def testConfusionMatrix(self):
spec = audio_spec.BrowserFFTSpec()
temp_folder = self.get_temp_dir()
cat1 = write_sample(temp_folder, 'cat', '1.wav', 44100, duration_sec=1)
cat2 = write_sample(temp_folder, 'cat', '2.wav', 44100, duration_sec=2)
dog1 = write_sample(temp_folder, 'dog', '1.wav', 44100, duration_sec=3)
dog2 = write_sample(temp_folder, 'dog', '2.wav', 44100, duration_sec=4)
index_to_labels = ['cat', 'dog']
# Prepare data.
ds = tf.data.Dataset.from_tensor_slices(([cat1, cat2, dog1,
dog2], [0, 0, 1, 1]))
data_loader = audio_dataloader.DataLoader(ds, len(ds), index_to_labels,
spec)
# Train a floating point model.
task = audio_classifier.create(data_loader,
spec,
batch_size=1,
epochs=15)
confusion_matrx = task.confusion_matrix(data_loader)
# BrowserFFTSpec generates 1 sample for 1 second audio so there are
# 10 samples in total.
self.assertEqual(tf.math.reduce_sum(confusion_matrx), 10)
# confusion_matrix is of shape (truth, predication)
# We have 2 classes, 3 cat samples and 7 dog samples.
self.assertEqual(confusion_matrx.shape, (2, 2))
self.assertAllEqual(
tf.math.reduce_sum(confusion_matrx, axis=-1).numpy(),
np.array([3, 7]))
def test_dynamic_range_quantization(self):
self._train_and_export(
audio_spec.BrowserFFTSpec(),
num_classes=2,
filename='binary_classification.tflite',
expected_model_size=1 * 1000 * 1000,
quantization_config=configs.QuantizationConfig.for_dynamic())
def testBrowserFFT(self):
def pcm(shape):
# Convert random number between (0, 1] to int16
return np.random.rand(*shape) * (1 << 15)
np.random.seed(123)
spec = audio_spec.BrowserFFTSpec()
dataset_shape = (1, spec.expected_waveform_len)
sounds = [pcm(dataset_shape) for category in range(2)]
labels = list(range(2))
index_to_labels = ['sound1', 'sound2']
ds = tf.data.Dataset.zip((tf.data.Dataset.from_tensor_slices(sounds),
tf.data.Dataset.from_tensor_slices(labels)))
ds = ds.map(spec.preprocess)
data_loader = audio_dataloader.DataLoader(ds, len(ds), index_to_labels)
task = audio_classifier.create(data_loader,
spec,
batch_size=1,
epochs=100)
_, acc = task.evaluate(data_loader)
# Better than random guessing.
self.assertGreater(acc, .5)
# Export the model to saved model.
saved_model_output_path = os.path.join(spec.model_dir, 'saved_model')
task.export(spec.model_dir, export_format=ExportFormat.SAVED_MODEL)
self.assertTrue(os.path.isdir(saved_model_output_path))
self.assertNotEqual(len(os.listdir(saved_model_output_path)), 0)
def test_spec(self):
folder_path = self._get_folder_path('test_examples_helper')
write_sample(folder_path, 'unknown', '2s.wav', 44100, 2, value=1)
spec = audio_spec.YAMNetSpec()
audio_dataloader.DataLoader.from_folder(spec, folder_path)
spec = audio_spec.BrowserFFTSpec()
audio_dataloader.DataLoader.from_folder(spec, folder_path)
def test_spec(self):
folder_path = self._get_folder_path('test_examples_helper')
spec = audio_spec.YAMNetSpec()
audio_dataloader.DataLoader.from_esc50(spec, folder_path)
spec = audio_spec.BrowserFFTSpec()
with self.assertRaises(AssertionError):
audio_dataloader.DataLoader.from_esc50(spec, folder_path)
def testBrowserFFT(self):
def pcm(shape):
# Convert random number between (0, 1] to int16
return np.random.rand(*shape) * (1 << 15)
np.random.seed(123)
tf.random.set_seed(123)
# Prepare data.
spec = audio_spec.BrowserFFTSpec()
dataset_shape = (1, spec.expected_waveform_len)
sounds = [pcm(dataset_shape) for category in range(2)]
labels = list(range(2))
index_to_labels = ['sound1', 'sound2']
ds = tf.data.Dataset.zip((tf.data.Dataset.from_tensor_slices(sounds),
tf.data.Dataset.from_tensor_slices(labels)))
preprocessed_ds = ds.map(spec.preprocess)
data_loader = audio_dataloader.DataLoader(preprocessed_ds,
len(preprocessed_ds),
index_to_labels)
# Train a floating point model.
task = audio_classifier.create(data_loader, spec, batch_size=1, epochs=50)
# Evaluate trained model
_, acc = task.evaluate(data_loader)
# Better than random guessing.
self.assertGreater(acc, .5)
# Export the model to saved model.
output_path = os.path.join(spec.model_dir, 'saved_model')
task.export(spec.model_dir, export_format=ExportFormat.SAVED_MODEL)
self.assertTrue(os.path.isdir(output_path))
self.assertNotEqual(len(os.listdir(output_path)), 0)
# Export the model to TFLite.
output_path = os.path.join(spec.model_dir, 'float.tflite')
task.export(
spec.model_dir,
tflite_filename='float.tflite',
export_format=ExportFormat.TFLITE)
self.assertTrue(tf.io.gfile.exists(output_path))
self.assertGreater(os.path.getsize(output_path), 0)
