mindspore/tests/st/ops/gpu/test_maskedfill_op.py

144 lines
4.5 KiB
Python

# 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.
# ============================================================================
import numpy as np
import pytest
import mindspore.context as context
import mindspore.nn as nn
from mindspore import Tensor
from mindspore.ops import operations as P
context.set_context(mode=context.GRAPH_MODE, device_target="GPU")
class MaskedFillNet(nn.Cell):
def __init__(self):
super(MaskedFillNet, self).__init__()
self.maskedfill = P.MaskedFill()
def construct(self, inputs, mask, value):
return self.maskedfill(inputs, mask, value)
def maskedfill_fun(ntype):
maskedfill_net = MaskedFillNet()
inputs = Tensor(np.array([[1, 2, 3, 4], [5, 6, 7, 8]]).astype(ntype))
mask = Tensor(np.array([[True, True, False, True], [False, False, True, False]]).astype(np.bool))
value = Tensor(np.array(22).astype(ntype))
expect = np.array([[22, 22, 3, 22], [5, 6, 22, 8]]).astype(ntype)
output = maskedfill_net(inputs, mask, value)
assert (output.asnumpy() == expect).all()
mask = Tensor(np.array([[True, True, True, True], [True, True, True, True]]).astype(np.bool))
value = Tensor(np.array(1).astype(ntype))
expect = np.array([[1, 1, 1, 1], [1, 1, 1, 1]]).astype(ntype)
output = maskedfill_net(inputs, mask, value)
assert (output.asnumpy() == expect).all()
mask = Tensor(np.array([[False, False, False, False], [False, False, False, False]]).astype(np.bool))
value = Tensor(np.array(22).astype(ntype))
expect = np.array([[1, 2, 3, 4], [5, 6, 7, 8]]).astype(ntype)
output = maskedfill_net(inputs, mask, value)
assert (output.asnumpy() == expect).all()
# BroadCast
mask = Tensor(np.array([True, True, False, True]).astype(np.bool))
value = Tensor(np.array(22).astype(ntype))
expect = np.array([[22, 22, 3, 22], [22, 22, 7, 22]]).astype(ntype)
output = maskedfill_net(inputs, mask, value)
assert (output.asnumpy() == expect).all()
@pytest.mark.level0
@pytest.mark.platform_x86_gpu_training
@pytest.mark.env_onecard
def test_maskedfill_float():
"""
Feature: Test MaskedFill op.
Description: Test MaskedFill with float input.
Expectation: The result match to expect.
"""
maskedfill_fun(np.float32)
@pytest.mark.level0
@pytest.mark.platform_x86_gpu_training
@pytest.mark.env_onecard
def test_maskedfill_float16():
"""
Feature: Test MaskedFill op.
Description: Test MaskedFill with float16 input.
Expectation: The result match to expect.
"""
maskedfill_fun(np.float16)
@pytest.mark.level0
@pytest.mark.platform_x86_gpu_training
@pytest.mark.env_onecard
def test_maskedfill_int():
"""
Feature: Test MaskedFill op.
Description: Test MaskedFill with int input.
Expectation: The result match to expect.
"""
maskedfill_fun(np.int32)
@pytest.mark.level0
@pytest.mark.platform_x86_gpu_training
@pytest.mark.env_onecard
def test_maskedfill_int8():
"""
Feature: Test MaskedFill op.
Description: Test MaskedFill with int8 input.
Expectation: The result match to expect.
"""
maskedfill_fun(np.int8)
def maskedfill_value(value):
maskedfill_net = MaskedFillNet()
inputs = Tensor(np.array([1, 2, 3, 4]).astype(np.float32))
mask = Tensor(np.array([True, True, False, True]).astype(np.bool))
expect = np.array([0.5, 0.5, 3, 0.5]).astype(np.float32)
output = maskedfill_net(inputs, mask, value)
assert (output.asnumpy() == expect).all()
@pytest.mark.level0
@pytest.mark.platform_x86_gpu_training
@pytest.mark.env_onecard
def test_maskedfill_float_value():
"""
Feature: Test MaskedFill op.
Description: Test MaskedFill with float value.
Expectation: The result match to expect.
"""
maskedfill_value(0.5)
@pytest.mark.level0
@pytest.mark.platform_x86_gpu_training
@pytest.mark.env_onecard
def test_maskedfill_tensor_value():
"""
Feature: Test MaskedFill op.
Description: Test MaskedFill with tensor input.
Expectation: The result match to expect.
"""
maskedfill_value(Tensor(0.5))