# Copyright 2022 Huawei Technologies Co., Ltd # # 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. # ============================================================================== """ Testing MelScale op in DE """ import numpy as np import pytest import mindspore.dataset as ds import mindspore.dataset.audio.transforms as c_audio from mindspore import log as logger from mindspore.dataset.audio.utils import MelType, NormType CHANNEL = 1 FREQ = 20 TIME = 15 DEFAULT_N_MELS = 128 def gen(shape, dtype=np.float32): np.random.seed(0) data = np.random.random(shape) yield (np.array(data, dtype=dtype),) def count_unequal_element(data_expected, data_me, rtol, atol): assert data_expected.shape == data_me.shape total_count = len(data_expected.flatten()) error = np.abs(data_expected - data_me) greater = np.greater(error, atol + np.abs(data_expected) * rtol) loss_count = np.count_nonzero(greater) assert (loss_count / total_count) < rtol, \ "\ndata_expected_std:{0}\ndata_me_error:{1}\nloss:{2}". \ format(data_expected[greater], data_me[greater], error[greater]) def allclose_nparray(data_expected, data_me, rtol, atol, equal_nan=True): if np.any(np.isnan(data_expected)): assert np.allclose(data_me, data_expected, rtol, atol, equal_nan=equal_nan) elif not np.allclose(data_me, data_expected, rtol, atol, equal_nan=equal_nan): count_unequal_element(data_expected, data_me, rtol, atol) def test_mel_scale_pipeline(): """ Feature: MelScale Description: test MelScale cpp op in pipeline Expectation: equal results from Mindspore and benchmark """ in_data = np.array([[[[-0.34207549691200256, -2.0971477031707764, -0.9462487101554871], [1.2536851167678833, -1.3225716352462769, -0.06942684203386307], [-0.4859708547592163, -0.4990693926811218, 0.2322249710559845], [-0.7589328289031982, -2.218672513961792, -0.8374152779579163]], [[1.0313602685928345, -1.5596215724945068, 0.46823829412460327], [0.14756731688976288, 0.35987502336502075, -1.3228634595870972], [-0.7677955627441406, -0.059919968247413635, 0.7958201766014099], [-0.6194286942481995, -0.5878928899765015, 0.3874965310096741]]]]).astype(np.float32) out_expect = np.array([[[-0.24386560916900635, -5.417530059814453, -1.4391992092132568], [-0.08942853659391403, -0.7199308276176453, -0.18166661262512207]], [[-0.0856514573097229, -1.6701887845993042, 0.25840121507644653], [-0.12264516949653625, -0.1773705929517746, 0.07029043138027191]]]).astype(np.float32) dataset = ds.NumpySlicesDataset(in_data, column_names=["multi_dimensional_data"], shuffle=False) transforms = [c_audio.MelScale(n_mels=2, sample_rate=10, f_min=-50, f_max=100, n_stft=4)] dataset = dataset.map(operations=transforms, input_columns=["multi_dimensional_data"]) for item in dataset.create_dict_iterator(num_epochs=1, output_numpy=True): out_put = item["multi_dimensional_data"] assert out_put.shape == (2, 2, 3) allclose_nparray(out_put, out_expect, 0.001, 0.001) def test_mel_scale_pipeline_invalid_param(): """ Feature: MelScale Description: test MelScale with invalid input parameters Expectation: throw ValueError or TypeError """ logger.info("test MelScale op with default values") generator = gen([CHANNEL, FREQ, TIME]) data1 = ds.GeneratorDataset(source=generator, column_names=["multi_dimensional_data"]) with pytest.raises(ValueError, match="MelScale: f_max should be greater than f_min."): transforms = [c_audio.MelScale(n_mels=128, sample_rate=16200, f_min=1000, f_max=1000)] data1 = data1.map(operations=transforms, input_columns=["multi_dimensional_data"]) for item in data1.create_dict_iterator(num_epochs=1, output_numpy=True): _ = item["multi_dimensional_data"] with pytest.raises(ValueError, match=r"Input n_mels is not within the required interval of \[1, 2147483647\]."): transforms = [c_audio.MelScale(n_mels=-1, sample_rate=16200, f_min=10, f_max=1000)] data1 = data1.map(operations=transforms, input_columns=["multi_dimensional_data"]) with pytest.raises(ValueError, match=r"Input sample_rate is not within the required interval of \[1, 2147483647\]."): transforms = [c_audio.MelScale(n_mels=128, sample_rate=0, f_min=10, f_max=1000)] data1 = data1.map(operations=transforms, input_columns=["multi_dimensional_data"]) with pytest.raises(ValueError, match=r"Input f_max is not within the required interval of \(0, 16777216\]."): transforms = [c_audio.MelScale(n_mels=128, sample_rate=16200, f_min=10, f_max=-10)] data1 = data1.map(operations=transforms, input_columns=["multi_dimensional_data"]) with pytest.raises(TypeError, match=r"Argument norm with value slaney is not of type \[\], " + "but got ."): transforms = [c_audio.MelScale(n_mels=128, sample_rate=16200, f_min=10, f_max=1000, norm="slaney", mel_type=MelType.SLANEY)] data1 = data1.map(operations=transforms, input_columns=["multi_dimensional_data"]) with pytest.raises(TypeError, match=r"Argument mel_type with value SLANEY is not of type \[\], " + "but got ."): transforms = [c_audio.MelScale(n_mels=128, sample_rate=16200, f_min=10, f_max=1000, norm=NormType.NONE, mel_type="SLANEY")] data1 = data1.map(operations=transforms, input_columns=["multi_dimensional_data"]) def test_mel_scale_eager(): """ Feature: MelScale Description: test MelScale cpp op with eage mode Expectation: equal results from Mindspore and benchmark """ spectrogram = np.array([[[-0.7010437250137329, 1.1184569597244263, -1.4936821460723877], [0.4603022038936615, -0.556514322757721, 0.8629537224769592]], [[0.41759368777275085, 1.0594186782836914, -0.07423319667577744], [0.47624683380126953, -0.33720797300338745, 2.0135815143585205]], [[-0.6765501499176025, 0.8924005031585693, 1.0404413938522339], [-0.5578446984291077, -0.349029004573822, 0.0370720773935318]]]) spectrogram = spectrogram.astype(np.float32) out_ms = c_audio.MelScale(n_mels=2, sample_rate=10, f_min=-50, f_max=100, n_stft=2)(spectrogram) out_expect = np.array([[[-0.27036190032958984, 0.579207181930542, -0.6739760637283325], [0.029620330780744553, -0.017264455556869507, 0.043247632682323456]], [[0.7849390506744385, 0.706536054611206, 1.6048823595046997], [0.10890152305364609, 0.01567467674612999, 0.33446595072746277]], [[-1.0940029621124268, 0.5411258339881897, 1.000023603439331], [-0.14039191603660583, 0.002245672047138214, 0.07748986035585403]]]).astype(np.float32) allclose_nparray(out_ms, out_expect, 0.001, 0.001) assert out_ms.shape == (3, 2, 3) if __name__ == "__main__": test_mel_scale_pipeline() test_mel_scale_pipeline_invalid_param() test_mel_scale_eager()