forked from mindspore-Ecosystem/mindspore
315 lines
19 KiB
Python
315 lines
19 KiB
Python
# Copyright 2021 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.
|
|
# ============================================================================
|
|
|
|
import mindspore
|
|
from mindspore.common.initializer import initializer, Identity, Dirac, Sparse, VarianceScaling, Orthogonal
|
|
import numpy as np
|
|
|
|
|
|
def test_sparse():
|
|
"""
|
|
Feature: Test sparse initializer.
|
|
Description: Initialize a 2 dimension sparse matrix to fill the input tensor.
|
|
Expectation: The Tensor is initialized with a 2 dimension sparse matrix.
|
|
"""
|
|
tensor1 = initializer(Sparse(sparsity=0.1, sigma=0.01), [5, 8], mindspore.float32)
|
|
tensor1.init_data()
|
|
|
|
|
|
def test_orthogonal():
|
|
"""
|
|
Feature: Test orthogonal initializer.
|
|
Description: Initialize a (semi) orthogonal matrix to fill the input tensor.
|
|
Expectation: The Tensor is initialized with values from orthogonal matrix.
|
|
"""
|
|
tensor1 = initializer(Orthogonal(gain=2.), [2, 3, 4], mindspore.float32)
|
|
tensor2 = initializer('orthogonal', [2, 3, 4], mindspore.float32)
|
|
tensor1.init_data()
|
|
tensor2.init_data()
|
|
|
|
|
|
def test_variancescaling():
|
|
"""
|
|
Feature: Test varianceScaling initializer.
|
|
Description: Randomly initialize an array with scaling to fill the input tensor.
|
|
Expectation: The Tensor is initialized successfully.
|
|
"""
|
|
tensor1 = initializer('varianceScaling', [2, 3], mindspore.float32)
|
|
tensor2 = initializer(VarianceScaling(scale=1.0, mode='fan_out', distribution='untruncated_normal'), [2, 3],
|
|
mindspore.float32)
|
|
tensor3 = initializer(VarianceScaling(scale=2.0, mode='fan_in', distribution='truncated_normal'), [2, 3],
|
|
mindspore.float32)
|
|
tensor4 = initializer(VarianceScaling(scale=3.0, mode='fan_avg', distribution='uniform'), [2, 3],
|
|
mindspore.float32)
|
|
tensor1.init_data()
|
|
tensor2.init_data()
|
|
tensor3.init_data()
|
|
tensor4.init_data()
|
|
|
|
|
|
def test_identity():
|
|
"""
|
|
Feature: Test identity initializer.
|
|
Description: Initialize an identity matrix to fill a Tensor.
|
|
Expectation: The Tensor is initialized with identity matrix.
|
|
"""
|
|
tensor1 = initializer(Identity(), [3, 3], mindspore.float32)
|
|
tensor2 = initializer('identity', [3, 4], mindspore.float32)
|
|
tensor3 = initializer('identity', [4, 3], mindspore.float32)
|
|
expect1 = np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1]], dtype=np.float32)
|
|
expect2 = np.array([[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0]], dtype=np.float32)
|
|
expect3 = np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1], [0, 0, 0]], dtype=np.float32)
|
|
assert (tensor1.asnumpy() == expect1).all()
|
|
assert (tensor2.asnumpy() == expect2).all()
|
|
assert (tensor3.asnumpy() == expect3).all()
|
|
|
|
|
|
def test_dirac():
|
|
"""
|
|
Feature: Test dirac initializer.
|
|
Description: Initialize input tensor with the Dirac delta function.
|
|
Expectation: The Tensor is correctly initialized.
|
|
"""
|
|
tensor3_1 = initializer(Dirac(groups=1), [6, 2, 3], mindspore.float32)
|
|
tensor3_2 = initializer(Dirac(groups=2), [6, 2, 3], mindspore.float32)
|
|
tensor3_3 = initializer(Dirac(groups=3), [6, 2, 3], mindspore.float32)
|
|
|
|
tensor4_1 = initializer(Dirac(groups=1), [6, 4, 3, 3], mindspore.float32)
|
|
tensor4_2 = initializer(Dirac(groups=2), [6, 4, 3, 3], mindspore.float32)
|
|
tensor4_3 = initializer(Dirac(groups=3), [6, 4, 3, 3], mindspore.float32)
|
|
|
|
tensor5_1 = initializer(Dirac(groups=1), [6, 2, 3, 3, 3], mindspore.float32)
|
|
tensor5_2 = initializer(Dirac(groups=2), [6, 2, 3, 3, 3], mindspore.float32)
|
|
tensor5_3 = initializer(Dirac(groups=3), [6, 2, 3, 3, 3], mindspore.float32)
|
|
|
|
expectation3_1 = np.array([[[0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.]]], dtype=np.float32)
|
|
|
|
expectation3_2 = np.array([[[0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.]]], dtype=np.float32)
|
|
|
|
expectation3_3 = np.array([[[0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.]],
|
|
[[0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.]],
|
|
[[0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.]]], dtype=np.float32)
|
|
|
|
expectation4_1 = np.array([[[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]], dtype=np.float32)
|
|
|
|
expectation4_2 = np.array([[[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]], dtype=np.float32)
|
|
|
|
expectation4_3 = np.array([[[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]], dtype=np.float32)
|
|
|
|
expectation5_1 = np.array([[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]]], dtype=np.float32)
|
|
|
|
expectation5_2 = np.array([[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]]], dtype=np.float32)
|
|
|
|
expectation5_3 = np.array([[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]],
|
|
[[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]],
|
|
[[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 1., 0.], [0., 0., 0.]],
|
|
[[0., 0., 0.], [0., 0., 0.], [0., 0., 0.]]]]], dtype=np.float32)
|
|
|
|
assert (tensor3_1.asnumpy() == expectation3_1).all()
|
|
assert (tensor3_2.asnumpy() == expectation3_2).all()
|
|
assert (tensor3_3.asnumpy() == expectation3_3).all()
|
|
|
|
assert (tensor4_1.asnumpy() == expectation4_1).all()
|
|
assert (tensor4_2.asnumpy() == expectation4_2).all()
|
|
assert (tensor4_3.asnumpy() == expectation4_3).all()
|
|
|
|
assert (tensor5_1.asnumpy() == expectation5_1).all()
|
|
assert (tensor5_2.asnumpy() == expectation5_2).all()
|
|
assert (tensor5_3.asnumpy() == expectation5_3).all()
|