-
Notifications
You must be signed in to change notification settings - Fork 49
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
first round of automatic improvements
- Loading branch information
Showing
13 changed files
with
317 additions
and
775 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,3 +1,4 @@ | ||
| from setuptools import setup | ||
|
|
||
| # Setup configuration using Python Build Reasonableness (PBR) | ||
| setup(setup_requires=["pbr"], pbr=True) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,139 +1,108 @@ | ||
| import numpy as np | ||
| import pytest | ||
| from tonic.audio_transforms import (AddNoise, AmplitudeScale, Bin, FixLength, | ||
| LinearButterFilterBank, | ||
| MelButterFilterBank, RobustAmplitudeScale, | ||
| SwapAxes) | ||
|
|
||
|
|
||
| def test_standardize_data_length(): | ||
| from tonic.audio_transforms import FixLength | ||
| class DummyNoiseDataset: | ||
| """Dummy dataset for generating random noise signals of varying lengths.""" | ||
|
|
||
| def __len__(self) -> int: | ||
| return 1000 | ||
|
|
||
| def __getitem__(self, item: int) -> tuple[np.ndarray, int]: | ||
| sig_len = np.random.randint(12000, 20000) | ||
| return np.random.random((1, sig_len)), 0 | ||
|
|
||
|
|
||
| def test_standardize_data_length(): | ||
| """Test FixLength transform for both extending and truncating data.""" | ||
| sdl = FixLength(100, 1) | ||
|
|
||
| # Data is longer | ||
| data = np.ones((1, 120)) | ||
| assert sdl(data).shape == (1, 100) | ||
| data_long = np.ones((1, 120)) | ||
| assert sdl(data_long).shape == (1, 100) | ||
|
|
||
| # Data is shorter | ||
| data = np.ones((1, 80)) | ||
| assert sdl(data).shape == (1, 100) | ||
| data_short = np.ones((1, 80)) | ||
| assert sdl(data_short).shape == (1, 100) | ||
|
|
||
|
|
||
| def test_bin(): | ||
| import numpy as np | ||
|
|
||
| from tonic.audio_transforms import Bin | ||
| """Test Bin transform for re-sampling data frequency.""" | ||
| bin_transform = Bin(orig_freq=16000, new_freq=100, axis=1) | ||
|
|
||
| bin_transform = Bin(orig_freq=16_000, new_freq=100, axis=1) | ||
|
|
||
| data = np.random.random((1, 8 * 16_000)) | ||
| data = np.random.random((1, 8 * 16000)) | ||
| data_binned = bin_transform(data) | ||
|
|
||
| assert data_binned.shape == (1, 8 * 100) | ||
| assert data.sum() == pytest.approx(data_binned.sum(), 1e-7) | ||
|
|
||
|
|
||
| def test_linear_butter_filter_bank(): | ||
| import numpy as np | ||
|
|
||
| from tonic.audio_transforms import LinearButterFilterBank | ||
|
|
||
| """Test LinearButterFilterBank for filtering data.""" | ||
| fb = LinearButterFilterBank( | ||
| order=2, low_freq=100, sampling_freq=16000, num_filters=16 | ||
| ) | ||
| data = np.random.random((1, 16_000)) | ||
| data = np.random.random((1, 16000)) | ||
|
|
||
| filter_out = fb(data) | ||
| assert filter_out.shape == (16, 16_000) | ||
| assert filter_out.shape == (16, 16000) | ||
|
|
||
|
|
||
| def test_mel_butter_filter_bank(): | ||
| import numpy as np | ||
|
|
||
| from tonic.audio_transforms import MelButterFilterBank | ||
|
|
||
| """Test MelButterFilterBank for filtering data using Mel scale.""" | ||
| fb = MelButterFilterBank(order=2, low_freq=100, sampling_freq=16000, num_filters=16) | ||
| data = np.random.random((1, 16_000)) | ||
| data = np.random.random((1, 16000)) | ||
|
|
||
| filter_out = fb(data) | ||
| assert filter_out.shape == (16, 16_000) | ||
| assert filter_out.shape == (16, 16000) | ||
|
|
||
|
|
||
| def test_add_noise(): | ||
| import numpy as np | ||
|
|
||
