
+        batch_size=48,
+        learning_rate=3e-5,
+        distribution_strategy='off'),
+    name='MobileBert',
+    uri='https://tfhub.dev/tensorflow/mobilebert_en_uncased_L-24_H-128_B-512_A-4_F-4_OPT/1',
+    tflite_input_name={
+        'ids': 'serving_default_input_1:0',
+        'mask': 'serving_default_input_3:0',
+        'segment_ids': 'serving_default_input_2:0'
+    },
+)
+
+
+@enum.unique
+class SupportedModels(enum.Enum):
+  """Predefined text classifier model specs supported by Model Maker."""
+  AVERAGE_WORD_EMBEDDING_CLASSIFIER = average_word_embedding_classifier_spec
+  MOBILEBERT_CLASSIFIER = mobilebert_classifier_spec

mediapipe/model_maker/python/text/text_classifier/model_spec_test.py

@@ -0,0 +1,118 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Tests for model_spec."""
+
+import os
+
+import tensorflow as tf
+
+from mediapipe.model_maker.python.core import hyperparameters as hp
+from mediapipe.model_maker.python.text.text_classifier import model_options as classifier_model_options
+from mediapipe.model_maker.python.text.text_classifier import model_spec as ms
+
+
+class ModelSpecTest(tf.test.TestCase):
+
+  def test_predefined_bert_spec(self):
+    model_spec_obj = ms.SupportedModels.MOBILEBERT_CLASSIFIER.value()
+    self.assertIsInstance(model_spec_obj, ms.BertClassifierSpec)
+    self.assertEqual(model_spec_obj.name, 'MobileBert')
+    self.assertEqual(
+        model_spec_obj.uri, 'https://tfhub.dev/tensorflow/'
+        'mobilebert_en_uncased_L-24_H-128_B-512_A-4_F-4_OPT/1')
+    self.assertTrue(model_spec_obj.do_lower_case)
+    self.assertEqual(
+        model_spec_obj.tflite_input_name, {
+            'ids': 'serving_default_input_1:0',
+            'mask': 'serving_default_input_3:0',
+            'segment_ids': 'serving_default_input_2:0'
+        })
+    self.assertEqual(
+        model_spec_obj.model_options,
+        classifier_model_options.BertClassifierOptions(
+            seq_len=128, do_fine_tuning=True, dropout_rate=0.1))
+    self.assertEqual(
+        model_spec_obj.hparams,
+        hp.BaseHParams(
+            epochs=3,
+            batch_size=48,
+            learning_rate=3e-5,
+            distribution_strategy='off'))
+
+  def test_predefined_average_word_embedding_spec(self):
+    model_spec_obj = (
+        ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER.value())
+    self.assertIsInstance(model_spec_obj, ms.AverageWordEmbeddingClassifierSpec)
+    self.assertEqual(model_spec_obj.name, 'AverageWordEmbedding')
+    self.assertEqual(
+        model_spec_obj.model_options,
+        classifier_model_options.AverageWordEmbeddingClassifierOptions(
+            seq_len=256,
+            wordvec_dim=16,
+            do_lower_case=True,
+            vocab_size=10000,
+            dropout_rate=0.2))
+    self.assertEqual(
+        model_spec_obj.hparams,
+        hp.BaseHParams(
+            epochs=10,
+            batch_size=32,
+            learning_rate=0,
+            steps_per_epoch=None,
+            shuffle=False,
+            distribution_strategy='off',
+            num_gpus=-1,
+            tpu=''))
+
+  def test_custom_bert_spec(self):
+    custom_bert_classifier_options = (
+        classifier_model_options.BertClassifierOptions(
+            seq_len=512, do_fine_tuning=False, dropout_rate=0.3))
+    model_spec_obj = (
+        ms.SupportedModels.MOBILEBERT_CLASSIFIER.value(
+            model_options=custom_bert_classifier_options))
+    self.assertEqual(model_spec_obj.model_options,
+                     custom_bert_classifier_options)
+
+  def test_custom_average_word_embedding_spec(self):
+    custom_hparams = hp.BaseHParams(
+        learning_rate=0.4,
+        batch_size=64,
+        epochs=10,
+        steps_per_epoch=10,
+        shuffle=True,
+        export_dir='foo/bar',
+        distribution_strategy='mirrored',
+        num_gpus=3,
+        tpu='tpu/address')
+    custom_average_word_embedding_model_options = (
+        classifier_model_options.AverageWordEmbeddingClassifierOptions(
+            seq_len=512,
+            wordvec_dim=32,
+            do_lower_case=False,
+            vocab_size=5000,
+            dropout_rate=0.5))
+    model_spec_obj = (
+        ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER.value(
+            model_options=custom_average_word_embedding_model_options,
+            hparams=custom_hparams))
+    self.assertEqual(model_spec_obj.model_options,
+                     custom_average_word_embedding_model_options)
+    self.assertEqual(model_spec_obj.hparams, custom_hparams)
+
+
+if __name__ == '__main__':
+  # Load compressed models from tensorflow_hub
+  os.environ['TFHUB_MODEL_LOAD_FORMAT'] = 'COMPRESSED'
+  tf.test.main()

mediapipe/model_maker/python/text/text_classifier/preprocessor.py

@@ -0,0 +1,285 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+"""Preprocessors for text classification."""
+
+import collections
+import os
+import re
+import tempfile
+from typing import Mapping, Sequence, Tuple, Union
+
+import tensorflow as tf
+import tensorflow_hub
+
+from mediapipe.model_maker.python.text.text_classifier import dataset as text_classifier_ds
+from official.nlp.data import classifier_data_lib
+from official.nlp.tools import tokenization
+
+
+def _validate_text_and_label(text: tf.Tensor, label: tf.Tensor) -> None:
+  """Validates the shape and type of `text` and `label`.
+
+  Args:
+    text: Stores text data. Should have shape [1] and dtype tf.string.
+    label: Stores the label for the corresponding `text`. Should have shape [1]
+      and dtype tf.int64.
+
+  Raises:
+    ValueError: If either tensor has the wrong shape or type.
+  """
+  if text.shape != [1]:
+    raise ValueError(f"`text` should have shape [1], got {text.shape}")
+  if text.dtype != tf.string:
+    raise ValueError(f"Expected dtype string for `text`, got {text.dtype}")
+  if label.shape != [1]:
+    raise ValueError(f"`label` should have shape [1], got {text.shape}")
+  if label.dtype != tf.int64:
+    raise ValueError(f"Expected dtype int64 for `label`, got {label.dtype}")
+
+
+def _decode_record(
+    record: tf.Tensor, name_to_features: Mapping[str, tf.io.FixedLenFeature]
+) -> Tuple[Mapping[str, tf.Tensor], tf.Tensor]:
+  """Decodes a record into input for a BERT model.
+
+  Args:
+    record: Stores serialized example.
+    name_to_features: Maps record keys to feature types.
+
+  Returns:
+    BERT model input features and label for the record.
+  """
+  example = tf.io.parse_single_example(record, name_to_features)
+
+  # tf.Example only supports tf.int64, but the TPU only supports tf.int32.
+  for name in list(example.keys()):
+    example[name] = tf.cast(example[name], tf.int32)
+
+  bert_features = {
+      "input_word_ids": example["input_ids"],
+      "input_mask": example["input_mask"],
+      "input_type_ids": example["segment_ids"]
+  }
+  return bert_features, example["label_ids"]
+
+
+def _single_file_dataset(
+    input_file: str, name_to_features: Mapping[str, tf.io.FixedLenFeature]
+) -> tf.data.TFRecordDataset:
+  """Creates a single-file dataset to be passed for BERT custom training.
+
+  Args:
+    input_file: Filepath for the dataset.
+    name_to_features: Maps record keys to feature types.
+
+  Returns:
+    Dataset containing BERT model input features and labels.
+  """
+  d = tf.data.TFRecordDataset(input_file)
+  d = d.map(
+      lambda record: _decode_record(record, name_to_features),
+      num_parallel_calls=tf.data.AUTOTUNE)
+  return d
+
+
+class AverageWordEmbeddingClassifierPreprocessor:
+  """Preprocessor for an Average Word Embedding model.
+
+  Takes (text, label) data and applies regex tokenization and padding to the
+  text to generate (token IDs, label) data.
+
+  Attributes:
+    seq_len: Length of the input sequence to the model.
+    do_lower_case: Whether text inputs should be converted to lower-case.
+    vocab: Vocabulary of tokens used by the model.
+  """
+
+  PAD: str = "<PAD>"  # Index: 0
+  START: str = "<START>"  # Index: 1
+  UNKNOWN: str = "<UNKNOWN>"  # Index: 2
+
+  def __init__(self, seq_len: int, do_lower_case: bool, texts: Sequence[str],
+               vocab_size: int):
+    self._seq_len = seq_len
+    self._do_lower_case = do_lower_case
+    self._vocab = self._gen_vocab(texts, vocab_size)
+
+  def _gen_vocab(self, texts: Sequence[str],
+                 vocab_size: int) -> Mapping[str, int]:
+    """Generates vocabulary list in `texts` with size `vocab_size`.
+
+    Args:
+      texts: All texts (across training and validation data) that will be
+        preprocessed by the model.
+      vocab_size: Size of the vocab.
+
+    Returns:
+      The vocab mapping tokens to IDs.
+    """
+    vocab_counter = collections.Counter()
+
+    for text in texts:
+      tokens = self._regex_tokenize(text)
+      for token in tokens:
+        vocab_counter[token] += 1
+
+    vocab_freq = vocab_counter.most_common(vocab_size)
+    vocab_list = [self.PAD, self.START, self.UNKNOWN
+                 ] + [word for word, _ in vocab_freq]
+    return collections.OrderedDict(((v, i) for i, v in enumerate(vocab_list)))
+
+  def get_vocab(self) -> Mapping[str, int]:
+    """Returns the vocab of the AverageWordEmbeddingClassifierPreprocessor."""
+    return self._vocab
+
+  # TODO: Align with MediaPipe's RegexTokenizer.
+  def _regex_tokenize(self, text: str) -> Sequence[str]:
+    """Splits `text` by words but does not split on single quotes.
+
+    Args:
+      text: Text to be tokenized.
+
+    Returns:
+      List of tokens.
+    """
+    text = tf.compat.as_text(text)
+    if self._do_lower_case:
+      text = text.lower()
+    tokens = re.compile(r"[^\w\']+").split(text.strip())
+    # Filters out any empty strings in `tokens`.
+    return list(filter(None, tokens))
+
+  def _tokenize_and_pad(self, text: str) -> Sequence[int]:
+    """Tokenizes `text` and pads the tokens to `seq_len`.
+
+    Args:
+      text: Text to be tokenized and padded.
+
+    Returns:
+      List of token IDs padded to have length `seq_len`.
+    """
+    tokens = self._regex_tokenize(text)
+
+    # Gets ids for START, PAD and UNKNOWN tokens.
+    start_id = self._vocab[self.START]
+    pad_id = self._vocab[self.PAD]
+    unknown_id = self._vocab[self.UNKNOWN]
+
+    token_ids = [self._vocab.get(token, unknown_id) for token in tokens]
+    token_ids = [start_id] + token_ids
+
+    if len(token_ids) < self._seq_len:
+      pad_length = self._seq_len - len(token_ids)
+      token_ids = token_ids + pad_length * [pad_id]
+    else:
+      token_ids = token_ids[:self._seq_len]
+    return token_ids
+
+  def preprocess(
+      self, dataset: text_classifier_ds.Dataset) -> text_classifier_ds.Dataset:
+    """Preprocesses data into input for an Average Word Embedding model.
+
+    Args:
+      dataset: Stores (text, label) data.
+
+    Returns:
+      Dataset containing (token IDs, label) data.
+    """
+    token_ids_list = []
+    labels_list = []
+    for text, label in dataset.gen_tf_dataset():
+      _validate_text_and_label(text, label)
+      token_ids = self._tokenize_and_pad(text.numpy()[0].decode("utf-8"))
+      token_ids_list.append(token_ids)
+      labels_list.append(label.numpy()[0])
+
+    token_ids_ds = tf.data.Dataset.from_tensor_slices(token_ids_list)
+    labels_ds = tf.data.Dataset.from_tensor_slices(labels_list)
+    preprocessed_ds = tf.data.Dataset.zip((token_ids_ds, labels_ds))
+    return text_classifier_ds.Dataset(
+        dataset=preprocessed_ds,
+        size=dataset.size,
+        label_names=dataset.label_names)
+
+
+class BertClassifierPreprocessor:
+  """Preprocessor for a BERT-based classifier.
+
+  Attributes:
+    seq_len: Length of the input sequence to the model.
+    vocab_file: File containing the BERT vocab.
+    tokenizer: BERT tokenizer.
+  """
+
+  def __init__(self, seq_len: int, do_lower_case: bool, uri: str):
+    self._seq_len = seq_len
+    # Vocab filepath is tied to the BERT module's URI.
+    self._vocab_file = os.path.join(
+        tensorflow_hub.resolve(uri), "assets", "vocab.txt")
+    self._tokenizer = tokenization.FullTokenizer(self._vocab_file,
+                                                 do_lower_case)
+
+  def _get_name_to_features(self):
+    """Gets the dictionary mapping record keys to feature types."""
+    return {
+        "input_ids": tf.io.FixedLenFeature([self._seq_len], tf.int64),
+        "input_mask": tf.io.FixedLenFeature([self._seq_len], tf.int64),
+        "segment_ids": tf.io.FixedLenFeature([self._seq_len], tf.int64),
+        "label_ids": tf.io.FixedLenFeature([], tf.int64),
+    }
+
+  def get_vocab_file(self) -> str:
+    """Returns the vocab file of the BertClassifierPreprocessor."""
+    return self._vocab_file
+
+  def preprocess(
+      self, dataset: text_classifier_ds.Dataset) -> text_classifier_ds.Dataset:
+    """Preprocesses data into input for a BERT-based classifier.
+
+    Args:
+      dataset: Stores (text, label) data.
+
+    Returns:
+      Dataset containing (bert_features, label) data.
+    """
+    examples = []
+    for index, (text, label) in enumerate(dataset.gen_tf_dataset()):
+      _validate_text_and_label(text, label)
+      examples.append(
+          classifier_data_lib.InputExample(
+              guid=str(index),
+              text_a=text.numpy()[0].decode("utf-8"),
+              text_b=None,
+              # InputExample expects the label name rather than the int ID
+              label=dataset.label_names[label.numpy()[0]]))
+
+    tfrecord_file = os.path.join(tempfile.mkdtemp(), "bert_features.tfrecord")
+    classifier_data_lib.file_based_convert_examples_to_features(
+        examples=examples,
+        label_list=dataset.label_names,
+        max_seq_length=self._seq_len,
+        tokenizer=self._tokenizer,
+        output_file=tfrecord_file)
+    preprocessed_ds = _single_file_dataset(tfrecord_file,
+                                           self._get_name_to_features())
+    return text_classifier_ds.Dataset(
+        dataset=preprocessed_ds,
+        size=dataset.size,
+        label_names=dataset.label_names)
+
+
+TextClassifierPreprocessor = (
+    Union[BertClassifierPreprocessor,
+          AverageWordEmbeddingClassifierPreprocessor])

mediapipe/model_maker/python/text/text_classifier/preprocessor_test.py

@@ -0,0 +1,96 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import csv
+import os
+
+import numpy as np
+import numpy.testing as npt
+import tensorflow as tf
+
+from mediapipe.model_maker.python.text.text_classifier import dataset as text_classifier_ds
+from mediapipe.model_maker.python.text.text_classifier import model_spec
+from mediapipe.model_maker.python.text.text_classifier import preprocessor
+
+
+class PreprocessorTest(tf.test.TestCase):
+  CSV_PARAMS_ = text_classifier_ds.CSVParameters(
+      text_column='text', label_column='label')
+
+  def _get_csv_file(self):
+    labels_and_text = (('pos', 'super super super super good'),
+                       (('neg', 'really bad')))
+    csv_file = os.path.join(self.get_temp_dir(), 'data.csv')
+    if os.path.exists(csv_file):
+      return csv_file
+    fieldnames = ['text', 'label']
+    with open(csv_file, 'w') as f:
+      writer = csv.DictWriter(f, fieldnames=fieldnames)
+      writer.writeheader()
+      for label, text in labels_and_text:
+        writer.writerow({'text': text, 'label': label})
+    return csv_file
+
+  def test_average_word_embedding_preprocessor(self):
+    csv_file = self._get_csv_file()
+    dataset = text_classifier_ds.Dataset.from_csv(
+        filename=csv_file, csv_params=self.CSV_PARAMS_)
+    average_word_embedding_preprocessor = (
+        preprocessor.AverageWordEmbeddingClassifierPreprocessor(
+            seq_len=5,
+            do_lower_case=True,
+            texts=['super super super super good', 'really bad'],
+            vocab_size=7))
+    preprocessed_dataset = (
+        average_word_embedding_preprocessor.preprocess(dataset))
+    labels = []
+    features_list = []
+    for features, label in preprocessed_dataset.gen_tf_dataset():
+      self.assertEqual(label.shape, [1])
+      labels.append(label.numpy()[0])
+      self.assertEqual(features.shape, [1, 5])
+      features_list.append(features.numpy()[0])
+    self.assertEqual(labels, [1, 0])
+    npt.assert_array_equal(
+        np.stack(features_list), np.array([[1, 3, 3, 3, 3], [1, 5, 6, 0, 0]]))
+
+  def test_bert_preprocessor(self):
+    csv_file = self._get_csv_file()
+    dataset = text_classifier_ds.Dataset.from_csv(
+        filename=csv_file, csv_params=self.CSV_PARAMS_)
+    bert_spec = model_spec.SupportedModels.MOBILEBERT_CLASSIFIER.value()
+    bert_preprocessor = preprocessor.BertClassifierPreprocessor(
+        seq_len=5, do_lower_case=bert_spec.do_lower_case, uri=bert_spec.uri)
+    preprocessed_dataset = bert_preprocessor.preprocess(dataset)
+    labels = []
+    input_masks = []
+    for features, label in preprocessed_dataset.gen_tf_dataset():
+      self.assertEqual(label.shape, [1])
+      labels.append(label.numpy()[0])
+      self.assertSameElements(
+          features.keys(), ['input_word_ids', 'input_mask', 'input_type_ids'])
+      for feature in features.values():
+        self.assertEqual(feature.shape, [1, 5])
+      input_masks.append(features['input_mask'].numpy()[0])
+      npt.assert_array_equal(features['input_type_ids'].numpy()[0],
+                             [0, 0, 0, 0, 0])
+    npt.assert_array_equal(
+        np.stack(input_masks), np.array([[1, 1, 1, 1, 1], [1, 1, 1, 1, 0]]))
+    self.assertEqual(labels, [1, 0])
+
+
+if __name__ == '__main__':
+  # Load compressed models from tensorflow_hub
+  os.environ['TFHUB_MODEL_LOAD_FORMAT'] = 'COMPRESSED'
+  tf.test.main()

mediapipe/model_maker/python/text/text_classifier/testdata/BUILD

@@ -0,0 +1,23 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+package(
+    default_visibility = ["//mediapipe/model_maker/python/text/text_classifier:__subpackages__"],
+    licenses = ["notice"],  # Apache 2.0
+)
+
+filegroup(
+    name = "testdata",
+    srcs = ["average_word_embedding_metadata.json"],
+)

mediapipe/model_maker/python/text/text_classifier/testdata/average_word_embedding_metadata.json

