wehub-resource-sync
306 lines
12 KiB
Python
306 lines
12 KiB
Python
# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# 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 unittest
|
|
|
|
import numpy as np
|
|
|
|
import paddle
|
|
from paddle import base
|
|
|
|
|
|
class TestCompatMedianAPI(unittest.TestCase):
|
|
def test_compat_median_basic(self):
|
|
paddle.disable_static()
|
|
|
|
x = paddle.to_tensor([[1, 2, 3], [4, 5, 6], [7, 8, 9]], dtype='float32')
|
|
|
|
result = paddle.compat.median(x)
|
|
expected = paddle.to_tensor(5, dtype='float32')
|
|
np.testing.assert_allclose(result.numpy(), expected.numpy())
|
|
|
|
values, indices = paddle.compat.median(x, dim=1)
|
|
expected_values = paddle.to_tensor([2, 5, 8], dtype='float32')
|
|
expected_indices = paddle.to_tensor([1, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
result = paddle.compat.median(x, dim=1)
|
|
np.testing.assert_allclose(
|
|
result.values.numpy(), expected_values.numpy()
|
|
)
|
|
np.testing.assert_allclose(
|
|
result.indices.numpy(), expected_indices.numpy()
|
|
)
|
|
|
|
values, indices = paddle.compat.median(x, dim=1, keepdim=True)
|
|
expected_values = paddle.to_tensor([[2], [5], [8]], dtype='float32')
|
|
expected_indices = paddle.to_tensor([[1], [1], [1]], dtype='int64')
|
|
np.testing.assert_allclose(values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
paddle.enable_static()
|
|
|
|
def test_compat_median_out(self):
|
|
paddle.disable_static()
|
|
|
|
x = paddle.to_tensor([[1, 2, 3], [4, 5, 6], [7, 8, 9]], dtype='float32')
|
|
|
|
out = paddle.zeros([], dtype='float32')
|
|
result = paddle.compat.median(x, out=out)
|
|
expected = paddle.to_tensor(5, dtype='float32')
|
|
np.testing.assert_allclose(result.numpy(), expected.numpy())
|
|
np.testing.assert_allclose(out.numpy(), expected.numpy())
|
|
self.assertIs(result, out)
|
|
|
|
out_values = paddle.zeros([3], dtype='float32')
|
|
out_indices = paddle.zeros([3], dtype='int64')
|
|
result_values, result_indices = paddle.compat.median(
|
|
x, dim=1, out=(out_values, out_indices)
|
|
)
|
|
expected_values = paddle.to_tensor([2, 5, 8], dtype='float32')
|
|
expected_indices = paddle.to_tensor([1, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(
|
|
result_values.numpy(), expected_values.numpy()
|
|
)
|
|
np.testing.assert_allclose(
|
|
result_indices.numpy(), expected_indices.numpy()
|
|
)
|
|
np.testing.assert_allclose(out_values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(
|
|
out_indices.numpy(), expected_indices.numpy()
|
|
)
|
|
self.assertIs(result_values, out_values)
|
|
self.assertIs(result_indices, out_indices)
|
|
|
|
paddle.enable_static()
|
|
|
|
def test_compat_median_different_dims(self):
|
|
paddle.disable_static()
|
|
|
|
x = paddle.to_tensor([[1, 2, 3], [4, 5, 6], [7, 8, 9]], dtype='float32')
|
|
|
|
values, indices = paddle.compat.median(x, dim=0)
|
|
expected_values = paddle.to_tensor([4, 5, 6], dtype='float32')
|
|
expected_indices = paddle.to_tensor([1, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
values, indices = paddle.compat.median(x, dim=1)
|
|
expected_values = paddle.to_tensor([2, 5, 8], dtype='float32')
|
|
expected_indices = paddle.to_tensor([1, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
values, indices = paddle.compat.median(x, dim=-1)
|
|
expected_values = paddle.to_tensor([2, 5, 8], dtype='float32')