| class DummyNoiseDataset: | ||
| def __len__(self): | ||
| return 1000 | ||
|
|
||
| def __getitem__(self, item): | ||
| sig_len = np.random.randint(12000, 20000, (1,)).item() | ||
| return np.random.random((1, sig_len)), 0 | ||
|
|
||
| data = np.sin(np.arange(0, 16_000 * 0.001, 0.001))[None, ...] | ||
| print(data.shape) | ||
|
|
||
| from tonic.audio_transforms import AddNoise | ||
|
|
||
| """Test AddNoise transform for adding noise to a signal.""" | ||
| data = np.sin(np.linspace(0, 2 * np.pi, 16000))[None, ...] | ||
| noise_dataset = DummyNoiseDataset() | ||
|
|
||
| print(noise_dataset[0][0].shape) | ||
|
|
||
| add_noise = AddNoise(noise_dataset, 10, normed=True) | ||
|
|
||
| signal = add_noise(data) | ||
| assert signal.shape == (1, 16_000) | ||
| assert signal.shape == (1, 16000) | ||
|
|
||
|
|
||
| def test_swap_axes(): | ||
| """Tests SwapAxes transform with synthetic data.""" | ||
| from tonic.audio_transforms import SwapAxes | ||
|
|
||
| np.random.seed(123) | ||
| sr = 16_000 # sample rate | ||
| sl = 1 # sample length | ||
| data = np.random.rand(1, sr * sl) | ||
| ax1, ax2 = 0, 1 | ||
| swap_ax = SwapAxes(ax1=ax1, ax2=ax2) | ||
| """Test SwapAxes transform for swapping axes of the data.""" | ||
| data = np.random.rand(1, 16000) | ||
| swap_ax = SwapAxes(ax1=0, ax2=1) | ||
| swaped = swap_ax(data) | ||
|
|
||
| assert swaped.shape[0] == data.shape[1] | ||
| assert swaped.shape[1] == data.shape[0] | ||
| assert swaped.shape == (16000, 1) | ||
|
|
||
|
|
||
| def test_amplitude_scale(): | ||
| """Tests the amplitude scaling transform with synthetic data.""" | ||
| from tonic.audio_transforms import AmplitudeScale | ||
|
|
||
| np.random.seed(123) | ||
| sr = 16_000 # sample rate | ||
| sl = 1 # sample length | ||
| data = np.random.rand(1, sr * sl) | ||
| """Test AmplitudeScale transform for scaling amplitude of the data.""" | ||
| data = np.random.rand(1, 16000) | ||
| max_amps = np.random.rand(10) | ||
|
|
||
| for amp in max_amps: | ||
| AmpScale = AmplitudeScale(max_amplitude=amp) | ||
| transformed = AmpScale(data) | ||
| assert data.shape[1] == transformed.shape[1] | ||
| amp_scale = AmplitudeScale(max_amplitude=amp) | ||
| transformed = amp_scale(data) | ||
| assert transformed.max() == amp | ||
|
|
||
|
|
||
| def test_robust_amplitude_scale(): | ||
| """Tests robust amplitude scaling transform with a synthetic data.""" | ||
| from tonic.audio_transforms import RobustAmplitudeScale | ||
|
|
||
| np.random.seed(123) | ||
| sr = 16_000 # sample rate | ||
| sl = 1 # sample length | ||
| data = np.random.rand(1, sr * sl) | ||
| """Test RobustAmplitudeScale transform for scaling amplitude considering outliers.""" | ||
| data = np.random.rand(1, 16000) | ||
| max_amps = np.random.rand(10) | ||
| percent = 0.01 | ||
|
|
||
| for amp in max_amps: | ||
| RobustAmpScale = RobustAmplitudeScale( | ||
| robust_amp_scale = RobustAmplitudeScale( | ||
| max_robust_amplitude=amp, outlier_percent=percent | ||
| ) | ||
| transformed = RobustAmpScale(data) | ||
| transformed = robust_amp_scale(data) | ||
| sorted_transformed = np.sort(np.abs(transformed.ravel())) | ||
| non_outlier = sorted_transformed[ | ||
| 0 : int(np.floor(len(sorted_transformed)) * (1 - percent)) | ||
| ] | ||
| print(non_outlier) | ||
| assert data.shape[1] == transformed.shape[1] | ||
| non_outlier = sorted_transformed[: int(len(sorted_transformed) * (1 - percent))] | ||
|
|
||
| assert np.all(non_outlier <= amp) |
This file was deleted.
Oops, something went wrong.
Oops, something went wrong.