@@ -0,0 +1,63 @@
+{
+  "name": "TextClassifier",
+  "description": "Classify the input text into a set of known categories.",
+  "subgraph_metadata": [
+    {
+      "input_tensor_metadata": [
+        {
+          "name": "input_text",
+          "description": "Embedding vectors representing the input text to be processed.",
+          "content": {
+            "content_properties_type": "FeatureProperties",
+            "content_properties": {
+            }
+          },
+          "process_units": [
+            {
+              "options_type": "RegexTokenizerOptions",
+              "options": {
+                "delim_regex_pattern": "[^\\w\\']+",
+                "vocab_file": [
+                  {
+                    "name": "vocab.txt",
+                    "description": "Vocabulary file to convert natural language words to embedding vectors.",
+                    "type": "VOCABULARY"
+                  }
+                ]
+              }
+            }
+          ],
+          "stats": {
+          }
+        }
+      ],
+      "output_tensor_metadata": [
+        {
+          "name": "score",
+          "description": "Score of the labels respectively.",
+          "content": {
+            "content_properties_type": "FeatureProperties",
+            "content_properties": {
+            }
+          },
+          "stats": {
+            "max": [
+              1.0
+            ],
+            "min": [
+              0.0
+            ]
+          },
+          "associated_files": [
+            {
+              "name": "labels.txt",
+              "description": "Labels for categories that the model can recognize.",
+              "type": "TENSOR_AXIS_LABELS"
+            }
+          ]
+        }
+      ]
+    }
+  ],
+  "min_parser_version": "1.2.1"
+}

mediapipe/model_maker/python/text/text_classifier/text_classifier.py

@@ -0,0 +1,437 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""API for text classification."""
+
+import abc
+import os
+import tempfile
+from typing import Any, Optional, Sequence, Tuple
+
+import tensorflow as tf
+import tensorflow_hub as hub
+
+from mediapipe.model_maker.python.core import hyperparameters as hp
+from mediapipe.model_maker.python.core.data import dataset as ds
+from mediapipe.model_maker.python.core.tasks import classifier
+from mediapipe.model_maker.python.core.utils import model_util
+from mediapipe.model_maker.python.core.utils import quantization
+from mediapipe.model_maker.python.text.text_classifier import dataset as text_ds
+from mediapipe.model_maker.python.text.text_classifier import model_options as mo
+from mediapipe.model_maker.python.text.text_classifier import model_spec as ms
+from mediapipe.model_maker.python.text.text_classifier import preprocessor
+from mediapipe.model_maker.python.text.text_classifier import text_classifier_options
+from mediapipe.tasks.python.metadata.metadata_writers import metadata_writer
+from mediapipe.tasks.python.metadata.metadata_writers import text_classifier as text_classifier_writer
+from official.nlp import optimization
+
+
+def _validate(options: text_classifier_options.TextClassifierOptions):
+  """Validates that `model_options` and `supported_model` are compatible.
+
+  Args:
+    options: Options for creating and training a text classifier.
+
+  Raises:
+    ValueError if there is a mismatch between `model_options` and
+    `supported_model`.
+  """
+
+  if options.model_options is None:
+    return
+
+  if (isinstance(options.model_options,
+                 mo.AverageWordEmbeddingClassifierOptions) and
+      (options.supported_model !=
+       ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER)):
+    raise ValueError("Expected AVERAGE_WORD_EMBEDDING_CLASSIFIER,"
+                     f" got {options.supported_model}")
+  if (isinstance(options.model_options, mo.BertClassifierOptions) and
+      (options.supported_model != ms.SupportedModels.MOBILEBERT_CLASSIFIER)):
+    raise ValueError(
+        f"Expected MOBILEBERT_CLASSIFIER, got {options.supported_model}")
+
+
+class TextClassifier(classifier.Classifier):
+  """API for creating and training a text classification model."""
+
+  def __init__(self, model_spec: Any, hparams: hp.BaseHParams,
+               label_names: Sequence[str]):
+    super().__init__(
+        model_spec=model_spec, label_names=label_names, shuffle=hparams.shuffle)
+    self._model_spec = model_spec
+    self._hparams = hparams
+    self._callbacks = model_util.get_default_callbacks(self._hparams.export_dir)
+    self._text_preprocessor: preprocessor.TextClassifierPreprocessor = None
+
+  @classmethod
+  def create(
+      cls, train_data: text_ds.Dataset, validation_data: text_ds.Dataset,
+      options: text_classifier_options.TextClassifierOptions
+  ) -> "TextClassifier":
+    """Factory function that creates and trains a text classifier.
+
+    Note that `train_data` and `validation_data` are expected to share the same
+    `label_names` since they should be split from the same dataset.
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data.
+      options: Options for creating and training the text classifier.
+
+    Returns:
+      A text classifier.
+
+    Raises:
+      ValueError if `train_data` and `validation_data` do not have the
+      same label_names or `options` contains an unknown `supported_model`
+    """
+    if train_data.label_names != validation_data.label_names:
+      raise ValueError(
+          f"Training data label names {train_data.label_names} not equal to "
+          f"validation data label names {validation_data.label_names}")
+
+    _validate(options)
+    if options.model_options is None:
+      options.model_options = options.supported_model.value().model_options
+
+    if options.hparams is None:
+      options.hparams = options.supported_model.value().hparams
+
+    if options.supported_model == ms.SupportedModels.MOBILEBERT_CLASSIFIER:
+      text_classifier = (
+          _BertClassifier.create_bert_classifier(train_data, validation_data,
+                                                 options,
+                                                 train_data.label_names))
+    elif (options.supported_model ==
+          ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER):
+      text_classifier = (
+          _AverageWordEmbeddingClassifier
+          .create_average_word_embedding_classifier(train_data, validation_data,
+                                                    options,
+                                                    train_data.label_names))
+    else:
+      raise ValueError(f"Unknown model {options.supported_model}")
+
+    return text_classifier
+
+  def evaluate(self, data: ds.Dataset, batch_size: int = 32) -> Any:
+    """Overrides Classifier.evaluate().
+
+    Args:
+      data: Evaluation dataset. Must be a TextClassifier Dataset.
+      batch_size: Number of samples per evaluation step.
+
+    Returns:
+      The loss value and accuracy.
+
+    Raises:
+      ValueError if `data` is not a TextClassifier Dataset.
+    """
+    # This override is needed because TextClassifier preprocesses its data
+    # outside of the `gen_tf_dataset()` method. The preprocess call also
+    # requires a TextClassifier Dataset instead of a core Dataset.
+    if not isinstance(data, text_ds.Dataset):
+      raise ValueError("Need a TextClassifier Dataset.")
+
+    processed_data = self._text_preprocessor.preprocess(data)
+    dataset = processed_data.gen_tf_dataset(batch_size, is_training=False)
+    return self._model.evaluate(dataset)
+
+  def export_model(
+      self,
+      model_name: str = "model.tflite",
+      quantization_config: Optional[quantization.QuantizationConfig] = None):
+    """Converts and saves the model to a TFLite file with metadata included.
+
+    Note that only the TFLite file is needed for deployment. This function also
+    saves a metadata.json file to the same directory as the TFLite file which
+    can be used to interpret the metadata content in the TFLite file.
+
+    Args:
+      model_name: File name to save TFLite model with metadata. The full export
+        path is {self._hparams.export_dir}/{model_name}.
+      quantization_config: The configuration for model quantization.
+    """
+    if not tf.io.gfile.exists(self._hparams.export_dir):
+      tf.io.gfile.makedirs(self._hparams.export_dir)
+    tflite_file = os.path.join(self._hparams.export_dir, model_name)
+    metadata_file = os.path.join(self._hparams.export_dir, "metadata.json")
+
+    tflite_model = model_util.convert_to_tflite(
+        model=self._model, quantization_config=quantization_config)
+    vocab_filepath = os.path.join(tempfile.mkdtemp(), "vocab.txt")
+    self._save_vocab(vocab_filepath)
+
+    writer = self._get_metadata_writer(tflite_model, vocab_filepath)
+    tflite_model_with_metadata, metadata_json = writer.populate()
+    model_util.save_tflite(tflite_model_with_metadata, tflite_file)
+    with open(metadata_file, "w") as f:
+      f.write(metadata_json)
+
+  @abc.abstractmethod
+  def _save_vocab(self, vocab_filepath: str):
+    """Saves the preprocessor's vocab to `vocab_filepath`."""
+
+  @abc.abstractmethod
+  def _get_metadata_writer(self, tflite_model: bytearray, vocab_filepath: str):
+    """Gets the metadata writer for the text classifier TFLite model."""
+
+
+class _AverageWordEmbeddingClassifier(TextClassifier):
+  """APIs to help create and train an Average Word Embedding text classifier."""
+
+  _DELIM_REGEX_PATTERN = r"[^\w\']+"
+
+  def __init__(self, model_spec: ms.AverageWordEmbeddingClassifierSpec,
+               model_options: mo.AverageWordEmbeddingClassifierOptions,
+               hparams: hp.BaseHParams, label_names: Sequence[str]):
+    super().__init__(model_spec, hparams, label_names)
+    self._model_options = model_options
+    self._loss_function = "sparse_categorical_crossentropy"
+    self._metric_function = "accuracy"
+    self._text_preprocessor: (
+        preprocessor.AverageWordEmbeddingClassifierPreprocessor) = None
+
+  @classmethod
+  def create_average_word_embedding_classifier(
+      cls, train_data: text_ds.Dataset, validation_data: text_ds.Dataset,
+      options: text_classifier_options.TextClassifierOptions,
+      label_names: Sequence[str]) -> "_AverageWordEmbeddingClassifier":
+    """Creates, trains, and returns an Average Word Embedding classifier.
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data.
+      options: Options for creating and training the text classifier.
+      label_names: Label names used in the data.
+
+    Returns:
+      An Average Word Embedding classifier.
+    """
+    average_word_embedding_classifier = _AverageWordEmbeddingClassifier(
+        model_spec=options.supported_model.value(),
+        model_options=options.model_options,
+        hparams=options.hparams,
+        label_names=train_data.label_names)
+    average_word_embedding_classifier._create_and_train_model(
+        train_data, validation_data)
+    return average_word_embedding_classifier
+
+  def _create_and_train_model(self, train_data: text_ds.Dataset,
+                              validation_data: text_ds.Dataset):
+    """Creates the Average Word Embedding classifier keras model and trains it.
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data.
+    """
+    (processed_train_data, processed_validation_data) = (
+        self._load_and_run_preprocessor(train_data, validation_data))
+    self._create_model()
+    self._optimizer = "rmsprop"
+    self._train_model(processed_train_data, processed_validation_data)
+
+  def _load_and_run_preprocessor(
+      self, train_data: text_ds.Dataset, validation_data: text_ds.Dataset
+  ) -> Tuple[text_ds.Dataset, text_ds.Dataset]:
+    """Runs an AverageWordEmbeddingClassifierPreprocessor on the data.
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data.
+
+    Returns:
+      Preprocessed training data and preprocessed validation data.
+    """
+    train_texts = [text.numpy()[0] for text, _ in train_data.gen_tf_dataset()]
+    validation_texts = [
+        text.numpy()[0] for text, _ in validation_data.gen_tf_dataset()
+    ]
+    self._text_preprocessor = (
+        preprocessor.AverageWordEmbeddingClassifierPreprocessor(
+            seq_len=self._model_options.seq_len,
+            do_lower_case=self._model_options.do_lower_case,
+            texts=train_texts + validation_texts,
+            vocab_size=self._model_options.vocab_size))
+    return self._text_preprocessor.preprocess(
+        train_data), self._text_preprocessor.preprocess(validation_data)
+
+  def _create_model(self):
+    """Creates an Average Word Embedding model."""
+    self._model = tf.keras.Sequential([
+        tf.keras.layers.InputLayer(
+            input_shape=[self._model_options.seq_len], dtype=tf.int32),
+        tf.keras.layers.Embedding(
+            len(self._text_preprocessor.get_vocab()),
+            self._model_options.wordvec_dim,
+            input_length=self._model_options.seq_len),
+        tf.keras.layers.GlobalAveragePooling1D(),
+        tf.keras.layers.Dense(
+            self._model_options.wordvec_dim, activation=tf.nn.relu),
+        tf.keras.layers.Dropout(self._model_options.dropout_rate),
+        tf.keras.layers.Dense(self._num_classes, activation="softmax")
+    ])
+
+  def _save_vocab(self, vocab_filepath: str):
+    with tf.io.gfile.GFile(vocab_filepath, "w") as f:
+      for token, index in self._text_preprocessor.get_vocab().items():
+        f.write(f"{token} {index}\n")
+
+  def _get_metadata_writer(self, tflite_model: bytearray, vocab_filepath: str):
+    return text_classifier_writer.MetadataWriter.create_for_regex_model(
+        model_buffer=tflite_model,
+        regex_tokenizer=metadata_writer.RegexTokenizer(
+            # TODO: Align with MediaPipe's RegexTokenizer.
+            delim_regex_pattern=self._DELIM_REGEX_PATTERN,
+            vocab_file_path=vocab_filepath),
+        labels=metadata_writer.Labels().add(list(self._label_names)))
+
+
+class _BertClassifier(TextClassifier):
+  """APIs to help create and train a BERT-based text classifier."""
+
+  _INITIALIZER_RANGE = 0.02
+
+  def __init__(self, model_spec: ms.BertClassifierSpec,
+               model_options: mo.BertClassifierOptions, hparams: hp.BaseHParams,
+               label_names: Sequence[str]):
+    super().__init__(model_spec, hparams, label_names)
+    self._model_options = model_options
+    self._loss_function = tf.keras.losses.SparseCategoricalCrossentropy()
+    self._metric_function = tf.keras.metrics.SparseCategoricalAccuracy(
+        "test_accuracy", dtype=tf.float32)
+    self._text_preprocessor: preprocessor.BertClassifierPreprocessor = None
+
+  @classmethod
+  def create_bert_classifier(
+      cls, train_data: text_ds.Dataset, validation_data: text_ds.Dataset,
+      options: text_classifier_options.TextClassifierOptions,
+      label_names: Sequence[str]) -> "_BertClassifier":
+    """Creates, trains, and returns a BERT-based classifier.
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data.
+      options: Options for creating and training the text classifier.
+      label_names: Label names used in the data.
+
+    Returns:
+      A BERT-based classifier.
+    """
+    bert_classifier = _BertClassifier(
+        model_spec=options.supported_model.value(),
+        model_options=options.model_options,
+        hparams=options.hparams,
+        label_names=train_data.label_names)
+    bert_classifier._create_and_train_model(train_data, validation_data)
+    return bert_classifier
+
+  def _create_and_train_model(self, train_data: text_ds.Dataset,
+                              validation_data: text_ds.Dataset):
+    """Creates the BERT-based classifier keras model and trains it.
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data.
+    """
+    (processed_train_data, processed_validation_data) = (
+        self._load_and_run_preprocessor(train_data, validation_data))
+    self._create_model()
+    self._create_optimizer(processed_train_data)
+    self._train_model(processed_train_data, processed_validation_data)
+
+  def _load_and_run_preprocessor(
+      self, train_data: text_ds.Dataset, validation_data: text_ds.Dataset
+  ) -> Tuple[text_ds.Dataset, text_ds.Dataset]:
+    """Loads a BertClassifierPreprocessor and runs it on the data.
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data.
+
+    Returns:
+      Preprocessed training data and preprocessed validation data.
+    """
+    self._text_preprocessor = preprocessor.BertClassifierPreprocessor(
+        seq_len=self._model_options.seq_len,
+        do_lower_case=self._model_spec.do_lower_case,
+        uri=self._model_spec.uri)
+    return (self._text_preprocessor.preprocess(train_data),
+            self._text_preprocessor.preprocess(validation_data))
+
+  def _create_model(self):
+    """Creates a BERT-based classifier model.
+
+    The model architecture consists of stacking a dense classification layer and
+    dropout layer on top of the BERT encoder outputs.
+    """
+    encoder_inputs = dict(
+        input_word_ids=tf.keras.layers.Input(
+            shape=(self._model_options.seq_len,), dtype=tf.int32),
+        input_mask=tf.keras.layers.Input(
+            shape=(self._model_options.seq_len,), dtype=tf.int32),
+        input_type_ids=tf.keras.layers.Input(
+            shape=(self._model_options.seq_len,), dtype=tf.int32),
+    )
+    encoder = hub.KerasLayer(
+        self._model_spec.uri, trainable=self._model_options.do_fine_tuning)
+    encoder_outputs = encoder(encoder_inputs)
+    pooled_output = encoder_outputs["pooled_output"]
+
+    output = tf.keras.layers.Dropout(rate=self._model_options.dropout_rate)(
+        pooled_output)
+    initializer = tf.keras.initializers.TruncatedNormal(
+        stddev=self._INITIALIZER_RANGE)
+    output = tf.keras.layers.Dense(
+        self._num_classes,
+        kernel_initializer=initializer,
+        name="output",
+        activation="softmax",
+        dtype=tf.float32)(
+            output)
+    self._model = tf.keras.Model(inputs=encoder_inputs, outputs=output)
+
+  def _create_optimizer(self, train_data: text_ds.Dataset):
+    """Loads an optimizer with a learning rate schedule.
+
+    The decay steps in the learning rate schedule depend on the
+    `steps_per_epoch` which may depend on the size of the training data.
+
+    Args:
+      train_data: Training data.
+    """
+    self._hparams.steps_per_epoch = model_util.get_steps_per_epoch(
+        steps_per_epoch=self._hparams.steps_per_epoch,
+        batch_size=self._hparams.batch_size,
+        train_data=train_data)
+    total_steps = self._hparams.steps_per_epoch * self._hparams.epochs
+    warmup_steps = int(total_steps * 0.1)
+    initial_lr = self._hparams.learning_rate
+    self._optimizer = optimization.create_optimizer(initial_lr, total_steps,
+                                                    warmup_steps)
+
+  def _save_vocab(self, vocab_filepath: str):
+    tf.io.gfile.copy(
+        self._text_preprocessor.get_vocab_file(),
+        vocab_filepath,
+        overwrite=True)
+
+  def _get_metadata_writer(self, tflite_model: bytearray, vocab_filepath: str):
+    return text_classifier_writer.MetadataWriter.create_for_bert_model(
+        model_buffer=tflite_model,
+        tokenizer=metadata_writer.BertTokenizer(vocab_filepath),
+        labels=metadata_writer.Labels().add(list(self._label_names)),
+        ids_name=self._model_spec.tflite_input_name["ids"],
+        mask_name=self._model_spec.tflite_input_name["mask"],
+        segment_name=self._model_spec.tflite_input_name["segment_ids"])