|
|
expected_indices = paddle.to_tensor([1, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
paddle.enable_static()
|
|
|
|
def test_compat_median_static(self):
|
|
paddle.enable_static()
|
|
|
|
with paddle.static.program_guard(
|
|
paddle.static.Program(), paddle.static.Program()
|
|
):
|
|
x = paddle.static.data(name='x', shape=[3, 3], dtype='float32')
|
|
values, indices = paddle.compat.median(x, dim=1)
|
|
|
|
exe = base.Executor(base.CPUPlace())
|
|
x_data = np.array(
|
|
[[1, 2, 3], [4, 5, 6], [7, 8, 9]], dtype='float32'
|
|
)
|
|
result_values, result_indices = exe.run(
|
|
feed={'x': x_data}, fetch_list=[values, indices]
|
|
)
|
|
|
|
expected_values = np.array([2, 5, 8], dtype='float32')
|
|
expected_indices = np.array([1, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(result_values, expected_values)
|
|
np.testing.assert_allclose(result_indices, expected_indices)
|
|
|
|
with paddle.static.program_guard(
|
|
paddle.static.Program(), paddle.static.Program()
|
|
):
|
|
x = paddle.static.data(name='x', shape=[3, 3], dtype='float32')
|
|
result = paddle.compat.median(x, dim=1)
|
|
|
|
exe = base.Executor(base.CPUPlace())
|
|
x_data = np.array(
|
|
[[1, 2, 3], [4, 5, 6], [7, 8, 9]], dtype='float32'
|
|
)
|
|
result_values, result_indices = exe.run(
|
|
feed={'x': x_data}, fetch_list=[result.values, result.indices]
|
|
)
|
|
|
|
expected_values = np.array([2, 5, 8], dtype='float32')
|
|
expected_indices = np.array([1, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(result_values, expected_values)
|
|
np.testing.assert_allclose(result_indices, expected_indices)
|
|
|
|
paddle.disable_static()
|
|
|
|
|
|
class TestCompatNanmedianAPI(unittest.TestCase):
|
|
def test_compat_nanmedian_basic(self):
|
|
paddle.disable_static()
|
|
|
|
x = paddle.to_tensor(
|
|
[[1, float('nan'), 3], [4, 5, 6], [float('nan'), 8, 9]],
|
|
dtype='float32',
|
|
)
|
|
|
|
result = paddle.compat.nanmedian(x)
|
|
expected = paddle.to_tensor(5.0, dtype='float32')
|
|
np.testing.assert_allclose(result.numpy(), expected.numpy())
|
|
|
|
values, indices = paddle.compat.nanmedian(x, dim=1)
|
|
expected_values = paddle.to_tensor([1.0, 5.0, 8.0], dtype='float32')
|
|
expected_indices = paddle.to_tensor([0, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
result = paddle.compat.nanmedian(x, dim=1)
|
|
np.testing.assert_allclose(
|
|
result.values.numpy(), expected_values.numpy()
|
|
)
|
|
np.testing.assert_allclose(
|
|
result.indices.numpy(), expected_indices.numpy()
|
|
)
|
|
|
|
values, indices = paddle.compat.nanmedian(x, dim=-1)
|
|
expected_values = paddle.to_tensor([1.0, 5.0, 8.0], dtype='float32')
|
|
expected_indices = paddle.to_tensor([0, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
values, indices = paddle.compat.nanmedian(x, dim=1, keepdim=True)
|
|
expected_values = paddle.to_tensor(
|
|
[[1.0], [5.0], [8.0]], dtype='float32'
|
|
)
|
|
expected_indices = paddle.to_tensor([[0], [1], [1]], dtype='int64')
|
|
np.testing.assert_allclose(values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
paddle.enable_static()
|
|
|
|
def test_compat_nanmedian_out(self):
|
|
paddle.disable_static()
|
|
|
|
x = paddle.to_tensor(
|
|
[[1, float('nan'), 3], [4, 5, 6], [float('nan'), 8, 9]],
|
|
dtype='float32',
|
|
)
|
|
|
|
out = paddle.zeros([], dtype='float32')
|
|