# Evaluate accurarcy on TFLite model.
# Create a new dataset without preprocessing since preprocessing has been
# packaged inside TFLite model.
squeezed_ds = ds.map(lambda x, y: (tf.squeeze(tf.cast(x, tf.float32)), y))
tflite_dataloader = audio_dataloader.DataLoader(squeezed_ds,
len(squeezed_ds),
index_to_labels)
# Evaluate accurarcy on float model.
result = task.evaluate_tflite(output_path, tflite_dataloader)
self.assertGreater(result['accuracy'], .5)
def testBrowserFFT(self):
temp_folder = self.get_temp_dir()
cat1 = write_sample(temp_folder, 'cat', '1.wav', 44100, duration_sec=1)
cat2 = write_sample(temp_folder, 'cat', '2.wav', 44100, duration_sec=2)
dog1 = write_sample(temp_folder, 'dog', '1.wav', 44100, duration_sec=3)
dog2 = write_sample(temp_folder, 'dog', '2.wav', 44100, duration_sec=4)
index_to_labels = ['cat', 'dog']
np.random.seed(123)
tf.random.set_seed(123)
# Prepare data.
spec = audio_spec.BrowserFFTSpec()
ds = tf.data.Dataset.from_tensor_slices(([cat1, cat2, dog1,
dog2], [0, 0, 1, 1]))
data_loader = audio_dataloader.DataLoader(ds, len(ds), index_to_labels,
spec)
# Train a floating point model.
task = audio_classifier.create(data_loader,
spec,
batch_size=1,
epochs=15)
# Evaluate trained model
_, acc = task.evaluate(data_loader)
# Better than random guessing.
self.assertGreater(acc, .5)
# Export the model to saved model.
output_path = os.path.join(spec.model_dir, 'saved_model')
task.export(spec.model_dir, export_format=ExportFormat.SAVED_MODEL)
self.assertTrue(os.path.isdir(output_path))
self.assertNotEqual(len(os.listdir(output_path)), 0)
# Export the model to TFLite.
output_path = os.path.join(spec.model_dir, 'float.tflite')
task.export(spec.model_dir,
tflite_filename='float.tflite',
export_format=ExportFormat.TFLITE)
self.assertTrue(tf.io.gfile.exists(output_path))
self.assertGreater(os.path.getsize(output_path), 0)
# Evaluate accurarcy on TFLite model.
# Create a new dataset without preprocessing since preprocessing has been
# packaged inside TFLite model.
spec = BrowserFFTWithoutPreprocessing()
tflite_dataloader = audio_dataloader.DataLoader(
ds, len(ds), index_to_labels, spec)
# Evaluate accurarcy on float model.
result = task.evaluate_tflite(output_path, tflite_dataloader)
self.assertGreaterEqual(result['accuracy'], .5)
def test_basic_training(self):
tflite_path = self._train_and_export(
audio_spec.BrowserFFTSpec(),
num_classes=5,
filename='basic_5_classes_training.tflite',
expected_model_size=6 * 1000 * 1000)
self.assertEqual(
model_util.extract_tflite_metadata_json(tflite_path), """{
"name": "AudioClassifier",
"description": "Identify the most prominent type in the audio clip from a known set of categories.",
"version": "v1",
"subgraph_metadata": [
{
"input_tensor_metadata": [
{
"name": "audio_clip",
"description": "Input audio clip to be classified.",
"content": {
"content_properties_type": "AudioProperties",
"content_properties": {
"sample_rate": 44100,
"channels": 1
}
},
"stats": {
}
}
],
"output_tensor_metadata": [
{
"name": "probability",
"description": "Scores of the labels respectively.",
"content": {
"content_properties_type": "FeatureProperties",
"content_properties": {
}
},
"stats": {
"max": [
1.0
],
"min": [
0.0
]
},
"associated_files": [
{
"name": "probability_labels.txt",
"description": "Labels for categories that the model can recognize.",
"type": "TENSOR_AXIS_LABELS"
}
]
}
]
}
],
"author": "TensorFlow Lite Model Maker",
"license": "Apache License. Version 2.0 http://www.apache.org/licenses/LICENSE-2.0.",
"min_parser_version": "1.3.0"
}
""")
def test_binary_classification(self):
self._train_and_export(audio_spec.BrowserFFTSpec(),
num_classes=2,
filename='binary_classification.tflite',
expected_model_size=6 * 1000 * 1000)
def setUpClass(cls):
super(BrowserFFTSpecTest, cls).setUpClass()
cls._spec = audio_spec.BrowserFFTSpec()
def testBrowserFFT(self):
self._test_spec(audio_spec.BrowserFFTSpec(),
BrowserFFTWithoutPreprocessing())