mediapipe/model_maker/python/text/text_classifier/text_classifier_demo.py

@@ -0,0 +1,108 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Demo for making a text classifier model by MediaPipe Model Maker."""
+
+import os
+import tempfile
+
+# Dependency imports
+
+from absl import app
+from absl import flags
+from absl import logging
+import tensorflow as tf
+
+from mediapipe.model_maker.python.core import hyperparameters as hp
+from mediapipe.model_maker.python.core.utils import quantization
+from mediapipe.model_maker.python.text.text_classifier import dataset as text_ds
+from mediapipe.model_maker.python.text.text_classifier import model_spec as ms
+from mediapipe.model_maker.python.text.text_classifier import text_classifier
+from mediapipe.model_maker.python.text.text_classifier import text_classifier_options
+
+FLAGS = flags.FLAGS
+
+
+def define_flags():
+  flags.DEFINE_string('export_dir', None,
+                      'The directory to save exported files.')
+  flags.DEFINE_enum('supported_model', 'average_word_embedding',
+                    ['average_word_embedding', 'bert'],
+                    'The text classifier to run.')
+  flags.mark_flag_as_required('export_dir')
+
+
+def download_demo_data():
+  """Downloads demo data, and returns directory path."""
+  data_path = tf.keras.utils.get_file(
+      fname='SST-2.zip',
+      origin='https://dl.fbaipublicfiles.com/glue/data/SST-2.zip',
+      extract=True)
+  return os.path.join(os.path.dirname(data_path), 'SST-2')  # folder name
+
+
+def run(data_dir,
+        export_dir=tempfile.mkdtemp(),
+        supported_model=ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER):
+  """Runs demo."""
+
+  # Gets training data and validation data.
+  csv_params = text_ds.CSVParameters(
+      text_column='sentence', label_column='label', delimiter='\t')
+  train_data = text_ds.Dataset.from_csv(
+      filename=os.path.join(os.path.join(data_dir, 'train.tsv')),
+      csv_params=csv_params)
+  validation_data = text_ds.Dataset.from_csv(
+      filename=os.path.join(os.path.join(data_dir, 'dev.tsv')),
+      csv_params=csv_params)
+
+  quantization_config = None
+  if supported_model == ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER:
+    hparams = hp.BaseHParams(
+        epochs=10, batch_size=32, learning_rate=0, export_dir=export_dir)
+  # Warning: This takes extremely long to run on CPU
+  elif supported_model == ms.SupportedModels.MOBILEBERT_CLASSIFIER:
+    quantization_config = quantization.QuantizationConfig.for_dynamic()
+    hparams = hp.BaseHParams(
+        epochs=3, batch_size=48, learning_rate=3e-5, export_dir=export_dir)
+
+  # Fine-tunes the model.
+  options = text_classifier_options.TextClassifierOptions(
+      supported_model=supported_model, hparams=hparams)
+  model = text_classifier.TextClassifier.create(train_data, validation_data,
+                                                options)
+
+  # Gets evaluation results.
+  _, acc = model.evaluate(validation_data)
+  print('Eval accuracy: %f' % acc)
+
+  model.export_model(quantization_config=quantization_config)
+  model.export_labels(export_dir=options.hparams.export_dir)
+
+
+def main(_):
+  logging.set_verbosity(logging.INFO)
+  data_dir = download_demo_data()
+  export_dir = os.path.expanduser(FLAGS.export_dir)
+
+  if FLAGS.supported_model == 'average_word_embedding':
+    supported_model = ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER
+  elif FLAGS.supported_model == 'bert':
+    supported_model = ms.SupportedModels.MOBILEBERT_CLASSIFIER
+
+  run(data_dir, export_dir, supported_model)
+
+
+if __name__ == '__main__':
+  define_flags()
+  app.run(main)

mediapipe/model_maker/python/text/text_classifier/text_classifier_options.py

@@ -0,0 +1,38 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""User-facing customization options to create and train a text classifier."""
+
+import dataclasses
+from typing import Optional
+
+from mediapipe.model_maker.python.core import hyperparameters as hp
+from mediapipe.model_maker.python.text.text_classifier import model_options as mo
+from mediapipe.model_maker.python.text.text_classifier import model_spec as ms
+
+
+@dataclasses.dataclass
+class TextClassifierOptions:
+  """User-facing options for creating the text classifier.
+
+  Attributes:
+    supported_model: A preconfigured model spec.
+    hparams: Training hyperparameters the user can set to override the ones in
+      `supported_model`.
+    model_options: Model options the user can set to override the ones in
+      `supported_model`. The model options type should be consistent with the
+      architecture of the `supported_model`.
+  """
+  supported_model: ms.SupportedModels
+  hparams: Optional[hp.BaseHParams] = None
+  model_options: Optional[mo.TextClassifierModelOptions] = None

mediapipe/model_maker/python/text/text_classifier/text_classifier_test.py

@@ -0,0 +1,138 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import csv
+import filecmp
+import os
+
+import tensorflow as tf
+
+from mediapipe.model_maker.python.core import hyperparameters as hp
+from mediapipe.model_maker.python.text.text_classifier import dataset
+from mediapipe.model_maker.python.text.text_classifier import model_options as mo
+from mediapipe.model_maker.python.text.text_classifier import model_spec as ms
+from mediapipe.model_maker.python.text.text_classifier import text_classifier
+from mediapipe.model_maker.python.text.text_classifier import text_classifier_options
+from mediapipe.tasks.python.test import test_utils
+
+
+class TextClassifierTest(tf.test.TestCase):
+
+  _AVERAGE_WORD_EMBEDDING_JSON_FILE = (
+      test_utils.get_test_data_path('average_word_embedding_metadata.json'))
+
+  def _get_data(self):
+    labels_and_text = (('pos', 'super good'), (('neg', 'really bad')))
+    csv_file = os.path.join(self.get_temp_dir(), 'data.csv')
+    if os.path.exists(csv_file):
+      return csv_file
+    fieldnames = ['text', 'label']
+    with open(csv_file, 'w') as f:
+      writer = csv.DictWriter(f, fieldnames=fieldnames)
+      writer.writeheader()
+      for label, text in labels_and_text:
+        writer.writerow({'text': text, 'label': label})
+    csv_params = dataset.CSVParameters(text_column='text', label_column='label')
+    all_data = dataset.Dataset.from_csv(
+        filename=csv_file, csv_params=csv_params)
+    return all_data.split(0.5)
+
+  def test_create_and_train_average_word_embedding_model(self):
+    train_data, validation_data = self._get_data()
+    options = text_classifier_options.TextClassifierOptions(
+        supported_model=ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER,
+        hparams=hp.BaseHParams(epochs=1, batch_size=1, learning_rate=0))
+    average_word_embedding_classifier = text_classifier.TextClassifier.create(
+        train_data, validation_data, options)
+
+    _, accuracy = average_word_embedding_classifier.evaluate(validation_data)
+    self.assertGreaterEqual(accuracy, 0.0)
+
+    # Test export_model
+    average_word_embedding_classifier.export_model()
+    output_metadata_file = os.path.join(options.hparams.export_dir,
+                                        'metadata.json')
+    output_tflite_file = os.path.join(options.hparams.export_dir,
+                                      'model.tflite')
+
+    self.assertTrue(os.path.exists(output_tflite_file))
+    self.assertGreater(os.path.getsize(output_tflite_file), 0)
+
+    self.assertTrue(os.path.exists(output_metadata_file))
+    self.assertGreater(os.path.getsize(output_metadata_file), 0)
+    self.assertTrue(
+        filecmp.cmp(output_metadata_file,
+                    self._AVERAGE_WORD_EMBEDDING_JSON_FILE))
+
+  def test_create_and_train_bert(self):
+    train_data, validation_data = self._get_data()
+    options = text_classifier_options.TextClassifierOptions(
+        supported_model=ms.SupportedModels.MOBILEBERT_CLASSIFIER,
+        model_options=mo.BertClassifierOptions(do_fine_tuning=False, seq_len=2),
+        hparams=hp.BaseHParams(
+            epochs=1,
+            batch_size=1,
+            learning_rate=3e-5,
+            distribution_strategy='off'))
+    bert_classifier = text_classifier.TextClassifier.create(
+        train_data, validation_data, options)
+
+    _, accuracy = bert_classifier.evaluate(validation_data)
+    self.assertGreaterEqual(accuracy, 0.0)
+    # TODO: Add a unit test that does not run OOM.
+
+  def test_label_mismatch(self):
+    options = (
+        text_classifier_options.TextClassifierOptions(
+            supported_model=ms.SupportedModels.MOBILEBERT_CLASSIFIER))
+    train_tf_dataset = tf.data.Dataset.from_tensor_slices([[0]])
+    train_data = dataset.Dataset(train_tf_dataset, 1, ['foo'])
+    validation_tf_dataset = tf.data.Dataset.from_tensor_slices([[0]])
+    validation_data = dataset.Dataset(validation_tf_dataset, 1, ['bar'])
+    with self.assertRaisesRegex(
+        ValueError,
+        'Training data label names .* not equal to validation data label names'
+    ):
+      text_classifier.TextClassifier.create(train_data, validation_data,
+                                            options)
+
+  def test_options_mismatch(self):
+    train_data, validation_data = self._get_data()
+
+    avg_options = (
+        text_classifier_options.TextClassifierOptions(
+            supported_model=ms.SupportedModels.MOBILEBERT_CLASSIFIER,
+            model_options=mo.AverageWordEmbeddingClassifierOptions()))
+    with self.assertRaisesRegex(
+        ValueError, 'Expected AVERAGE_WORD_EMBEDDING_CLASSIFIER, got'
+        ' SupportedModels.MOBILEBERT_CLASSIFIER'):
+      text_classifier.TextClassifier.create(train_data, validation_data,
+                                            avg_options)
+
+    bert_options = (
+        text_classifier_options.TextClassifierOptions(
+            supported_model=(
+                ms.SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER),
+            model_options=mo.BertClassifierOptions()))
+    with self.assertRaisesRegex(
+        ValueError, 'Expected MOBILEBERT_CLASSIFIER, got'
+        ' SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER'):
+      text_classifier.TextClassifier.create(train_data, validation_data,
+                                            bert_options)
+
+
+if __name__ == '__main__':
+  # Load compressed models from tensorflow_hub
+  os.environ['TFHUB_MODEL_LOAD_FORMAT'] = 'COMPRESSED'
+  tf.test.main()

mediapipe/model_maker/python/vision/gesture_recognizer/BUILD

@@ -0,0 +1,165 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Placeholder for internal Python strict test compatibility macro.
+# Placeholder for internal Python strict library and test compatibility macro.
+
+licenses(["notice"])
+
+package(
+    default_visibility = ["//mediapipe:__subpackages__"],
+)
+
+# TODO: Remove the unncessary test data once the demo data are moved to an open-sourced
+# directory.
+filegroup(
+    name = "test_data",
+    srcs = glob([
+        "test_data/**",
+    ]),
+)
+
+py_library(
+    name = "constants",
+    srcs = ["constants.py"],
+)
+
+# TODO: Change to py_library after migrating the MediaPipe hand solution
+# library to MediaPipe hand task library.
+py_library(
+    name = "dataset",
+    srcs = ["dataset.py"],
+    deps = [
+        ":constants",
+        "//mediapipe/model_maker/python/core/data:classification_dataset",
+        "//mediapipe/model_maker/python/core/data:data_util",
+        "//mediapipe/model_maker/python/core/utils:model_util",
+        "//mediapipe/python/solutions:hands",
+    ],
+)
+
+py_test(
+    name = "dataset_test",
+    srcs = ["dataset_test.py"],
+    data = [
+        ":test_data",
+        "//mediapipe/model_maker/models/gesture_recognizer:models",
+    ],
+    deps = [
+        ":dataset",
+        "//mediapipe/python/solutions:hands",
+        "//mediapipe/tasks/python/test:test_utils",
+    ],
+)
+
+py_library(
+    name = "hyperparameters",
+    srcs = ["hyperparameters.py"],
+    deps = [
+        "//mediapipe/model_maker/python/core:hyperparameters",
+    ],
+)
+
+py_library(
+    name = "model_options",
+    srcs = ["model_options.py"],
+)
+
+py_library(
+    name = "gesture_recognizer_options",
+    srcs = ["gesture_recognizer_options.py"],
+    deps = [
+        ":hyperparameters",
+        ":model_options",
+    ],
+)
+
+py_library(
+    name = "gesture_recognizer",
+    srcs = ["gesture_recognizer.py"],
+    data = ["//mediapipe/model_maker/models/gesture_recognizer:models"],
+    deps = [
+        ":constants",
+        ":gesture_recognizer_options",
+        ":hyperparameters",
+        ":metadata_writer",
+        ":model_options",
+        "//mediapipe/model_maker/python/core/data:classification_dataset",
+        "//mediapipe/model_maker/python/core/tasks:classifier",
+        "//mediapipe/model_maker/python/core/utils:loss_functions",
+        "//mediapipe/model_maker/python/core/utils:model_util",
+        "//mediapipe/tasks/python/metadata/metadata_writers:metadata_writer",
+    ],
+)
+
+py_library(
+    name = "gesture_recognizer_import",
+    srcs = ["__init__.py"],
+    deps = [
+        ":dataset",
+        ":gesture_recognizer",
+        ":gesture_recognizer_options",
+        ":hyperparameters",
+        ":model_options",
+    ],
+)
+
+py_library(
+    name = "metadata_writer",
+    srcs = ["metadata_writer.py"],
+    deps = [
+        "//mediapipe/tasks/python/metadata/metadata_writers:metadata_writer",
+        "//mediapipe/tasks/python/metadata/metadata_writers:writer_utils",
+    ],
+)
+
+py_test(
+    name = "gesture_recognizer_test",
+    size = "large",
+    srcs = ["gesture_recognizer_test.py"],
+    data = [
+        ":test_data",
+        "//mediapipe/model_maker/models/gesture_recognizer:models",
+    ],
+    shard_count = 2,
+    deps = [
+        ":gesture_recognizer_import",
+        "//mediapipe/model_maker/python/core/utils:test_util",
+        "//mediapipe/tasks/python/test:test_utils",
+    ],
+)
+
+py_test(
+    name = "metadata_writer_test",
+    srcs = ["metadata_writer_test.py"],
+    data = [
+        ":test_data",
+    ],
+    deps = [
+        ":metadata_writer",
+        "//mediapipe/tasks/python/metadata/metadata_writers:metadata_writer",
+        "//mediapipe/tasks/python/test:test_utils",
+    ],
+)
+
+py_binary(
+    name = "gesture_recognizer_demo",
+    srcs = ["gesture_recognizer_demo.py"],
+    data = [
+        ":test_data",
+        "//mediapipe/model_maker/models/gesture_recognizer:models",
+    ],
+    python_version = "PY3",
+    deps = [":gesture_recognizer_import"],
+)

mediapipe/model_maker/python/vision/gesture_recognizer/__init__.py

@@ -0,0 +1,27 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""MediaPipe Model Maker Python Public API For Gesture Recognizer."""
+
+from mediapipe.model_maker.python.vision.gesture_recognizer import dataset
+from mediapipe.model_maker.python.vision.gesture_recognizer import gesture_recognizer
+from mediapipe.model_maker.python.vision.gesture_recognizer import gesture_recognizer_options
+from mediapipe.model_maker.python.vision.gesture_recognizer import hyperparameters
+from mediapipe.model_maker.python.vision.gesture_recognizer import model_options
+
+GestureRecognizer = gesture_recognizer.GestureRecognizer
+ModelOptions = model_options.GestureRecognizerModelOptions
+HParams = hyperparameters.HParams
+Dataset = dataset.Dataset
+HandDataPreprocessingParams = dataset.HandDataPreprocessingParams
+GestureRecognizerOptions = gesture_recognizer_options.GestureRecognizerOptions

mediapipe/model_maker/python/vision/gesture_recognizer/constants.py

@@ -0,0 +1,20 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Gesture recognition constants."""
+
+GESTURE_EMBEDDER_KERAS_MODEL_PATH = 'mediapipe/model_maker/models/gesture_recognizer/gesture_embedder'
+GESTURE_EMBEDDER_TFLITE_MODEL_FILE = 'mediapipe/model_maker/models/gesture_recognizer/gesture_embedder.tflite'
+HAND_DETECTOR_TFLITE_MODEL_FILE = 'mediapipe/model_maker/models/gesture_recognizer/palm_detection_full.tflite'
+HAND_LANDMARKS_DETECTOR_TFLITE_MODEL_FILE = 'mediapipe/model_maker/models/gesture_recognizer/hand_landmark_full.tflite'
+CANNED_GESTURE_CLASSIFIER_TFLITE_MODEL_FILE = 'mediapipe/model_maker/models/gesture_recognizer/canned_gesture_classifier.tflite'