result = paddle.compat.nanmedian(x, out=out)
|
|
expected = paddle.to_tensor(5.0, dtype='float32')
|
|
np.testing.assert_allclose(result.numpy(), expected.numpy())
|
|
np.testing.assert_allclose(out.numpy(), expected.numpy())
|
|
self.assertIs(result, out)
|
|
|
|
out_values = paddle.zeros([3], dtype='float32')
|
|
out_indices = paddle.zeros([3], dtype='int64')
|
|
result_values, result_indices = paddle.compat.nanmedian(
|
|
x, dim=1, out=(out_values, out_indices)
|
|
)
|
|
expected_values = paddle.to_tensor([1.0, 5.0, 8.0], dtype='float32')
|
|
expected_indices = paddle.to_tensor([0, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(
|
|
result_values.numpy(), expected_values.numpy()
|
|
)
|
|
np.testing.assert_allclose(
|
|
result_indices.numpy(), expected_indices.numpy()
|
|
)
|
|
np.testing.assert_allclose(out_values.numpy(), expected_values.numpy())
|
|
np.testing.assert_allclose(
|
|
out_indices.numpy(), expected_indices.numpy()
|
|
)
|
|
self.assertIs(result_values, out_values)
|
|
self.assertIs(result_indices, out_indices)
|
|
|
|
paddle.enable_static()
|
|
|
|
def test_compat_nanmedian_all_nan(self):
|
|
paddle.disable_static()
|
|
|
|
x = paddle.to_tensor(
|
|
[[1, 2, 3], [float('nan'), float('nan'), float('nan')], [7, 8, 9]],
|
|
dtype='float32',
|
|
)
|
|
|
|
values, indices = paddle.compat.nanmedian(x, dim=1)
|
|
expected_values = paddle.to_tensor(
|
|
[2.0, float('nan'), 8.0], dtype='float32'
|
|
)
|
|
expected_indices = paddle.to_tensor([1, 0, 1], dtype='int64')
|
|
np.testing.assert_allclose(
|
|
values.numpy(), expected_values.numpy(), equal_nan=True
|
|
)
|
|
np.testing.assert_allclose(indices.numpy(), expected_indices.numpy())
|
|
|
|
paddle.enable_static()
|
|
|
|
def test_compat_nanmedian_static(self):
|
|
paddle.enable_static()
|
|
|
|
with paddle.static.program_guard(
|
|
paddle.static.Program(), paddle.static.Program()
|
|
):
|
|
x = paddle.static.data(name='x', shape=[3, 3], dtype='float32')
|
|
values, indices = paddle.compat.nanmedian(x, dim=1)
|
|
|
|
exe = base.Executor(base.CPUPlace())
|
|
x_data = np.array(
|
|
[[1, float('nan'), 3], [4, 5, 6], [float('nan'), 8, 9]],
|
|
dtype='float32',
|
|
)
|
|
result_values, result_indices = exe.run(
|
|
feed={'x': x_data}, fetch_list=[values, indices]
|
|
)
|
|
|
|
expected_values = np.array([1.0, 5.0, 8.0], dtype='float32')
|
|
expected_indices = np.array([0, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(result_values, expected_values)
|
|
np.testing.assert_allclose(result_indices, expected_indices)
|
|
|
|
with paddle.static.program_guard(
|
|
paddle.static.Program(), paddle.static.Program()
|
|
):
|
|
x = paddle.static.data(name='x', shape=[3, 3], dtype='float32')
|
|
result = paddle.compat.nanmedian(x, dim=1)
|
|
|
|
exe = base.Executor(base.CPUPlace())
|
|
x_data = np.array(
|
|
[[1, float('nan'), 3], [4, 5, 6], [float('nan'), 8, 9]],
|
|
dtype='float32',
|
|
)
|
|
result_values, result_indices = exe.run(
|
|
feed={'x': x_data}, fetch_list=[result.values, result.indices]
|
|
)
|
|
|
|
expected_values = np.array([1.0, 5.0, 8.0], dtype='float32')
|
|
expected_indices = np.array([0, 1, 1], dtype='int64')
|
|
np.testing.assert_allclose(result_values, expected_values)
|
|
np.testing.assert_allclose(result_indices, expected_indices)
|
|
|
|
paddle.disable_static()
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|