mediapipe/model_maker/python/vision/gesture_recognizer/dataset.py

@@ -0,0 +1,238 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Gesture recognition dataset library."""
+
+import dataclasses
+import os
+import random
+from typing import List, NamedTuple, Optional
+
+import cv2
+import tensorflow as tf
+
+from mediapipe.model_maker.python.core.data import classification_dataset
+from mediapipe.model_maker.python.core.data import data_util
+from mediapipe.model_maker.python.core.utils import model_util
+from mediapipe.model_maker.python.vision.gesture_recognizer import constants
+from mediapipe.python.solutions import hands as mp_hands
+
+
+@dataclasses.dataclass
+class HandDataPreprocessingParams:
+  """A dataclass wraps the hand data preprocessing hyperparameters.
+
+  Attributes:
+    shuffle: A boolean controlling if shuffle the dataset. Default to true.
+    min_detection_confidence: confidence threshold for hand detection.
+  """
+  shuffle: bool = True
+  min_detection_confidence: float = 0.7
+
+
+@dataclasses.dataclass
+class HandData:
+  """A dataclass represents hand data for training gesture recognizer model.
+
+  See https://google.github.io/mediapipe/solutions/hands#mediapipe-hands for
+  more details of the hand gesture data API.
+
+  Attributes:
+    hand: normalized hand landmarks of shape 21x3 from the screen based
+      hand-landmark model.
+    world_hand: hand landmarks of shape 21x3 in world coordinates.
+    handedness: Collection of handedness confidence of the detected hands (i.e.
+      is it a left or right hand).
+  """
+  hand: List[List[float]]
+  world_hand: List[List[float]]
+  handedness: List[float]
+
+
+def _validate_data_sample(data: NamedTuple) -> bool:
+  """Validates the input hand data sample.
+
+  Args:
+    data: input hand data sample.
+
+  Returns:
+    False if the input data namedtuple does not contain the fields including
+    'multi_hand_landmarks' or 'multi_hand_world_landmarks' or 'multi_handedness'
+    or any of these attributes' values are none. Otherwise, True.
+  """
+  if (not hasattr(data, 'multi_hand_landmarks') or
+      data.multi_hand_landmarks is None):
+    return False
+  if (not hasattr(data, 'multi_hand_world_landmarks') or
+      data.multi_hand_world_landmarks is None):
+    return False
+  if not hasattr(data, 'multi_handedness') or data.multi_handedness is None:
+    return False
+  return True
+
+
+def _get_hand_data(all_image_paths: List[str],
+                   min_detection_confidence: float) -> Optional[HandData]:
+  """Computes hand data (landmarks and handedness) in the input image.
+
+  Args:
+    all_image_paths: all input image paths.
+    min_detection_confidence: hand detection confidence threshold
+
+  Returns:
+    A HandData object. Returns None if no hand is detected.
+  """
+  hand_data_result = []
+  with mp_hands.Hands(
+      static_image_mode=True,
+      max_num_hands=1,
+      min_detection_confidence=min_detection_confidence) as hands:
+    for path in all_image_paths:
+      tf.compat.v1.logging.info('Loading image %s', path)
+      image = data_util.load_image(path)
+      # Flip image around y-axis for correct handedness output
+      image = cv2.flip(image, 1)
+      data = hands.process(image)
+      if not _validate_data_sample(data):
+        hand_data_result.append(None)
+        continue
+      hand_landmarks = [[
+          hand_landmark.x, hand_landmark.y, hand_landmark.z
+      ] for hand_landmark in data.multi_hand_landmarks[0].landmark]
+      hand_world_landmarks = [[
+          hand_landmark.x, hand_landmark.y, hand_landmark.z
+      ] for hand_landmark in data.multi_hand_world_landmarks[0].landmark]
+      handedness_scores = [
+          handedness.score
+          for handedness in data.multi_handedness[0].classification
+      ]
+      hand_data_result.append(
+          HandData(
+              hand=hand_landmarks,
+              world_hand=hand_world_landmarks,
+              handedness=handedness_scores))
+  return hand_data_result
+
+
+class Dataset(classification_dataset.ClassificationDataset):
+  """Dataset library for hand gesture recognizer."""
+
+  @classmethod
+  def from_folder(
+      cls,
+      dirname: str,
+      hparams: Optional[HandDataPreprocessingParams] = None
+  ) -> classification_dataset.ClassificationDataset:
+    """Loads images and labels from the given directory.
+
+    Directory contents are expected to be in the format:
+    <root_dir>/<gesture_name>/*.jpg". One of the `gesture_name` must be `none`
+    (case insensitive). The `none` sub-directory is expected to contain images
+    of hands that don't belong to other gesture classes in <root_dir>. Assumes
+    the image data of the same label are in the same subdirectory.
+
+    Args:
+      dirname: Name of the directory containing the data files.
+      hparams: Optional hyperparameters for processing input hand gesture
+        images.
+
+    Returns:
+      Dataset containing landmarks, labels, and other related info.
+
+    Raises:
+      ValueError: if the input data directory is empty or the label set does not
+        contain label 'none' (case insensitive).
+    """
+    data_root = os.path.abspath(dirname)
+
+    # Assumes the image data of the same label are in the same subdirectory,
+    # gets image path and label names.
+    all_image_paths = list(tf.io.gfile.glob(data_root + r'/*/*'))
+    if not all_image_paths:
+      raise ValueError('Image dataset directory is empty.')
+
+    if not hparams:
+      hparams = HandDataPreprocessingParams()
+
+    if hparams.shuffle:
+      # Random shuffle data.
+      random.shuffle(all_image_paths)
+
+    label_names = sorted(
+        name for name in os.listdir(data_root)
+        if os.path.isdir(os.path.join(data_root, name)))
+    if 'none' not in [v.lower() for v in label_names]:
+      raise ValueError('Label set does not contain label "None".')
+    # Move label 'none' to the front of label list.
+    none_idx = [v.lower() for v in label_names].index('none')
+    none_value = label_names.pop(none_idx)
+    label_names.insert(0, none_value)
+
+    index_by_label = dict(
+        (name, index) for index, name in enumerate(label_names))
+    all_gesture_indices = [
+        index_by_label[os.path.basename(os.path.dirname(path))]
+        for path in all_image_paths
+    ]
+
+    # Compute hand data (including local hand landmark, world hand landmark, and
+    # handedness) for all the input images.
+    hand_data = _get_hand_data(
+        all_image_paths=all_image_paths,
+        min_detection_confidence=hparams.min_detection_confidence)
+
+    # Get a list of the valid hand landmark sample in the hand data list.
+    valid_indices = [
+        i for i in range(len(hand_data)) if hand_data[i] is not None
+    ]
+    # Remove 'None' element from the hand data and label list.
+    valid_hand_data = [dataclasses.asdict(hand_data[i]) for i in valid_indices]
+    if not valid_hand_data:
+      raise ValueError('No valid hand is detected.')
+
+    valid_label = [all_gesture_indices[i] for i in valid_indices]
+
+    # Convert list of dictionaries to dictionary of lists.
+    hand_data_dict = {
+        k: [lm[k] for lm in valid_hand_data] for k in valid_hand_data[0]
+    }
+    hand_ds = tf.data.Dataset.from_tensor_slices(hand_data_dict)
+
+    embedder_model = model_util.load_keras_model(
+        constants.GESTURE_EMBEDDER_KERAS_MODEL_PATH)
+
+    hand_ds = hand_ds.batch(batch_size=1)
+    hand_embedding_ds = hand_ds.map(
+        map_func=lambda feature: embedder_model(dict(feature)),
+        num_parallel_calls=tf.data.experimental.AUTOTUNE)
+    hand_embedding_ds = hand_embedding_ds.unbatch()
+
+    # Create label dataset
+    label_ds = tf.data.Dataset.from_tensor_slices(
+        tf.cast(valid_label, tf.int64))
+
+    label_one_hot_ds = label_ds.map(
+        map_func=lambda index: tf.one_hot(index, len(label_names)),
+        num_parallel_calls=tf.data.experimental.AUTOTUNE)
+
+    # Create a dataset with (hand_embedding, one_hot_label) pairs
+    hand_embedding_label_ds = tf.data.Dataset.zip(
+        (hand_embedding_ds, label_one_hot_ds))
+
+    tf.compat.v1.logging.info(
+        'Load valid hands with size: {}, num_label: {}, labels: {}.'.format(
+            len(valid_hand_data), len(label_names), ','.join(label_names)))
+    return Dataset(
+        dataset=hand_embedding_label_ds,
+        size=len(valid_hand_data),
+        label_names=label_names)

mediapipe/model_maker/python/vision/gesture_recognizer/dataset_test.py

@@ -0,0 +1,161 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.s
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import collections
+import os
+import shutil
+from typing import NamedTuple
+import unittest
+
+from absl import flags
+from absl.testing import parameterized
+import tensorflow as tf
+
+from mediapipe.model_maker.python.vision.gesture_recognizer import dataset
+from mediapipe.python.solutions import hands as mp_hands
+from mediapipe.tasks.python.test import test_utils
+
+FLAGS = flags.FLAGS
+
+_TEST_DATA_DIRNAME = 'raw_data'
+
+
+class DatasetTest(tf.test.TestCase, parameterized.TestCase):
+
+  def test_split(self):
+    input_data_dir = test_utils.get_test_data_path(_TEST_DATA_DIRNAME)
+    data = dataset.Dataset.from_folder(
+        dirname=input_data_dir, hparams=dataset.HandDataPreprocessingParams())
+    train_data, test_data = data.split(0.5)
+
+    self.assertLen(train_data, 17)
+    for _, elem in enumerate(train_data.gen_tf_dataset(is_training=True)):
+      self.assertEqual(elem[0].shape, (1, 128))
+      self.assertEqual(elem[1].shape, ([1, 4]))
+    self.assertEqual(train_data.num_classes, 4)
+    self.assertEqual(train_data.label_names, ['none', 'call', 'four', 'rock'])
+
+    self.assertLen(test_data, 18)
+    for _, elem in enumerate(test_data.gen_tf_dataset(is_training=True)):
+      self.assertEqual(elem[0].shape, (1, 128))
+      self.assertEqual(elem[1].shape, ([1, 4]))
+    self.assertEqual(test_data.num_classes, 4)
+    self.assertEqual(test_data.label_names, ['none', 'call', 'four', 'rock'])
+
+  def test_from_folder(self):
+    input_data_dir = test_utils.get_test_data_path(_TEST_DATA_DIRNAME)
+    data = dataset.Dataset.from_folder(
+        dirname=input_data_dir, hparams=dataset.HandDataPreprocessingParams())
+    for _, elem in enumerate(data.gen_tf_dataset(is_training=True)):
+      self.assertEqual(elem[0].shape, (1, 128))
+      self.assertEqual(elem[1].shape, ([1, 4]))
+    self.assertLen(data, 35)
+    self.assertEqual(data.num_classes, 4)
+    self.assertEqual(data.label_names, ['none', 'call', 'four', 'rock'])
+
+  def test_create_dataset_from_empty_folder_raise_value_error(self):
+    with self.assertRaisesRegex(ValueError, 'Image dataset directory is empty'):
+      dataset.Dataset.from_folder(
+          dirname=self.get_temp_dir(),
+          hparams=dataset.HandDataPreprocessingParams())
+
+  def test_create_dataset_from_folder_without_none_raise_value_error(self):
+    input_data_dir = test_utils.get_test_data_path(_TEST_DATA_DIRNAME)
+    tmp_dir = self.create_tempdir()
+    # Copy input dataset to a temporary directory and skip 'None' directory.
+    for name in os.listdir(input_data_dir):
+      if name == 'none':
+        continue
+      src_dir = os.path.join(input_data_dir, name)
+      dst_dir = os.path.join(tmp_dir, name)
+      shutil.copytree(src_dir, dst_dir)
+
+    with self.assertRaisesRegex(ValueError,
+                                'Label set does not contain label "None"'):
+      dataset.Dataset.from_folder(
+          dirname=tmp_dir, hparams=dataset.HandDataPreprocessingParams())
+
+  def test_create_dataset_from_folder_with_capital_letter_in_folder_name(self):
+    input_data_dir = test_utils.get_test_data_path(_TEST_DATA_DIRNAME)
+    tmp_dir = self.create_tempdir()
+    # Copy input dataset to a temporary directory and change the base folder
+    # name to upper case letter, e.g. 'none' -> 'NONE'
+    for name in os.listdir(input_data_dir):
+      src_dir = os.path.join(input_data_dir, name)
+      dst_dir = os.path.join(tmp_dir, name.upper())
+      shutil.copytree(src_dir, dst_dir)
+
+    upper_base_folder_name = list(os.listdir(tmp_dir))
+    self.assertCountEqual(upper_base_folder_name,
+                          ['CALL', 'FOUR', 'NONE', 'ROCK'])
+
+    data = dataset.Dataset.from_folder(
+        dirname=tmp_dir, hparams=dataset.HandDataPreprocessingParams())
+    for _, elem in enumerate(data.gen_tf_dataset(is_training=True)):
+      self.assertEqual(elem[0].shape, (1, 128))
+      self.assertEqual(elem[1].shape, ([1, 4]))
+    self.assertLen(data, 35)
+    self.assertEqual(data.num_classes, 4)
+    self.assertEqual(data.label_names, ['NONE', 'CALL', 'FOUR', 'ROCK'])
+
+  @parameterized.named_parameters(
+      dict(
+          testcase_name='invalid_field_name_multi_hand_landmark',
+          hand=collections.namedtuple('Hand', [
+              'multi_hand_landmark', 'multi_hand_world_landmarks',
+              'multi_handedness'
+          ])(1, 2, 3)),
+      dict(
+          testcase_name='invalid_field_name_multi_hand_world_landmarks',
+          hand=collections.namedtuple('Hand', [
+              'multi_hand_landmarks', 'multi_hand_world_landmark',
+              'multi_handedness'
+          ])(1, 2, 3)),
+      dict(
+          testcase_name='invalid_field_name_multi_handed',
+          hand=collections.namedtuple('Hand', [
+              'multi_hand_landmarks', 'multi_hand_world_landmarks',
+              'multi_handed'
+          ])(1, 2, 3)),
+      dict(
+          testcase_name='multi_hand_landmarks_is_none',
+          hand=collections.namedtuple('Hand', [
+              'multi_hand_landmarks', 'multi_hand_world_landmarks',
+              'multi_handedness'
+          ])(None, 2, 3)),
+      dict(
+          testcase_name='multi_hand_world_landmarks_is_none',
+          hand=collections.namedtuple('Hand', [
+              'multi_hand_landmarks', 'multi_hand_world_landmarks',
+              'multi_handedness'
+          ])(1, None, 3)),
+      dict(
+          testcase_name='multi_handedness_is_none',
+          hand=collections.namedtuple('Hand', [
+              'multi_hand_landmarks', 'multi_hand_world_landmarks',
+              'multi_handedness'
+          ])(1, 2, None)),
+  )
+  def test_create_dataset_from_invalid_hand_data(self, hand: NamedTuple):
+    with unittest.mock.patch.object(
+        mp_hands.Hands, 'process', return_value=hand):
+      input_data_dir = test_utils.get_test_data_path(_TEST_DATA_DIRNAME)
+      with self.assertRaisesRegex(ValueError, 'No valid hand is detected'):
+        dataset.Dataset.from_folder(
+            dirname=input_data_dir,
+            hparams=dataset.HandDataPreprocessingParams())
+
+
+if __name__ == '__main__':
+  tf.test.main()

mediapipe/model_maker/python/vision/gesture_recognizer/gesture_recognizer.py

@@ -0,0 +1,239 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""APIs to train gesture recognizer model."""
+
+import os
+from typing import List
+
+import tensorflow as tf
+
+from mediapipe.model_maker.python.core.data import classification_dataset as classification_ds
+from mediapipe.model_maker.python.core.tasks import classifier
+from mediapipe.model_maker.python.core.utils import loss_functions
+from mediapipe.model_maker.python.core.utils import model_util
+from mediapipe.model_maker.python.vision.gesture_recognizer import constants
+from mediapipe.model_maker.python.vision.gesture_recognizer import gesture_recognizer_options
+from mediapipe.model_maker.python.vision.gesture_recognizer import hyperparameters as hp
+from mediapipe.model_maker.python.vision.gesture_recognizer import metadata_writer
+from mediapipe.model_maker.python.vision.gesture_recognizer import model_options as model_opt
+from mediapipe.tasks.python.metadata.metadata_writers import metadata_writer as base_metadata_writer
+
+_EMBEDDING_SIZE = 128
+
+
+class GestureRecognizer(classifier.Classifier):
+  """GestureRecognizer for building hand gesture recognizer model.
+
+  Attributes:
+    embedding_size: Size of the input gesture embedding vector.
+  """
+
+  def __init__(self, label_names: List[str],
+               model_options: model_opt.GestureRecognizerModelOptions,
+               hparams: hp.HParams):
+    """Initializes GestureRecognizer class.
+
+    Args:
+      label_names: A list of label names for the classes.
+      model_options: options to create gesture recognizer model.
+      hparams: The hyperparameters for training hand gesture recognizer model.
+    """
+    super().__init__(
+        model_spec=None, label_names=label_names, shuffle=hparams.shuffle)
+    self._model_options = model_options
+    self._hparams = hparams
+    self._history = None
+    self.embedding_size = _EMBEDDING_SIZE
+
+  @classmethod
+  def create(
+      cls,
+      train_data: classification_ds.ClassificationDataset,
+      validation_data: classification_ds.ClassificationDataset,
+      options: gesture_recognizer_options.GestureRecognizerOptions,
+  ) -> 'GestureRecognizer':
+    """Creates and trains a hand gesture recognizer with input datasets.
+
+    If a checkpoint file exists in the {options.hparams.export_dir}/checkpoint/
+    directory, the training process will load the weight from the checkpoint
+    file for continual training.
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data. If None, skips validation process.
+      options: options for creating and training gesture recognizer model.
+
+    Returns:
+      An instance of GestureRecognizer.
+    """
+    if options.model_options is None:
+      options.model_options = model_opt.GestureRecognizerModelOptions()
+
+    if options.hparams is None:
+      options.hparams = hp.HParams()
+
+    gesture_recognizer = cls(
+        label_names=train_data.label_names,
+        model_options=options.model_options,
+        hparams=options.hparams)
+
+    gesture_recognizer._create_model()
+
+    train_dataset = train_data.gen_tf_dataset(
+        batch_size=options.hparams.batch_size,
+        is_training=True,
+        shuffle=options.hparams.shuffle)
+    options.hparams.steps_per_epoch = model_util.get_steps_per_epoch(
+        steps_per_epoch=options.hparams.steps_per_epoch,
+        batch_size=options.hparams.batch_size,
+        train_data=train_data)
+    train_dataset = train_dataset.take(count=options.hparams.steps_per_epoch)
+
+    validation_dataset = validation_data.gen_tf_dataset(
+        batch_size=options.hparams.batch_size, is_training=False)
+
+    tf.compat.v1.logging.info('Training the gesture recognizer model...')
+    gesture_recognizer._train(
+        train_data=train_dataset, validation_data=validation_dataset)
+
+    return gesture_recognizer
+
+  def _train(self, train_data: tf.data.Dataset,
+             validation_data: tf.data.Dataset):
+    """Trains the model with input train_data.
+
+    The training results are recorded by a self.History object returned by
+    tf.keras.Model.fit().
+
+    Args:
+      train_data: Training data.
+      validation_data: Validation data.
+    """
+    hparams = self._hparams
+
+    scheduler = lambda epoch: hparams.learning_rate * (hparams.lr_decay**epoch)
+    scheduler_callback = tf.keras.callbacks.LearningRateScheduler(scheduler)
+
+    job_dir = hparams.export_dir
+    checkpoint_path = os.path.join(job_dir, 'epoch_models')
+    checkpoint_callback = tf.keras.callbacks.ModelCheckpoint(
+        os.path.join(checkpoint_path, 'model-{epoch:04d}'),
+        save_weights_only=True)
+
+    best_model_path = os.path.join(job_dir, 'best_model_weights')
+    best_model_callback = tf.keras.callbacks.ModelCheckpoint(
+        best_model_path,
+        monitor='val_loss',
+        mode='min',
+        save_best_only=True,
+        save_weights_only=True)
+
+    tensorboard_callback = tf.keras.callbacks.TensorBoard(
+        log_dir=os.path.join(job_dir, 'logs'))
+
+    self._model.compile(
+        optimizer='adam',
+        loss=loss_functions.FocalLoss(gamma=self._hparams.gamma),
+        metrics=['categorical_accuracy'])
+
+    latest_checkpoint = tf.train.latest_checkpoint(checkpoint_path)
+    if latest_checkpoint:
+      print(f'Resuming from {latest_checkpoint}')
+      self._model.load_weights(latest_checkpoint)
+
+    self._history = self._model.fit(
+        x=train_data,
+        epochs=hparams.epochs,
+        validation_data=validation_data,
+        validation_freq=1,
+        callbacks=[
+            checkpoint_callback, best_model_callback, scheduler_callback,
+            tensorboard_callback
+        ],
+    )
+
+  def _create_model(self):
+    """Creates the hand gesture recognizer model.
+
+    The gesture embedding model is pretrained and loaded from a tf.saved_model.
+    """
+    inputs = tf.keras.Input(
+        shape=[self.embedding_size],
+        batch_size=None,
+        dtype=tf.float32,
+        name='hand_embedding')
+
+    x = tf.keras.layers.BatchNormalization()(inputs)
+    x = tf.keras.layers.ReLU()(x)
+    dropout_rate = self._model_options.dropout_rate
+    x = tf.keras.layers.Dropout(rate=dropout_rate, name='dropout')(x)
+    outputs = tf.keras.layers.Dense(
+        self._num_classes,
+        activation='softmax',
+        name='custom_gesture_recognizer')(
+            x)
+
+    self._model = tf.keras.Model(inputs=inputs, outputs=outputs)
+
+    print(self._model.summary())
+
+  def export_model(self, model_name: str = 'gesture_recognizer.task'):
+    """Converts the model to TFLite and exports as a model bundle file.
+
+    Saves a model bundle file and metadata json file to hparams.export_dir. The
+    resulting model bundle file will contain necessary models for hand
+    detection, canned gesture classification, and customized gesture
+    classification. Only the model bundle file is needed for the downstream
+    gesture recognition task. The metadata.json file is saved only to
+    interpret the contents of the model bundle file.
+
+    The customized gesture model is in float without quantization. The model is
+    lightweight and there is no need to balance performance and efficiency by
+    quantization. The default score_thresholding is set to 0.5 as it can be
+    adjusted during inference.
+
+    Args:
+      model_name: File name to save model bundle file. The full export path is
+        {export_dir}/{model_name}.
+    """
+    # TODO: Convert keras embedder model instead of using tflite
+    gesture_embedding_model_buffer = model_util.load_tflite_model_buffer(
+        constants.GESTURE_EMBEDDER_TFLITE_MODEL_FILE)
+    hand_detector_model_buffer = model_util.load_tflite_model_buffer(
+        constants.HAND_DETECTOR_TFLITE_MODEL_FILE)
+    hand_landmarks_detector_model_buffer = model_util.load_tflite_model_buffer(
+        constants.HAND_LANDMARKS_DETECTOR_TFLITE_MODEL_FILE)
+    canned_gesture_model_buffer = model_util.load_tflite_model_buffer(
+        constants.CANNED_GESTURE_CLASSIFIER_TFLITE_MODEL_FILE)
+
+    if not tf.io.gfile.exists(self._hparams.export_dir):
+      tf.io.gfile.makedirs(self._hparams.export_dir)
+    model_bundle_file = os.path.join(self._hparams.export_dir, model_name)
+    metadata_file = os.path.join(self._hparams.export_dir, 'metadata.json')
+
+    gesture_classifier_options = metadata_writer.GestureClassifierOptions(
+        model_buffer=model_util.convert_to_tflite(self._model),
+        labels=base_metadata_writer.Labels().add(list(self._label_names)),
+        score_thresholding=base_metadata_writer.ScoreThresholding(
+            global_score_threshold=0.5))
+
+    writer = metadata_writer.MetadataWriter.create(
+        hand_detector_model_buffer, hand_landmarks_detector_model_buffer,
+        gesture_embedding_model_buffer, canned_gesture_model_buffer,
+        gesture_classifier_options)
+    model_bundle_content, metadata_json = writer.populate()
+    with open(model_bundle_file, 'wb') as f:
+      f.write(model_bundle_content)
+    with open(metadata_file, 'w') as f:
+      f.write(metadata_json)

mediapipe/model_maker/python/vision/gesture_recognizer/gesture_recognizer_demo.py

@@ -0,0 +1,78 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Demo for making an gesture recognizer model by Mediapipe Model Maker."""
+
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import os
+
+# Dependency imports
+
+from absl import app
+from absl import flags
+from absl import logging
+
+from mediapipe.model_maker.python.vision import gesture_recognizer
+
+FLAGS = flags.FLAGS
+
+# TODO: Move hand gesture recognizer demo dataset to an
+# open-sourced directory.
+TEST_DATA_DIR = 'mediapipe/model_maker/python/vision/gesture_recognizer/test_data/raw_data'
+
+
+def define_flags():
+  flags.DEFINE_string('export_dir', None,
+                      'The directory to save exported files.')
+  flags.DEFINE_string('input_data_dir', None,
+                      'The directory with input training data.')
+  flags.mark_flag_as_required('export_dir')
+
+
+def run(data_dir: str, export_dir: str):
+  """Runs demo."""
+  data = gesture_recognizer.Dataset.from_folder(dirname=data_dir)
+  train_data, rest_data = data.split(0.8)
+  validation_data, test_data = rest_data.split(0.5)
+
+  model = gesture_recognizer.GestureRecognizer.create(
+      train_data=train_data,
+      validation_data=validation_data,
+      options=gesture_recognizer.GestureRecognizerOptions(
+          hparams=gesture_recognizer.HParams(export_dir=export_dir)))
+
+  metric = model.evaluate(test_data, batch_size=2)
+  print('Evaluation metric')
+  print(metric)
+
+  model.export_model()
+
+
+def main(_):
+  logging.set_verbosity(logging.INFO)
+
+  if FLAGS.input_data_dir is None:
+    data_dir = os.path.join(FLAGS.test_srcdir, TEST_DATA_DIR)
+  else:
+    data_dir = FLAGS.input_data_dir
+
+  export_dir = os.path.expanduser(FLAGS.export_dir)
+  run(data_dir=data_dir, export_dir=export_dir)
+
+
+if __name__ == '__main__':
+  define_flags()
+  app.run(main)

mediapipe/model_maker/python/vision/gesture_recognizer/gesture_recognizer_options.py

@@ -0,0 +1,32 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Options for building gesture recognizer."""
+
+import dataclasses
+from typing import Optional
+
+from mediapipe.model_maker.python.vision.gesture_recognizer import hyperparameters
+from mediapipe.model_maker.python.vision.gesture_recognizer import model_options as model_opt
+
+
+@dataclasses.dataclass
+class GestureRecognizerOptions:
+  """Configurable options for building gesture recognizer.
+
+  Attributes:
+    model_options: A set of options for configuring the selected model.
+    hparams: A set of hyperparameters used to train the gesture recognizer.
+  """
+  model_options: Optional[model_opt.GestureRecognizerModelOptions] = None
+  hparams: Optional[hyperparameters.HParams] = None

mediapipe/model_maker/python/vision/gesture_recognizer/gesture_recognizer_test.py

@@ -0,0 +1,132 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the 'License');
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an 'AS IS' BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import io
+import os
+from unittest import mock as unittest_mock
+import zipfile
+
+import mock
+import tensorflow as tf
+
+from mediapipe.model_maker.python.core.utils import test_util
+from mediapipe.model_maker.python.vision import gesture_recognizer
+from mediapipe.tasks.python.test import test_utils
+
+_TEST_DATA_DIR = 'mediapipe/model_maker/python/vision/gesture_recognizer/test_data'
+
+
+class GestureRecognizerTest(tf.test.TestCase):
+
+  def _load_data(self):
+    input_data_dir = test_utils.get_test_data_path(
+        os.path.join(_TEST_DATA_DIR, 'raw_data'))
+
+    data = gesture_recognizer.Dataset.from_folder(
+        dirname=input_data_dir,
+        hparams=gesture_recognizer.HandDataPreprocessingParams(shuffle=True))
+    return data
+
+  def setUp(self):
+    super().setUp()
+    self._model_options = gesture_recognizer.ModelOptions()
+    self._hparams = gesture_recognizer.HParams(epochs=2)
+    self._gesture_recognizer_options = (
+        gesture_recognizer.GestureRecognizerOptions(
+            model_options=self._model_options, hparams=self._hparams))
+    all_data = self._load_data()
+    # Splits data, 90% data for training, 10% for testing
+    self._train_data, self._test_data = all_data.split(0.9)
+
+  def test_gesture_recognizer_model(self):
+    model = gesture_recognizer.GestureRecognizer.create(
+        train_data=self._train_data,
+        validation_data=self._test_data,
+        options=self._gesture_recognizer_options)
+
+    self._test_accuracy(model)
+
+  def test_export_gesture_recognizer_model(self):
+    model = gesture_recognizer.GestureRecognizer.create(
+        train_data=self._train_data,
+        validation_data=self._test_data,
+        options=self._gesture_recognizer_options)
+    model.export_model()
+    model_bundle_file = os.path.join(self._hparams.export_dir,
+                                     'gesture_recognizer.task')
+    with zipfile.ZipFile(model_bundle_file) as zf:
+      self.assertEqual(
+          set(zf.namelist()),
+          set(['hand_landmarker.task', 'hand_gesture_recognizer.task']))
+      zf.extractall(self.get_temp_dir())
+    hand_gesture_recognizer_bundle_file = os.path.join(
+        self.get_temp_dir(), 'hand_gesture_recognizer.task')
+    with zipfile.ZipFile(hand_gesture_recognizer_bundle_file) as zf:
+      self.assertEqual(
+          set(zf.namelist()),
+          set([
+              'canned_gesture_classifier.tflite',
+              'custom_gesture_classifier.tflite', 'gesture_embedder.tflite'
+          ]))
+      zf.extractall(self.get_temp_dir())
+    gesture_classifier_tflite_file = os.path.join(
+        self.get_temp_dir(), 'custom_gesture_classifier.tflite')
+    test_util.test_tflite_file(
+        keras_model=model._model,
+        tflite_file=gesture_classifier_tflite_file,
+        size=[1, model.embedding_size])
+
+  def _test_accuracy(self, model, threshold=0.5):
+    _, accuracy = model.evaluate(self._test_data)
+    tf.compat.v1.logging.info(f'accuracy: {accuracy}')
+    self.assertGreaterEqual(accuracy, threshold)
+
+  @unittest_mock.patch.object(
+      gesture_recognizer.hyperparameters,
+      'HParams',
+      autospec=True,
+      return_value=gesture_recognizer.HParams(epochs=1))
+  @unittest_mock.patch.object(
+      gesture_recognizer.model_options,
+      'GestureRecognizerModelOptions',
+      autospec=True,
+      return_value=gesture_recognizer.ModelOptions())
+  def test_create_hparams_and_model_options_if_none_in_image_classifier_options(
+      self, mock_hparams, mock_model_options):
+    options = gesture_recognizer.GestureRecognizerOptions()
+    gesture_recognizer.GestureRecognizer.create(
+        train_data=self._train_data,
+        validation_data=self._test_data,
+        options=options)
+    mock_hparams.assert_called_once()
+    mock_model_options.assert_called_once()
+
+  def test_continual_training_by_loading_checkpoint(self):
+    mock_stdout = io.StringIO()
+    with mock.patch('sys.stdout', mock_stdout):
+      model = gesture_recognizer.GestureRecognizer.create(
+          train_data=self._train_data,
+          validation_data=self._test_data,
+          options=self._gesture_recognizer_options)
+      model = gesture_recognizer.GestureRecognizer.create(
+          train_data=self._train_data,
+          validation_data=self._test_data,
+          options=self._gesture_recognizer_options)
+      self._test_accuracy(model)
+
+    self.assertRegex(mock_stdout.getvalue(), 'Resuming from')
+
+
+if __name__ == '__main__':
+  tf.test.main()

mediapipe/model_maker/python/vision/gesture_recognizer/hyperparameters.py

@@ -0,0 +1,40 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Hyperparameters for training customized gesture recognizer models."""
+
+import dataclasses
+
+from mediapipe.model_maker.python.core import hyperparameters as hp
+
+
+@dataclasses.dataclass
+class HParams(hp.BaseHParams):
+  """The hyperparameters for training gesture recognizer.
+
+  Attributes:
+    learning_rate: Learning rate to use for gradient descent training.
+    batch_size: Batch size for training.
+    epochs: Number of training iterations over the dataset.
+    lr_decay: Learning rate decay to use for gradient descent training.
+    gamma: Gamma parameter for focal loss.
+  """
+  # Parameters from BaseHParams class.
+  learning_rate: float = 0.001
+  batch_size: int = 2
+  epochs: int = 10
+
+  # Parameters about training configuration
+  # TODO: Move lr_decay to hp.baseHParams.
+  lr_decay: float = 0.99
+  gamma: int = 2

mediapipe/model_maker/python/vision/gesture_recognizer/metadata_writer.py

@@ -0,0 +1,193 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+"""Writes metadata and creates model asset bundle for gesture recognizer."""
+
+import dataclasses
+import os
+import tempfile
+from typing import Union
+
+import tensorflow as tf
+from mediapipe.tasks.python.metadata.metadata_writers import metadata_writer
+from mediapipe.tasks.python.metadata.metadata_writers import writer_utils
+
+_HAND_DETECTOR_TFLITE_NAME = "hand_detector.tflite"
+_HAND_LANDMARKS_DETECTOR_TFLITE_NAME = "hand_landmarks_detector.tflite"
+_HAND_LANDMARKER_BUNDLE_NAME = "hand_landmarker.task"
+_HAND_GESTURE_RECOGNIZER_BUNDLE_NAME = "hand_gesture_recognizer.task"
+_GESTURE_EMBEDDER_TFLITE_NAME = "gesture_embedder.tflite"
+_CANNED_GESTURE_CLASSIFIER_TFLITE_NAME = "canned_gesture_classifier.tflite"
+_CUSTOM_GESTURE_CLASSIFIER_TFLITE_NAME = "custom_gesture_classifier.tflite"
+
+_MODEL_NAME = "HandGestureRecognition"
+_MODEL_DESCRIPTION = "Recognize the hand gesture in the image."
+
+_INPUT_NAME = "embedding"
+_INPUT_DESCRIPTION = "Embedding feature vector from gesture embedder."
+_OUTPUT_NAME = "scores"
+_OUTPUT_DESCRIPTION = "Hand gesture category scores."
+
+
+@dataclasses.dataclass
+class GestureClassifierOptions:
+  """Options to write metadata for gesture classifier.
+
+  Attributes:
+    model_buffer: Gesture classifier TFLite model buffer.
+    labels: Labels for the gesture classifier.
+    score_thresholding: Parameters to performs thresholding on output tensor
+      values [1].
+    [1]:
+      https://github.com/google/mediapipe/blob/f8af41b1eb49ff4bdad756ff19d1d36f486be614/mediapipe/tasks/metadata/metadata_schema.fbs#L468
+  """
+  model_buffer: bytearray
+  labels: metadata_writer.Labels
+  score_thresholding: metadata_writer.ScoreThresholding
+
+
+def read_file(file_path: str, mode: str = "rb") -> Union[str, bytes]:
+  with tf.io.gfile.GFile(file_path, mode) as f:
+    return f.read()
+
+
+class MetadataWriter:
+  """MetadataWriter to write the metadata and the model asset bundle."""
+
+  def __init__(
+      self, hand_detector_model_buffer: bytearray,
+      hand_landmarks_detector_model_buffer: bytearray,
+      gesture_embedder_model_buffer: bytearray,
+      canned_gesture_classifier_model_buffer: bytearray,
+      custom_gesture_classifier_metadata_writer: metadata_writer.MetadataWriter
+  ) -> None:
+    """Initialize MetadataWriter to write the metadata and model asset bundle.
+
+    Args:
+      hand_detector_model_buffer: A valid flatbuffer *with* metadata loaded from
+        the TFLite hand detector model file.
+      hand_landmarks_detector_model_buffer: A valid flatbuffer *with* metadata
+        loaded from the TFLite hand landmarks detector model file.
+      gesture_embedder_model_buffer: A valid flatbuffer *with* metadata loaded
+        from the TFLite gesture embedder model file.
+      canned_gesture_classifier_model_buffer: A valid flatbuffer *with* metadata
+        loaded from the TFLite canned gesture classifier model file.
+      custom_gesture_classifier_metadata_writer: Metadata writer to write custom
+        gesture classifier metadata into the TFLite file.
+    """
+    self._hand_detector_model_buffer = hand_detector_model_buffer
+    self._hand_landmarks_detector_model_buffer = hand_landmarks_detector_model_buffer
+    self._gesture_embedder_model_buffer = gesture_embedder_model_buffer
+    self._canned_gesture_classifier_model_buffer = canned_gesture_classifier_model_buffer
+    self._custom_gesture_classifier_metadata_writer = custom_gesture_classifier_metadata_writer
+    self._temp_folder = tempfile.TemporaryDirectory()
+
+  def __del__(self):
+    if os.path.exists(self._temp_folder.name):
+      self._temp_folder.cleanup()
+
+  @classmethod
+  def create(
+      cls,
+      hand_detector_model_buffer: bytearray,
+      hand_landmarks_detector_model_buffer: bytearray,
+      gesture_embedder_model_buffer: bytearray,
+      canned_gesture_classifier_model_buffer: bytearray,
+      custom_gesture_classifier_options: GestureClassifierOptions,
+  ) -> "MetadataWriter":
+    """Creates MetadataWriter to write the metadata for gesture recognizer.
+
+    Args:
+      hand_detector_model_buffer: A valid flatbuffer *with* metadata loaded from
+        the TFLite hand detector model file.
+      hand_landmarks_detector_model_buffer: A valid flatbuffer *with* metadata
+        loaded from the TFLite hand landmarks detector model file.
+      gesture_embedder_model_buffer: A valid flatbuffer *with* metadata loaded
+        from the TFLite gesture embedder model file.
+      canned_gesture_classifier_model_buffer: A valid flatbuffer *with* metadata
+        loaded from the TFLite canned gesture classifier model file.
+      custom_gesture_classifier_options: Custom gesture classifier options to
+        write custom gesture classifier metadata into the TFLite file.
+
+    Returns:
+      An MetadataWrite object.
+    """
+    writer = metadata_writer.MetadataWriter.create(
+        custom_gesture_classifier_options.model_buffer)
+    writer.add_general_info(_MODEL_NAME, _MODEL_DESCRIPTION)
+    writer.add_feature_input(name=_INPUT_NAME, description=_INPUT_DESCRIPTION)
+    writer.add_classification_output(
+        labels=custom_gesture_classifier_options.labels,
+        score_thresholding=custom_gesture_classifier_options.score_thresholding,
+        name=_OUTPUT_NAME,
+        description=_OUTPUT_DESCRIPTION)
+    return cls(hand_detector_model_buffer, hand_landmarks_detector_model_buffer,
+               gesture_embedder_model_buffer,
+               canned_gesture_classifier_model_buffer, writer)
+
+  def populate(self):
+    """Populates the metadata and creates model asset bundle.
+
+    Note that only the output model asset bundle is used for deployment.
+    The output JSON content is used to interpret the custom gesture classifier
+    metadata content.
+
+    Returns:
+      A tuple of (model_asset_bundle_in_bytes, metadata_json_content)
+    """
+    # Creates the model asset bundle for hand landmarker task.
+    landmark_models = {
+        _HAND_DETECTOR_TFLITE_NAME:
+            self._hand_detector_model_buffer,
+        _HAND_LANDMARKS_DETECTOR_TFLITE_NAME:
+            self._hand_landmarks_detector_model_buffer
+    }
+    output_hand_landmarker_path = os.path.join(self._temp_folder.name,
+                                               _HAND_LANDMARKER_BUNDLE_NAME)
+    writer_utils.create_model_asset_bundle(landmark_models,
+                                           output_hand_landmarker_path)
+
+    # Write metadata into custom gesture classifier model.
+    self._custom_gesture_classifier_model_buffer, custom_gesture_classifier_metadata_json = self._custom_gesture_classifier_metadata_writer.populate(
+    )
+    # Creates the model asset bundle for hand gesture recognizer sub graph.
+    hand_gesture_recognizer_models = {
+        _GESTURE_EMBEDDER_TFLITE_NAME:
+            self._gesture_embedder_model_buffer,
+        _CANNED_GESTURE_CLASSIFIER_TFLITE_NAME:
+            self._canned_gesture_classifier_model_buffer,
+        _CUSTOM_GESTURE_CLASSIFIER_TFLITE_NAME:
+            self._custom_gesture_classifier_model_buffer
+    }
+    output_hand_gesture_recognizer_path = os.path.join(
+        self._temp_folder.name, _HAND_GESTURE_RECOGNIZER_BUNDLE_NAME)
+    writer_utils.create_model_asset_bundle(hand_gesture_recognizer_models,
+                                           output_hand_gesture_recognizer_path)
+
+    # Creates the model asset bundle for end-to-end hand gesture recognizer
+    # graph.
+    gesture_recognizer_models = {
+        _HAND_LANDMARKER_BUNDLE_NAME:
+            read_file(output_hand_landmarker_path),
+        _HAND_GESTURE_RECOGNIZER_BUNDLE_NAME:
+            read_file(output_hand_gesture_recognizer_path),
+    }
+
+    output_file_path = os.path.join(self._temp_folder.name,
+                                    "gesture_recognizer.task")
+    writer_utils.create_model_asset_bundle(gesture_recognizer_models,
+                                           output_file_path)
+    with open(output_file_path, "rb") as f:
+      gesture_recognizer_model_buffer = f.read()
+    return gesture_recognizer_model_buffer, custom_gesture_classifier_metadata_json

mediapipe/model_maker/python/vision/gesture_recognizer/metadata_writer_test.py

@@ -0,0 +1,90 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+"""Tests for metadata_writer."""
+
+import os
+import zipfile
+
+import tensorflow as tf
+
+from mediapipe.model_maker.python.vision.gesture_recognizer import metadata_writer
+from mediapipe.tasks.python.metadata.metadata_writers import metadata_writer as base_metadata_writer
+from mediapipe.tasks.python.test import test_utils
+
+_TEST_DATA_DIR = "mediapipe/model_maker/python/vision/gesture_recognizer/test_data/metadata"
+
+_EXPECTED_JSON = test_utils.get_test_data_path(
+    os.path.join(_TEST_DATA_DIR, "custom_gesture_classifier_meta.json"))
+_CUSTOM_GESTURE_CLASSIFIER_PATH = test_utils.get_test_data_path(
+    os.path.join(_TEST_DATA_DIR, "custom_gesture_classifier.tflite"))
+
+
+class MetadataWriterTest(tf.test.TestCase):
+
+  def test_write_metadata_and_create_model_asset_bundle_successful(self):
+    # Use dummy model buffer for unit test only.
+    hand_detector_model_buffer = b"\x11\x12"
+    hand_landmarks_detector_model_buffer = b"\x22"
+    gesture_embedder_model_buffer = b"\x33"
+    canned_gesture_classifier_model_buffer = b"\x44"
+    custom_gesture_classifier_metadata_writer = metadata_writer.GestureClassifierOptions(
+        model_buffer=metadata_writer.read_file(_CUSTOM_GESTURE_CLASSIFIER_PATH),
+        labels=base_metadata_writer.Labels().add(
+            ["None", "Paper", "Rock", "Scissors"]),
+        score_thresholding=base_metadata_writer.ScoreThresholding(
+            global_score_threshold=0.5))
+    writer = metadata_writer.MetadataWriter.create(
+        hand_detector_model_buffer, hand_landmarks_detector_model_buffer,
+        gesture_embedder_model_buffer, canned_gesture_classifier_model_buffer,
+        custom_gesture_classifier_metadata_writer)
+    model_bundle_content, metadata_json = writer.populate()
+    with open(_EXPECTED_JSON, "r") as f:
+      expected_json = f.read()
+    self.assertEqual(metadata_json, expected_json)
+
+    # Checks the top-level model bundle can be extracted successfully.
+    model_bundle_filepath = os.path.join(self.get_temp_dir(),
+                                         "gesture_recognition.task")
+
+    with open(model_bundle_filepath, "wb") as f:
+      f.write(model_bundle_content)
+
+    with zipfile.ZipFile(model_bundle_filepath) as zf:
+      self.assertEqual(
+          set(zf.namelist()),
+          set(["hand_landmarker.task", "hand_gesture_recognizer.task"]))
+      zf.extractall(self.get_temp_dir())
+
+    # Checks the model bundles for sub-task can be extracted successfully.
+    hand_landmarker_bundle_filepath = os.path.join(self.get_temp_dir(),
+                                                   "hand_landmarker.task")
+    with zipfile.ZipFile(hand_landmarker_bundle_filepath) as zf:
+      self.assertEqual(
+          set(zf.namelist()),
+          set(["hand_landmarks_detector.tflite", "hand_detector.tflite"]))
+
+    hand_gesture_recognizer_bundle_filepath = os.path.join(
+        self.get_temp_dir(), "hand_gesture_recognizer.task")
+    with zipfile.ZipFile(hand_gesture_recognizer_bundle_filepath) as zf:
+      self.assertEqual(
+          set(zf.namelist()),
+          set([
+              "canned_gesture_classifier.tflite",
+              "custom_gesture_classifier.tflite", "gesture_embedder.tflite"
+          ]))
+
+
+if __name__ == "__main__":
+  tf.test.main()

mediapipe/model_maker/python/vision/gesture_recognizer/model_options.py

@@ -0,0 +1,27 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Configurable model options for gesture recognizer models."""
+
+import dataclasses
+
+
+@dataclasses.dataclass
+class GestureRecognizerModelOptions:
+  """Configurable options for gesture recognizer model.
+
+  Attributes:
+    dropout_rate: The fraction of the input units to drop, used in dropout
+      layer.
+  """
+  dropout_rate: float = 0.05

mediapipe/model_maker/python/vision/gesture_recognizer/test_data/metadata/custom_gesture_classifier_meta.json

@@ -0,0 +1,56 @@
+{
+  "name": "HandGestureRecognition",
+  "description": "Recognize the hand gesture in the image.",
+  "subgraph_metadata": [
+    {
+      "input_tensor_metadata": [
+        {
+          "name": "embedding",
+          "description": "Embedding feature vector from gesture embedder.",
+          "content": {
+            "content_properties_type": "FeatureProperties",
+            "content_properties": {
+            }
+          },
+          "stats": {
+          }
+        }
+      ],
+      "output_tensor_metadata": [
+        {
+          "name": "scores",
+          "description": "Hand gesture category scores.",
+          "content": {
+            "content_properties_type": "FeatureProperties",
+            "content_properties": {
+            }
+          },
+          "process_units": [
+            {
+              "options_type": "ScoreThresholdingOptions",
+              "options": {
+                "global_score_threshold": 0.5
+              }
+            }
+          ],
+          "stats": {
+            "max": [
+              1.0
+            ],
+            "min": [
+              0.0
+            ]
+          },
+          "associated_files": [
+            {
+              "name": "labels.txt",
+              "description": "Labels for categories that the model can recognize.",
+              "type": "TENSOR_AXIS_LABELS"
+            }
+          ]
+        }
+      ]
+    }
+  ],
+  "min_parser_version": "1.0.0"
+}

mediapipe/python/BUILD

@@ -121,6 +121,34 @@ py_library(
     ],
 )
 
+py_library(
+    name = "solution_base",
+    srcs = ["solution_base.py"],
+    srcs_version = "PY3",
+    visibility = [
+        "//mediapipe/python:__subpackages__",
+    ],
+    deps = [
+        ":_framework_bindings",
+        ":packet_creator",
+        ":packet_getter",
+        "//mediapipe/calculators/core:constant_side_packet_calculator_py_pb2",
+        "//mediapipe/calculators/image:image_transformation_calculator_py_pb2",
+        "//mediapipe/calculators/tensor:tensors_to_detections_calculator_py_pb2",
+        "//mediapipe/calculators/util:landmarks_smoothing_calculator_py_pb2",
+        "//mediapipe/calculators/util:logic_calculator_py_pb2",
+        "//mediapipe/calculators/util:thresholding_calculator_py_pb2",
+        "//mediapipe/framework:calculator_py_pb2",
+        "//mediapipe/framework/formats:classification_py_pb2",
+        "//mediapipe/framework/formats:detection_py_pb2",
+        "//mediapipe/framework/formats:landmark_py_pb2",
+        "//mediapipe/framework/formats:rect_py_pb2",
+        "//mediapipe/modules/objectron/calculators:annotation_py_pb2",
+        "//mediapipe/modules/objectron/calculators:lift_2d_frame_annotation_to_3d_calculator_py_pb2",
+        "@com_google_protobuf//:protobuf_python",
+    ],
+)
+
 py_test(
     name = "calculator_graph_test",
     srcs = ["calculator_graph_test.py"],
@@ -176,3 +204,24 @@ py_test(
         ":_framework_bindings",
     ],
 )
+
+py_test(
+    name = "solution_base_test",
+    srcs = ["solution_base_test.py"],
+    python_version = "PY3",
+    srcs_version = "PY3",
+    deps = [
+        ":solution_base",
+        "//file/google_src",
+        "//file/localfile",
+        "//mediapipe/calculators/core:gate_calculator",
+        "//mediapipe/calculators/core:pass_through_calculator",
+        "//mediapipe/calculators/core:side_packet_to_stream_calculator",
+        "//mediapipe/calculators/image:image_transformation_calculator",
+        "//mediapipe/calculators/util:detection_unique_id_calculator",
+        "//mediapipe/calculators/util:to_image_calculator",
+        "//mediapipe/framework:calculator_py_pb2",
+        "//mediapipe/framework/formats:detection_py_pb2",
+        "@com_google_protobuf//:protobuf_python",
+    ],
+)

mediapipe/python/solution_base.py

@@ -40,7 +40,6 @@ from mediapipe.calculators.util import landmarks_smoothing_calculator_pb2
 from mediapipe.calculators.util import logic_calculator_pb2
 from mediapipe.calculators.util import thresholding_calculator_pb2
 from mediapipe.framework import calculator_pb2
-from mediapipe.framework.formats import body_rig_pb2
 from mediapipe.framework.formats import classification_pb2
 from mediapipe.framework.formats import detection_pb2
 from mediapipe.framework.formats import landmark_pb2

mediapipe/python/solutions/BUILD

@@ -0,0 +1,105 @@
+# Copyright 2020 The MediaPipe Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+licenses(["notice"])
+
+package(default_visibility = ["//visibility:public"])
+
+py_library(
+    name = "hands",
+    srcs = [
+        "hands.py",
+        "hands_connections.py",
+    ],
+    data = [
+        "//mediapipe/modules/hand_landmark:hand_landmark_full.tflite",
+        "//mediapipe/modules/hand_landmark:hand_landmark_lite.tflite",
+        "//mediapipe/modules/hand_landmark:hand_landmark_tracking_cpu_graph",
+        "//mediapipe/modules/hand_landmark:handedness.txt",
+        "//mediapipe/modules/palm_detection:palm_detection_full.tflite",
+        "//mediapipe/modules/palm_detection:palm_detection_lite.tflite",
+    ],
+    srcs_version = "PY3",
+    deps = [
+        "//mediapipe/calculators/core:constant_side_packet_calculator_py_pb2",
+        "//mediapipe/calculators/core:gate_calculator_py_pb2",
+        "//mediapipe/calculators/core:split_vector_calculator_py_pb2",
+        "//mediapipe/calculators/tensor:image_to_tensor_calculator_py_pb2",
+        "//mediapipe/calculators/tensor:inference_calculator_py_pb2",
+        "//mediapipe/calculators/tensor:tensors_to_classification_calculator_py_pb2",
+        "//mediapipe/calculators/tensor:tensors_to_detections_calculator_py_pb2",
+        "//mediapipe/calculators/tensor:tensors_to_landmarks_calculator_py_pb2",
+        "//mediapipe/calculators/tflite:ssd_anchors_calculator_py_pb2",
+        "//mediapipe/calculators/util:association_calculator_py_pb2",
+        "//mediapipe/calculators/util:detections_to_rects_calculator_py_pb2",
+        "//mediapipe/calculators/util:logic_calculator_py_pb2",
+        "//mediapipe/calculators/util:non_max_suppression_calculator_py_pb2",
+        "//mediapipe/calculators/util:rect_transformation_calculator_py_pb2",
+        "//mediapipe/calculators/util:thresholding_calculator_py_pb2",
+        "//mediapipe/python:solution_base",
+    ],
+)
+
+py_library(
+    name = "drawing_styles",
+    srcs = ["drawing_styles.py"],
+    srcs_version = "PY3",
+    deps = [
+        "drawing_utils",
+        "face_mesh",
+        "hands",
+        "pose",
+    ],
+)
+
+py_library(
+    name = "drawing_utils",
+    srcs = ["drawing_utils.py"],
+    srcs_version = "PY3",
+    deps = [
+        "//mediapipe/framework/formats:detection_py_pb2",
+        "//mediapipe/framework/formats:landmark_py_pb2",
+        "//mediapipe/framework/formats:location_data_py_pb2",
+    ],
+)
+
+py_test(
+    name = "drawing_utils_test",
+    srcs = ["drawing_utils_test.py"],
+    python_version = "PY3",
+    srcs_version = "PY3",
+    deps = [
+        ":drawing_utils",
+        "//mediapipe/framework/formats:detection_py_pb2",
+        "//mediapipe/framework/formats:landmark_py_pb2",
+        "@com_google_protobuf//:protobuf_python",
+    ],
+)
+
+py_test(
+    name = "hands_test",
+    srcs = ["hands_test.py"],
+    data = [
+        ":testdata/asl_hand.25fps.mp4",
+        ":testdata/asl_hand.full.npz",
+        ":testdata/hands.jpg",
+    ],
+    python_version = "PY3",
+    srcs_version = "PY3",
+    deps = [
+        ":drawing_styles",
+        ":drawing_utils",
+        ":hands",
+    ],
+)

mediapipe/tasks/cc/audio/audio_classifier/proto/audio_classifier_graph_options.proto

@@ -21,6 +21,9 @@ import "mediapipe/framework/calculator.proto";
 import "mediapipe/tasks/cc/components/processors/proto/classifier_options.proto";
 import "mediapipe/tasks/cc/core/proto/base_options.proto";
 
+option java_package = "com.google.mediapipe.tasks.audio.audioclassifier.proto";
+option java_outer_classname = "AudioClassifierGraphOptionsProto";
+
 message AudioClassifierGraphOptions {
   extend mediapipe.CalculatorOptions {
     optional AudioClassifierGraphOptions ext = 451755788;

mediapipe/tasks/cc/components/containers/BUILD

@@ -21,15 +21,6 @@ cc_library(
     hdrs = ["rect.h"],
 )
 
-cc_library(
-    name = "gesture_recognition_result",
-    hdrs = ["gesture_recognition_result.h"],
-    deps = [
-        "//mediapipe/framework/formats:classification_cc_proto",
-        "//mediapipe/framework/formats:landmark_cc_proto",
-    ],
-)
-
 cc_library(
     name = "category",
     srcs = ["category.cc"],

mediapipe/tasks/cc/vision/gesture_recognizer/BUILD

@@ -124,12 +124,22 @@ cc_library(
     alwayslink = 1,
 )
 
+cc_library(
+    name = "gesture_recognizer_result",
+    hdrs = ["gesture_recognizer_result.h"],
+    deps = [
+        "//mediapipe/framework/formats:classification_cc_proto",
+        "//mediapipe/framework/formats:landmark_cc_proto",
+    ],
+)
+
 cc_library(
     name = "gesture_recognizer",
     srcs = ["gesture_recognizer.cc"],
     hdrs = ["gesture_recognizer.h"],
     deps = [
         ":gesture_recognizer_graph",
+        ":gesture_recognizer_result",
         ":hand_gesture_recognizer_graph",
         "//mediapipe/framework:packet",
         "//mediapipe/framework/api2:builder",
@@ -140,7 +150,6 @@ cc_library(
         "//mediapipe/framework/formats:rect_cc_proto",
         "//mediapipe/tasks/cc:common",
         "//mediapipe/tasks/cc/components:image_preprocessing",
-        "//mediapipe/tasks/cc/components/containers:gesture_recognition_result",
         "//mediapipe/tasks/cc/components/processors:classifier_options",
         "//mediapipe/tasks/cc/components/processors/proto:classifier_options_cc_proto",
         "//mediapipe/tasks/cc/core:base_options",

mediapipe/tasks/cc/vision/gesture_recognizer/gesture_recognizer.cc

@@ -58,8 +58,6 @@ namespace {
 using GestureRecognizerGraphOptionsProto = ::mediapipe::tasks::vision::
     gesture_recognizer::proto::GestureRecognizerGraphOptions;
 
-using ::mediapipe::tasks::components::containers::GestureRecognitionResult;
-
 constexpr char kHandGestureSubgraphTypeName[] =
     "mediapipe.tasks.vision.gesture_recognizer.GestureRecognizerGraph";
 
@@ -214,7 +212,7 @@ absl::StatusOr<std::unique_ptr<GestureRecognizer>> GestureRecognizer::Create(
       std::move(packets_callback));
 }
 
-absl::StatusOr<GestureRecognitionResult> GestureRecognizer::Recognize(
+absl::StatusOr<GestureRecognizerResult> GestureRecognizer::Recognize(
     mediapipe::Image image,
     std::optional<core::ImageProcessingOptions> image_processing_options) {
   if (image.UsesGpu()) {
@@ -250,7 +248,7 @@ absl::StatusOr<GestureRecognitionResult> GestureRecognizer::Recognize(
   };
 }
 
-absl::StatusOr<GestureRecognitionResult> GestureRecognizer::RecognizeForVideo(
+absl::StatusOr<GestureRecognizerResult> GestureRecognizer::RecognizeForVideo(
     mediapipe::Image image, int64 timestamp_ms,
     std::optional<core::ImageProcessingOptions> image_processing_options) {
   if (image.UsesGpu()) {

mediapipe/tasks/cc/vision/gesture_recognizer/gesture_recognizer.h

@@ -24,12 +24,12 @@ limitations under the License.
 #include "mediapipe/framework/formats/classification.pb.h"
 #include "mediapipe/framework/formats/image.h"
 #include "mediapipe/framework/formats/landmark.pb.h"
-#include "mediapipe/tasks/cc/components/containers/gesture_recognition_result.h"
 #include "mediapipe/tasks/cc/components/processors/classifier_options.h"
 #include "mediapipe/tasks/cc/core/base_options.h"
 #include "mediapipe/tasks/cc/vision/core/base_vision_task_api.h"
 #include "mediapipe/tasks/cc/vision/core/image_processing_options.h"
 #include "mediapipe/tasks/cc/vision/core/running_mode.h"
+#include "mediapipe/tasks/cc/vision/gesture_recognizer/gesture_recognizer_result.h"
 
 namespace mediapipe {
 namespace tasks {
@@ -81,9 +81,8 @@ struct GestureRecognizerOptions {
   // The user-defined result callback for processing live stream data.
   // The result callback should only be specified when the running mode is set
   // to RunningMode::LIVE_STREAM.
-  std::function<void(
+  std::function<void(absl::StatusOr<GestureRecognizerResult>, const Image&,
-      absl::StatusOr<components::containers::GestureRecognitionResult>,
+                     int64)>
-      const Image&, int64)>
       result_callback = nullptr;
 };
 
@@ -104,7 +103,7 @@ struct GestureRecognizerOptions {
 //       'width' and 'height' fields is NOT supported and will result in an
 //       invalid argument error being returned.
 // Outputs:
-//   GestureRecognitionResult
+//   GestureRecognizerResult
 //     - The hand gesture recognition results.
 class GestureRecognizer : tasks::vision::core::BaseVisionTaskApi {
  public:
@@ -139,7 +138,7 @@ class GestureRecognizer : tasks::vision::core::BaseVisionTaskApi {
   // The image can be of any size with format RGB or RGBA.
   // TODO: Describes how the input image will be preprocessed
   // after the yuv support is implemented.
-  absl::StatusOr<components::containers::GestureRecognitionResult> Recognize(
+  absl::StatusOr<GestureRecognizerResult> Recognize(
       Image image,
       std::optional<core::ImageProcessingOptions> image_processing_options =
           std::nullopt);
@@ -157,10 +156,10 @@ class GestureRecognizer : tasks::vision::core::BaseVisionTaskApi {
   // The image can be of any size with format RGB or RGBA. It's required to
   // provide the video frame's timestamp (in milliseconds). The input timestamps
   // must be monotonically increasing.
-  absl::StatusOr<components::containers::GestureRecognitionResult>
+  absl::StatusOr<GestureRecognizerResult> RecognizeForVideo(
-  RecognizeForVideo(Image image, int64 timestamp_ms,
+      Image image, int64 timestamp_ms,
-                    std::optional<core::ImageProcessingOptions>
+      std::optional<core::ImageProcessingOptions> image_processing_options =
-                        image_processing_options = std::nullopt);
+          std::nullopt);
 
   // Sends live image data to perform gesture recognition, and the results will
   // be available via the "result_callback" provided in the
@@ -179,7 +178,7 @@ class GestureRecognizer : tasks::vision::core::BaseVisionTaskApi {
   // and will result in an invalid argument error being returned.
   //
   // The "result_callback" provides
-  //   - A vector of GestureRecognitionResult, each is the recognized results
+  //   - A vector of GestureRecognizerResult, each is the recognized results
   //     for a input frame.
   //   - The const reference to the corresponding input image that the gesture
   //     recognizer runs on. Note that the const reference to the image will no

mediapipe/tasks/cc/vision/gesture_recognizer/gesture_recognizer_result.h

@@ -13,20 +13,20 @@ See the License for the specific language governing permissions and
 limitations under the License.
 ==============================================================================*/
 
-#ifndef MEDIAPIPE_TASKS_CC_COMPONENTS_CONTAINERS_GESTURE_RECOGNITION_RESULT_H_
+#ifndef MEDIAPIPE_TASKS_CC_VISION_GESTURE_RECOGNIZER_GESTURE_RECOGNIZER_RESULT_H_
-#define MEDIAPIPE_TASKS_CC_COMPONENTS_CONTAINERS_GESTURE_RECOGNITION_RESULT_H_
+#define MEDIAPIPE_TASKS_CC_VISION_GESTURE_RECOGNIZER_GESTURE_RECOGNIZER_RESULT_H_
 
 #include "mediapipe/framework/formats/classification.pb.h"
 #include "mediapipe/framework/formats/landmark.pb.h"
 
 namespace mediapipe {
 namespace tasks {
-namespace components {
+namespace vision {
-namespace containers {
+namespace gesture_recognizer {
 
 // The gesture recognition result from GestureRecognizer, where each vector
 // element represents a single hand detected in the image.
-struct GestureRecognitionResult {
+struct GestureRecognizerResult {
   // Recognized hand gestures with sorted order such that the winning label is
   // the first item in the list.
   std::vector<mediapipe::ClassificationList> gestures;
@@ -38,9 +38,9 @@ struct GestureRecognitionResult {
   std::vector<mediapipe::LandmarkList> hand_world_landmarks;
 };
 
-}  // namespace containers
+}  // namespace gesture_recognizer
-}  // namespace components
+}  // namespace vision
 }  // namespace tasks
 }  // namespace mediapipe
 
-#endif  // MEDIAPIPE_TASKS_CC_COMPONENTS_CONTAINERS_GESTURE_RECOGNITION_RESULT_H_
+#endif  // MEDIAPIPE_TASKS_CC_VISION_GESTURE_RECOGNIZER_GESTURE_RECOGNIZER_RESULT_H_

mediapipe/tasks/cc/vision/hand_landmarker/hand_landmarker_graph.cc

@@ -276,33 +276,44 @@ class HandLandmarkerGraph : public core::ModelTaskGraph {
         .set_min_size(max_num_hands);
     auto has_enough_hands = min_size_node.Out("").Cast<bool>();
 
-    auto image_for_hand_detector =
-        DisallowIf(image_in, has_enough_hands, graph);
-    auto norm_rect_in_for_hand_detector =
-        DisallowIf(norm_rect_in, has_enough_hands, graph);
-
     auto& hand_detector =
         graph.AddNode("mediapipe.tasks.vision.hand_detector.HandDetectorGraph");
     hand_detector.GetOptions<HandDetectorGraphOptions>().CopyFrom(
         tasks_options.hand_detector_graph_options());
+    auto& clip_hand_rects =
+        graph.AddNode("ClipNormalizedRectVectorSizeCalculator");
+    clip_hand_rects.GetOptions<ClipVectorSizeCalculatorOptions>()
+        .set_max_vec_size(max_num_hands);
+
+    if (tasks_options.base_options().use_stream_mode()) {
+      // While in stream mode, skip hand detector graph when we successfully
+      // track the hands from the last frame.
+      auto image_for_hand_detector =
+          DisallowIf(image_in, has_enough_hands, graph);
+      auto norm_rect_in_for_hand_detector =
+          DisallowIf(norm_rect_in, has_enough_hands, graph);
       image_for_hand_detector >> hand_detector.In("IMAGE");
       norm_rect_in_for_hand_detector >> hand_detector.In("NORM_RECT");
       auto hand_rects_from_hand_detector = hand_detector.Out("HAND_RECTS");
-
       auto& hand_association = graph.AddNode("HandAssociationCalculator");
       hand_association.GetOptions<HandAssociationCalculatorOptions>()
-        .set_min_similarity_threshold(tasks_options.min_tracking_confidence());
+          .set_min_similarity_threshold(
+              tasks_options.min_tracking_confidence());
       prev_hand_rects_from_landmarks >>
           hand_association[Input<std::vector<NormalizedRect>>::Multiple("")][0];
       hand_rects_from_hand_detector >>
           hand_association[Input<std::vector<NormalizedRect>>::Multiple("")][1];
       auto hand_rects = hand_association.Out("");
-
-    auto& clip_hand_rects =
-        graph.AddNode("ClipNormalizedRectVectorSizeCalculator");
-    clip_hand_rects.GetOptions<ClipVectorSizeCalculatorOptions>()
-        .set_max_vec_size(max_num_hands);
       hand_rects >> clip_hand_rects.In("");
+    } else {
+      // While not in stream mode, the input images are not guaranteed to be in
+      // series, and we don't want to enable the tracking and hand associations
+      // between input images. Always use the hand detector graph.
+      image_in >> hand_detector.In("IMAGE");
+      norm_rect_in >> hand_detector.In("NORM_RECT");
+      auto hand_rects_from_hand_detector = hand_detector.Out("HAND_RECTS");
+      hand_rects_from_hand_detector >> clip_hand_rects.In("");
+    }
     auto clipped_hand_rects = clip_hand_rects.Out("");
 
     auto& hand_landmarks_detector_graph = graph.AddNode(

mediapipe/tasks/java/com/google/mediapipe/tasks/audio/BUILD

@@ -0,0 +1,84 @@
+# Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+licenses(["notice"])
+
+package(default_visibility = ["//mediapipe/tasks:internal"])
+
+android_library(
+    name = "core",
+    srcs = glob(["core/*.java"]),
+    javacopts = [
+        "-Xep:AndroidJdkLibsChecker:OFF",
+    ],
+    deps = [
+        ":libmediapipe_tasks_audio_jni_lib",
+        "//mediapipe/java/com/google/mediapipe/framework:android_framework_no_mff",
+        "//mediapipe/tasks/java/com/google/mediapipe/tasks/components/containers:audiodata",
+        "//mediapipe/tasks/java/com/google/mediapipe/tasks/core",
+        "@maven//:com_google_guava_guava",
+    ],
+)
+
+# The native library of all MediaPipe audio tasks.
+cc_binary(
+    name = "libmediapipe_tasks_audio_jni.so",
+    linkshared = 1,
+    linkstatic = 1,
+    deps = [
+        "//mediapipe/java/com/google/mediapipe/framework/jni:mediapipe_framework_jni",
+        "//mediapipe/tasks/cc/audio/audio_classifier:audio_classifier_graph",
+        "//mediapipe/tasks/java/com/google/mediapipe/tasks/core/jni:model_resources_cache_jni",
+    ],
+)
+
+cc_library(
+    name = "libmediapipe_tasks_audio_jni_lib",
+    srcs = [":libmediapipe_tasks_audio_jni.so"],
+    alwayslink = 1,
+)
+
+android_library(
+    name = "audioclassifier",
+    srcs = [
+        "audioclassifier/AudioClassifier.java",
+        "audioclassifier/AudioClassifierResult.java",
+    ],
+    javacopts = [
+        "-Xep:AndroidJdkLibsChecker:OFF",
+    ],
+    manifest = "audioclassifier/AndroidManifest.xml",
+    deps = [
+        ":core",
+        "//mediapipe/framework:calculator_options_java_proto_lite",
+        "//mediapipe/java/com/google/mediapipe/framework:android_framework",
+        "//mediapipe/tasks/cc/audio/audio_classifier/proto:audio_classifier_graph_options_java_proto_lite",
+        "//mediapipe/tasks/cc/components/containers/proto:classifications_java_proto_lite",
+        "//mediapipe/tasks/cc/core/proto:base_options_java_proto_lite",
+        "//mediapipe/tasks/java/com/google/mediapipe/tasks/components/containers:audiodata",
+        "//mediapipe/tasks/java/com/google/mediapipe/tasks/components/containers:classificationresult",
+        "//mediapipe/tasks/java/com/google/mediapipe/tasks/components/processors:classifieroptions",
+        "//mediapipe/tasks/java/com/google/mediapipe/tasks/core",
+        "//third_party:autovalue",
+        "@maven//:com_google_guava_guava",
+    ],
+)
+
+load("//mediapipe/tasks/java/com/google/mediapipe/tasks:mediapipe_tasks_aar.bzl", "mediapipe_tasks_audio_aar")
+
+mediapipe_tasks_audio_aar(
+    name = "tasks_audio",
+    srcs = glob(["**/*.java"]),
+    native_library = ":libmediapipe_tasks_audio_jni_lib",
+)

mediapipe/tasks/java/com/google/mediapipe/tasks/audio/audioclassifier/AndroidManifest.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="utf-8"?>
+<manifest xmlns:android="http://schemas.android.com/apk/res/android"
+    package="com.google.mediapipe.tasks.audio.audioclassifier">
+
+    <uses-sdk android:minSdkVersion="24"
+        android:targetSdkVersion="30" />
+
+</manifest>

mediapipe/tasks/java/com/google/mediapipe/tasks/audio/audioclassifier/AudioClassifier.java

@@ -0,0 +1,399 @@
+// Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+package com.google.mediapipe.tasks.audio.audioclassifier;
+
+import android.content.Context;
+import android.os.ParcelFileDescriptor;
+import com.google.auto.value.AutoValue;
+import com.google.mediapipe.proto.CalculatorOptionsProto.CalculatorOptions;
+import com.google.mediapipe.framework.MediaPipeException;
+import com.google.mediapipe.framework.Packet;
+import com.google.mediapipe.framework.PacketGetter;
+import com.google.mediapipe.framework.ProtoUtil;
+import com.google.mediapipe.tasks.audio.audioclassifier.proto.AudioClassifierGraphOptionsProto;
+import com.google.mediapipe.tasks.audio.core.BaseAudioTaskApi;
+import com.google.mediapipe.tasks.audio.core.RunningMode;
+import com.google.mediapipe.tasks.components.containers.AudioData;
+import com.google.mediapipe.tasks.components.containers.proto.ClassificationsProto;
+import com.google.mediapipe.tasks.components.processors.ClassifierOptions;
+import com.google.mediapipe.tasks.core.BaseOptions;
+import com.google.mediapipe.tasks.core.ErrorListener;
+import com.google.mediapipe.tasks.core.OutputHandler;
+import com.google.mediapipe.tasks.core.OutputHandler.PureResultListener;
+import com.google.mediapipe.tasks.core.OutputHandler.ResultListener;
+import com.google.mediapipe.tasks.core.TaskInfo;
+import com.google.mediapipe.tasks.core.TaskOptions;
+import com.google.mediapipe.tasks.core.TaskRunner;
+import com.google.mediapipe.tasks.core.proto.BaseOptionsProto;
+import java.io.File;
+import java.io.IOException;
+import java.nio.ByteBuffer;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+import java.util.Optional;
+
+/**
+ * Performs audio classification on audio clips or audio stream.
+ *
+ * <p>This API expects a TFLite model with mandatory TFLite Model Metadata that contains the
+ * mandatory AudioProperties of the solo input audio tensor and the optional (but recommended) label
+ * items as AssociatedFiles with type TENSOR_AXIS_LABELS per output classification tensor.
+ *
+ * <p>Input tensor: (kTfLiteFloat32)
+ *
+ * <ul>
+ *   <li>input audio buffer of size `[batch * samples]`.
+ *   <li>batch inference is not supported (`batch` is required to be 1).
+ *   <li>for multi-channel models, the channels need be interleaved.
+ * </ul>
+ *
+ * <p>At least one output tensor with: (kTfLiteFloat32)
+ *
+ * <ul>
+ *   <li>`[1 x N]` array with `N` represents the number of categories.
+ *   <li>optional (but recommended) label items as AssociatedFiles with type TENSOR_AXIS_LABELS,
+ *       containing one label per line. The first such AssociatedFile (if any) is used to fill the
+ *       `category_name` field of the results. The `display_name` field is filled from the
+ *       AssociatedFile (if any) whose locale matches the `display_names_locale` field of the
+ *       `AudioClassifierOptions` used at creation time ("en" by default, i.e. English). If none of
+ *       these are available, only the `index` field of the results will be filled.
+ * </ul>
+ */
+public final class AudioClassifier extends BaseAudioTaskApi {
+  private static final String TAG = AudioClassifier.class.getSimpleName();
+  private static final String AUDIO_IN_STREAM_NAME = "audio_in";
+  private static final String SAMPLE_RATE_IN_STREAM_NAME = "sample_rate_in";
+  private static final List<String> INPUT_STREAMS =
+      Collections.unmodifiableList(
+          Arrays.asList(
+              "AUDIO:" + AUDIO_IN_STREAM_NAME, "SAMPLE_RATE:" + SAMPLE_RATE_IN_STREAM_NAME));
+  private static final List<String> OUTPUT_STREAMS =
+      Collections.unmodifiableList(
+          Arrays.asList(
+              "CLASSIFICATIONS:classifications_out",
+              "TIMESTAMPED_CLASSIFICATIONS:timestamped_classifications_out"));
+  private static final int CLASSIFICATIONS_OUT_STREAM_INDEX = 0;
+  private static final int TIMESTAMPED_CLASSIFICATIONS_OUT_STREAM_INDEX = 1;
+  private static final String TASK_GRAPH_NAME =
+      "mediapipe.tasks.audio.audio_classifier.AudioClassifierGraph";
+  private static final long MICROSECONDS_PER_MILLISECOND = 1000;
+
+  static {
+    ProtoUtil.registerTypeName(
+        ClassificationsProto.ClassificationResult.class,
+        "mediapipe.tasks.components.containers.proto.ClassificationResult");
+  }
+
+  /**
+   * Creates an {@link AudioClassifier} instance from a model file and default {@link
+   * AudioClassifierOptions}.
+   *
+   * @param context an Android {@link Context}.
+   * @param modelPath path to the classification model in the assets.
+   * @throws MediaPipeException if there is an error during {@link AudioClassifier} creation.
+   */
+  public static AudioClassifier createFromFile(Context context, String modelPath) {
+    BaseOptions baseOptions = BaseOptions.builder().setModelAssetPath(modelPath).build();
+    return createFromOptions(
+        context, AudioClassifierOptions.builder().setBaseOptions(baseOptions).build());
+  }
+
+  /**
+   * Creates an {@link AudioClassifier} instance from a model file and default {@link
+   * AudioClassifierOptions}.
+   *
+   * @param context an Android {@link Context}.
+   * @param modelFile the classification model {@link File} instance.
+   * @throws IOException if an I/O error occurs when opening the tflite model file.
+   * @throws MediaPipeException if there is an error during {@link AudioClassifier} creation.
+   */
+  public static AudioClassifier createFromFile(Context context, File modelFile) throws IOException {
+    try (ParcelFileDescriptor descriptor =
+        ParcelFileDescriptor.open(modelFile, ParcelFileDescriptor.MODE_READ_ONLY)) {
+      BaseOptions baseOptions =
+          BaseOptions.builder().setModelAssetFileDescriptor(descriptor.getFd()).build();
+      return createFromOptions(
+          context, AudioClassifierOptions.builder().setBaseOptions(baseOptions).build());
+    }
+  }
+
+  /**
+   * Creates an {@link AudioClassifier} instance from a model buffer and default {@link
+   * AudioClassifierOptions}.
+   *
+   * @param context an Android {@link Context}.
+   * @param modelBuffer a direct {@link ByteBuffer} or a {@link MappedByteBuffer} of the
+   *     classification model.
+   * @throws MediaPipeException if there is an error during {@link AudioClassifier} creation.
+   */
+  public static AudioClassifier createFromBuffer(Context context, final ByteBuffer modelBuffer) {
+    BaseOptions baseOptions = BaseOptions.builder().setModelAssetBuffer(modelBuffer).build();
+    return createFromOptions(
+        context, AudioClassifierOptions.builder().setBaseOptions(baseOptions).build());
+  }
+
+  /**
+   * Creates an {@link AudioClassifier} instance from an {@link AudioClassifierOptions} instance.
+   *
+   * @param context an Android {@link Context}.
+   * @param options an {@link AudioClassifierOptions} instance.
+   * @throws MediaPipeException if there is an error during {@link AudioClassifier} creation.
+   */
+  public static AudioClassifier createFromOptions(Context context, AudioClassifierOptions options) {
+    OutputHandler<AudioClassifierResult, Void> handler = new OutputHandler<>();
+    handler.setOutputPacketConverter(
+        new OutputHandler.OutputPacketConverter<AudioClassifierResult, Void>() {
+          @Override
+          public AudioClassifierResult convertToTaskResult(List<Packet> packets) {
+            try {
+              if (!packets.get(CLASSIFICATIONS_OUT_STREAM_INDEX).isEmpty()) {
+                // For audio stream mode.
+                return AudioClassifierResult.createFromProto(
+                    PacketGetter.getProto(
+                        packets.get(CLASSIFICATIONS_OUT_STREAM_INDEX),
+                        ClassificationsProto.ClassificationResult.getDefaultInstance()),
+                    packets.get(CLASSIFICATIONS_OUT_STREAM_INDEX).getTimestamp()
+                        / MICROSECONDS_PER_MILLISECOND);
+              } else {
+                // For audio clips mode.
+                return AudioClassifierResult.createFromProtoList(
+                    PacketGetter.getProtoVector(
+                        packets.get(TIMESTAMPED_CLASSIFICATIONS_OUT_STREAM_INDEX),
+                        ClassificationsProto.ClassificationResult.parser()),
+                    -1);
+              }
+            } catch (IOException e) {
+              throw new MediaPipeException(
+                  MediaPipeException.StatusCode.INTERNAL.ordinal(), e.getMessage());
+            }
+          }
+
+          @Override
+          public Void convertToTaskInput(List<Packet> packets) {
+            return null;
+          }
+        });
+    if (options.resultListener().isPresent()) {
+      ResultListener<AudioClassifierResult, Void> resultListener =
+          new ResultListener<AudioClassifierResult, Void>() {
+            @Override
+            public void run(AudioClassifierResult audioClassifierResult, Void input) {
+              options.resultListener().get().run(audioClassifierResult);
+            }
+          };
+      handler.setResultListener(resultListener);
+    }
+    options.errorListener().ifPresent(handler::setErrorListener);
+    // Audio tasks should not drop input audio due to flow limiting, which may cause data
+    // inconsistency.
+    TaskRunner runner =
+        TaskRunner.create(
+            context,
+            TaskInfo.<AudioClassifierOptions>builder()
+                .setTaskGraphName(TASK_GRAPH_NAME)
+                .setInputStreams(INPUT_STREAMS)
+                .setOutputStreams(OUTPUT_STREAMS)
+                .setTaskOptions(options)
+                .setEnableFlowLimiting(false)
+                .build(),
+            handler);
+    return new AudioClassifier(runner, options.runningMode());
+  }
+
+  /**
+   * Constructor to initialize an {@link AudioClassifier} from a {@link TaskRunner} and {@link
+   * RunningMode}.
+   *
+   * @param taskRunner a {@link TaskRunner}.
+   * @param runningMode a mediapipe audio task {@link RunningMode}.
+   */
+  private AudioClassifier(TaskRunner taskRunner, RunningMode runningMode) {
+    super(taskRunner, runningMode, AUDIO_IN_STREAM_NAME, SAMPLE_RATE_IN_STREAM_NAME);
+  }
+
+  /*
+   * Performs audio classification on the provided audio clip. Only use this method when the
+   * AudioClassifier is created with the audio clips mode.
+   *
+   * <p>The audio clip is represented as a MediaPipe {@link AudioData} object  The method accepts
+   * audio clips with various length and audio sample rate.  It's required to provide the
+   * corresponding audio sample rate within the {@link AudioData} object.
+   *
+   * <p>The input audio clip may be longer than what the model is able to process in a single
+   * inference. When this occurs, the input audio clip is split into multiple chunks starting at
+   * different timestamps. For this reason, this function returns a vector of ClassificationResult
+   * objects, each associated with a timestamp corresponding to the start (in milliseconds) of the
+   * chunk data that was classified, e.g:
+   *
+   * ClassificationResult #0 (first chunk of data):
+   *  timestamp_ms: 0 (starts at 0ms)
+   *  classifications #0 (single head model):
+   *   category #0:
+   *    category_name: "Speech"
+   *    score: 0.6
+   *   category #1:
+   *    category_name: "Music"
+   *    score: 0.2
+   * ClassificationResult #1 (second chunk of data):
+   *  timestamp_ms: 800 (starts at 800ms)
+   *  classifications #0 (single head model):
+   *   category #0:
+   *    category_name: "Speech"
+   *    score: 0.5
+   *   category #1:
+   *    category_name: "Silence"
+   *    score: 0.1
+   *
+   * @param audioClip a MediaPipe {@link AudioData} object for processing.
+   * @throws MediaPipeException if there is an internal error.
+   */
+  public AudioClassifierResult classify(AudioData audioClip) {
+    return (AudioClassifierResult) processAudioClip(audioClip);
+  }
+
+  /*
+   * Sends audio data (a block in a continuous audio stream) to perform audio classification. Only
+   * use this method when the AudioClassifier is created with the audio stream mode.
+   *
+   * <p>The audio block is represented as a MediaPipe {@link AudioData} object. The audio data will
+   * be resampled, accumulated, and framed to the proper size for the underlying model to consume.
+   * It's required to provide the corresponding audio sample rate within {@link AudioData} object as
+   * well as a timestamp (in milliseconds) to indicate the start time of the input audio block. The
+   * timestamps must be monotonically increasing. This method will return immediately after
+   * the input audio data is accepted. The results will be available in the `resultListener`
+   * provided in the `AudioClassifierOptions`. The `classifyAsync` method is designed to process
+   * auido stream data such as microphone input.
+   *
+   * <p>The input audio block may be longer than what the model is able to process in a single
+   * inference. When this occurs, the input audio block is split into multiple chunks. For this
+   * reason, the callback may be called multiple times (once per chunk) for each call to this
+   * function.
+   *
+   * @param audioBlock a MediaPipe {@link AudioData} object for processing.
+   * @param timestampMs the input timestamp (in milliseconds).
+   * @throws MediaPipeException if there is an internal error.
+   */
+  public void classifyAsync(AudioData audioBlock, long timestampMs) {
+    checkOrSetSampleRate(audioBlock.getFormat().getSampleRate());
+    sendAudioStreamData(audioBlock, timestampMs);
+  }
+
+  /** Options for setting up and {@link AudioClassifier}. */
+  @AutoValue
+  public abstract static class AudioClassifierOptions extends TaskOptions {
+
+    /** Builder for {@link AudioClassifierOptions}. */
+    @AutoValue.Builder
+    public abstract static class Builder {
+      /** Sets the {@link BaseOptions} for the audio classifier task. */
+      public abstract Builder setBaseOptions(BaseOptions baseOptions);
+
+      /**
+       * Sets the {@link RunningMode} for the audio classifier task. Default to the audio clips
+       * mode. Image classifier has two modes:
+       *
+       * <ul>
+       *   <li>AUDIO_CLIPS: The mode for running audio classification on audio clips. Users feed
+       *       audio clips to the `classify` method, and will receive the classification results as
+       *       the return value.
+       *   <li>AUDIO_STREAM: The mode for running audio classification on the audio stream, such as
+       *       from microphone. Users call `classifyAsync` to push the audio data into the
+       *       AudioClassifier, the classification results will be available in the result callback
+       *       when the audio classifier finishes the work.
+       * </ul>
+       */
+      public abstract Builder setRunningMode(RunningMode runningMode);
+
+      /**
+       * Sets the optional {@link ClassifierOptions} controling classification behavior, such as
+       * score threshold, number of results, etc.
+       */
+      public abstract Builder setClassifierOptions(ClassifierOptions classifierOptions);
+
+      /**
+       * Sets the {@link ResultListener} to receive the classification results asynchronously when
+       * the audio classifier is in the audio stream mode.
+       */
+      public abstract Builder setResultListener(
+          PureResultListener<AudioClassifierResult> resultListener);
+
+      /** Sets an optional {@link ErrorListener}. */
+      public abstract Builder setErrorListener(ErrorListener errorListener);
+
+      abstract AudioClassifierOptions autoBuild();
+
+      /**
+       * Validates and builds the {@link AudioClassifierOptions} instance.
+       *
+       * @throws IllegalArgumentException if the result listener and the running mode are not
+       *     properly configured. The result listener should only be set when the audio classifier
+       *     is in the audio stream mode.
+       */
+      public final AudioClassifierOptions build() {
+        AudioClassifierOptions options = autoBuild();
+        if (options.runningMode() == RunningMode.AUDIO_STREAM) {
+          if (!options.resultListener().isPresent()) {
+            throw new IllegalArgumentException(
+                "The audio classifier is in the audio stream mode, a user-defined result listener"
+                    + " must be provided in the AudioClassifierOptions.");
+          }
+        } else if (options.resultListener().isPresent()) {
+          throw new IllegalArgumentException(
+              "The audio classifier is in the audio clips mode, a user-defined result listener"
+                  + " shouldn't be provided in AudioClassifierOptions.");
+        }
+        return options;
+      }
+    }
+
+    abstract BaseOptions baseOptions();
+
+    abstract RunningMode runningMode();
+
+    abstract Optional<ClassifierOptions> classifierOptions();
+
+    abstract Optional<PureResultListener<AudioClassifierResult>> resultListener();
+
+    abstract Optional<ErrorListener> errorListener();
+
+    public static Builder builder() {
+      return new AutoValue_AudioClassifier_AudioClassifierOptions.Builder()
+          .setRunningMode(RunningMode.AUDIO_CLIPS);
+    }
+
+    /**
+     * Converts a {@link AudioClassifierOptions} to a {@link CalculatorOptions} protobuf message.
+     */
+    @Override
+    public CalculatorOptions convertToCalculatorOptionsProto() {
+      BaseOptionsProto.BaseOptions.Builder baseOptionsBuilder =
+          BaseOptionsProto.BaseOptions.newBuilder();
+      baseOptionsBuilder.setUseStreamMode(runningMode() == RunningMode.AUDIO_STREAM);
+      baseOptionsBuilder.mergeFrom(convertBaseOptionsToProto(baseOptions()));
+      AudioClassifierGraphOptionsProto.AudioClassifierGraphOptions.Builder taskOptionsBuilder =
+          AudioClassifierGraphOptionsProto.AudioClassifierGraphOptions.newBuilder()
+              .setBaseOptions(baseOptionsBuilder);
+      if (classifierOptions().isPresent()) {
+        taskOptionsBuilder.setClassifierOptions(classifierOptions().get().convertToProto());
+      }
+      return CalculatorOptions.newBuilder()
+          .setExtension(
+              AudioClassifierGraphOptionsProto.AudioClassifierGraphOptions.ext,
+              taskOptionsBuilder.build())
+          .build();
+    }
+  }
+}

mediapipe/tasks/java/com/google/mediapipe/tasks/audio/audioclassifier/AudioClassifierResult.java

@@ -0,0 +1,76 @@
+// Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+package com.google.mediapipe.tasks.audio.audioclassifier;
+
+import com.google.auto.value.AutoValue;
+import com.google.mediapipe.tasks.components.containers.ClassificationResult;
+import com.google.mediapipe.tasks.components.containers.proto.ClassificationsProto;
+import com.google.mediapipe.tasks.core.TaskResult;
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Optional;
+
+/** Represents the classification results generated by {@link AudioClassifier}. */
+@AutoValue
+public abstract class AudioClassifierResult implements TaskResult {
+
+  /**
+   * Creates an {@link AudioClassifierResult} instance from a list of {@link
+   * ClassificationsProto.ClassificationResult} protobuf messages.
+   *
+   * @param protoList a list of {@link ClassificationsProto.ClassificationResult} protobuf message
+   *     to convert.
+   * @param timestampMs a timestamp for this result.
+   */
+  static AudioClassifierResult createFromProtoList(
+      List<ClassificationsProto.ClassificationResult> protoList, long timestampMs) {
+    List<ClassificationResult> classificationResultList = new ArrayList<>();
+    for (ClassificationsProto.ClassificationResult proto : protoList) {
+      classificationResultList.add(ClassificationResult.createFromProto(proto));
+    }
+    return new AutoValue_AudioClassifierResult(
+        Optional.of(classificationResultList), Optional.empty(), timestampMs);
+  }
+
+  /**
+   * Creates an {@link AudioClassifierResult} instance from a {@link
+   * ClassificationsProto.ClassificationResult} protobuf message.
+   *
+   * @param proto the {@link ClassificationsProto.ClassificationResult} protobuf message to convert.
+   * @param timestampMs a timestamp for this result.
+   */
+  static AudioClassifierResult createFromProto(
+      ClassificationsProto.ClassificationResult proto, long timestampMs) {
+    return new AutoValue_AudioClassifierResult(
+        Optional.empty(), Optional.of(ClassificationResult.createFromProto(proto)), timestampMs);
+  }
+
+  /**
+   * A list of of timpstamed {@link ClassificationResult} objects, each contains one set of results
+   * per classifier head. The list represents the audio classification result of an audio clip, and
+   * is only available when running with the audio clips mode.
+   */
+  public abstract Optional<List<ClassificationResult>> classificationResultList();
+
+  /**
+   * Contains one set of results per classifier head. A {@link ClassificationResult} usually
+   * represents one audio classification result in an audio stream, and s only available when
+   * running with the audio stream mode.
+   */
+  public abstract Optional<ClassificationResult> classificationResult();
+
+  @Override
+  public abstract long timestampMs();
+}

mediapipe/tasks/java/com/google/mediapipe/tasks/audio/core/BaseAudioTaskApi.java

@@ -0,0 +1,151 @@
+// Copyright 2022 The MediaPipe Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+package com.google.mediapipe.tasks.audio.core;
+
+import com.google.mediapipe.framework.MediaPipeException;
+import com.google.mediapipe.framework.Packet;
+import com.google.mediapipe.tasks.components.containers.AudioData;
+import com.google.mediapipe.tasks.core.TaskResult;
+import com.google.mediapipe.tasks.core.TaskRunner;
+import java.util.HashMap;
+import java.util.Map;
+
+/** The base class of MediaPipe audio tasks. */
+public class BaseAudioTaskApi implements AutoCloseable {
+  private static final long MICROSECONDS_PER_MILLISECOND = 1000;
+  private static final long PRESTREAM_TIMESTAMP = Long.MIN_VALUE + 2;
+
+  private final TaskRunner runner;
+  private final RunningMode runningMode;
+  private final String audioStreamName;
+  private final String sampleRateStreamName;
+  private double defaultSampleRate;
+
+  static {
+    System.loadLibrary("mediapipe_tasks_audio_jni");
+  }
+
+  /**
+   * Constructor to initialize a {@link BaseAudioTaskApi}.
+   *
+   * @param runner a {@link TaskRunner}.
+   * @param runningMode a mediapipe audio task {@link RunningMode}.
+   * @param audioStreamName the name of the input audio stream.
+   * @param sampleRateStreamName the name of the audio sample rate stream.
+   */
+  public BaseAudioTaskApi(
+      TaskRunner runner,
+      RunningMode runningMode,
+      String audioStreamName,
+      String sampleRateStreamName) {
+    this.runner = runner;
+    this.runningMode = runningMode;
+    this.audioStreamName = audioStreamName;
+    this.sampleRateStreamName = sampleRateStreamName;
+    this.defaultSampleRate = -1.0;
+  }
+
+  /**
+   * A synchronous method to process audio clips. The call blocks the current thread until a failure
+   * status or a successful result is returned.
+   *
+   * @param audioClip a MediaPipe {@link AudioDatra} object for processing.
+   * @throws MediaPipeException if the task is not in the audio clips mode.
+   */
+  protected TaskResult processAudioClip(AudioData audioClip) {
+    if (runningMode != RunningMode.AUDIO_CLIPS) {
+      throw new MediaPipeException(
+          MediaPipeException.StatusCode.FAILED_PRECONDITION.ordinal(),
+          "Task is not initialized with the audio clips mode. Current running mode:"
+              + runningMode.name());
+    }
+    Map<String, Packet> inputPackets = new HashMap<>();
+    inputPackets.put(
+        audioStreamName,
+        runner
+            .getPacketCreator()
+            .createMatrix(
+                audioClip.getFormat().getNumOfChannels(),
+                audioClip.getBufferLength(),
+                audioClip.getBuffer()));
+    inputPackets.put(
+        sampleRateStreamName,
+        runner.getPacketCreator().createFloat64(audioClip.getFormat().getSampleRate()));
+    return runner.process(inputPackets);
+  }
+
+  /**
+   * Checks or sets the audio sample rate in the audio stream mode.
+   *
+   * @param sampleRate the audio sample rate.
+   * @throws MediaPipeException if the task is not in the audio stream mode or the provided sample
+   *     rate is inconsisent with the previously recevied.
+   */
+  protected void checkOrSetSampleRate(double sampleRate) {
+    if (runningMode != RunningMode.AUDIO_STREAM) {
+      throw new MediaPipeException(
+          MediaPipeException.StatusCode.FAILED_PRECONDITION.ordinal(),
+          "Task is not initialized with the audio stream mode. Current running mode:"
+              + runningMode.name());
+    }
+    if (defaultSampleRate > 0) {
+      if (Double.compare(sampleRate, defaultSampleRate) != 0) {
+        throw new MediaPipeException(
+            MediaPipeException.StatusCode.INVALID_ARGUMENT.ordinal(),
+            "The input audio sample rate: "
+                + sampleRate
+                + " is inconsistent with the previously provided: "
+                + defaultSampleRate);
+      }
+    } else {
+      Map<String, Packet> inputPackets = new HashMap<>();
+      inputPackets.put(sampleRateStreamName, runner.getPacketCreator().createFloat64(sampleRate));
+      runner.send(inputPackets, PRESTREAM_TIMESTAMP);
+      defaultSampleRate = sampleRate;
+    }
+  }
+  /**
+   * An asynchronous method to send audio stream data to the {@link TaskRunner}. The results will be
+   * available in the user-defined result listener.
+   *
+   * @param audioClip a MediaPipe {@link AudioDatra} object for processing.
+   * @param timestampMs the corresponding timestamp of the input image in milliseconds.
+   * @throws MediaPipeException if the task is not in the stream mode.
+   */
+  protected void sendAudioStreamData(AudioData audioClip, long timestampMs) {
+    if (runningMode != RunningMode.AUDIO_STREAM) {
+      throw new MediaPipeException(
+          MediaPipeException.StatusCode.FAILED_PRECONDITION.ordinal(),
+          "Task is not initialized with the audio stream mode. Current running mode:"
+              + runningMode.name());
+    }
+    Map<String, Packet> inputPackets = new HashMap<>();
+    inputPackets.put(
+        audioStreamName,
+        runner
+            .getPacketCreator()
+            .createMatrix(
+                audioClip.getFormat().getNumOfChannels(),
+                audioClip.getBufferLength(),
+                audioClip.getBuffer()));
+    runner.send(inputPackets, timestampMs * MICROSECONDS_PER_MILLISECOND);
+  }
+
+  /** Closes and cleans up the MediaPipe audio task. */
+  @Override
+  public void close() {
+    runner.close();
+  }
+}