wehub-resource-sync
487 lines
14 KiB
Python
487 lines
14 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
|
|
from tensorrt_test_base import TensorRTBaseTest
|
|
|
|
import paddle
|
|
|
|
|
|
def pool2d_api(
|
|
x,
|
|
ksize=[],
|
|
strides=[],
|
|
paddings=[],
|
|
ceil_mode=False,
|
|
exclusive=True,
|
|
data_format="NCHW",
|
|
pooling_type="max",
|
|
global_pooling=False,
|
|
adaptive=False,
|
|
padding_algorithm="EXPLICIT",
|
|
):
|
|
return paddle._C_ops.pool2d(
|
|
x,
|
|
ksize,
|
|
strides,
|
|
paddings,
|
|
ceil_mode,
|
|
exclusive,
|
|
data_format,
|
|
pooling_type,
|
|
global_pooling,
|
|
adaptive,
|
|
padding_algorithm,
|
|
)
|
|
|
|
|
|
class TestPoolingTRTPattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = paddle.nn.AvgPool2D(kernel_size=2, stride=1)
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 1, 2, 3).astype("float32"),
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 1, 2, 3]}
|
|
self.opt_shape = {"x": [1, 1, 2, 3]}
|
|
self.max_shape = {"x": [5, 1, 2, 3]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTCase1Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 1, 2, 3).astype("float32"),
|
|
"ksize": [2, 3],
|
|
"strides": [1, 2],
|
|
"paddings": [0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": False,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": False,
|
|
"adaptive": False,
|
|
"padding_algorithm": "VALID",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 1, 2, 3]}
|
|
self.opt_shape = {"x": [1, 1, 2, 3]}
|
|
self.max_shape = {"x": [5, 1, 2, 3]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTCase2Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 1, 2, 3).astype("float32"),
|
|
"ksize": [2, 3],
|
|
"strides": [1, 2],
|
|
"paddings": [0, 0],
|
|
"ceil_mode": True,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "max",
|
|
"global_pooling": False,
|
|
"adaptive": False,
|
|
"padding_algorithm": "SAME",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 1, 2, 3]}
|
|
self.opt_shape = {"x": [1, 1, 2, 3]}
|
|
self.max_shape = {"x": [5, 1, 2, 3]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTCase3Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 1, 2, 3).astype("float32"),
|
|
"ksize": [2, 3],
|
|
"strides": [1, 2],
|
|
"paddings": [0, 0],
|
|
"ceil_mode": True,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "max",
|
|
"global_pooling": True,
|
|
"adaptive": False,
|
|
"padding_algorithm": "SAME",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 1, 2, 3]}
|
|
self.opt_shape = {"x": [1, 1, 2, 3]}
|
|
self.max_shape = {"x": [5, 1, 2, 3]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTCase4Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 1, 5, 5).astype("float32"),
|
|
"ksize": [3, 3],
|
|
"strides": [1, 1],
|
|
"paddings": [0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": False,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": True,
|
|
"adaptive": False,
|
|
"padding_algorithm": "SAME",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 1, 5, 5]}
|
|
self.opt_shape = {"x": [1, 1, 5, 5]}
|
|
self.max_shape = {"x": [5, 1, 5, 5]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTCase5Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 16, 56, 56).astype("float32"),
|
|
"ksize": [2, 2],
|
|
"strides": [1, 1],
|
|
"paddings": [0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": False,
|
|
"adaptive": True,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 16, 56, 56]}
|
|
self.opt_shape = {"x": [1, 16, 56, 56]}
|
|
self.max_shape = {"x": [5, 16, 56, 56]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTCase6Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 3, 5, 5).astype("float32"),
|
|
"ksize": [1, 1],
|
|
"strides": [1, 1],
|
|
"paddings": [0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": False,
|
|
"adaptive": True,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 5, 5]}
|
|
self.opt_shape = {"x": [1, 3, 5, 5]}
|
|
self.max_shape = {"x": [2, 3, 5, 5]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTCase7Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 3, 32, 32).astype("float32"),
|
|
"ksize": [2, 3],
|
|
"strides": [1, 2],
|
|
"paddings": [0, 2],
|
|
"ceil_mode": True,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "max",
|
|
"global_pooling": False,
|
|
"adaptive": False,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 32, 32]}
|
|
self.opt_shape = {"x": [1, 3, 32, 32]}
|
|
self.max_shape = {"x": [2, 3, 32, 32]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTCase8Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 3, 32, 32).astype("float32"),
|
|
"ksize": [2, 3],
|
|
"strides": [1, 2],
|
|
"paddings": [0, 2],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "max",
|
|
"global_pooling": False,
|
|
"adaptive": True,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 32, 32]}
|
|
self.opt_shape = {"x": [1, 3, 32, 32]}
|
|
self.max_shape = {"x": [2, 3, 32, 32]}
|
|
|
|
def test_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
|
|
class TestPoolingTRTMarker(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool2d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 3, 5, 5).astype("float32"),
|
|
"ksize": [6, 6],
|
|
"strides": [2, 2],
|
|
"paddings": [0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": False,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": False,
|
|
"adaptive": False,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.target_marker_op = "pd_op.pool2d"
|
|
|
|
def test_trt_result(self):
|
|
self.check_marker(expected_result=False)
|
|
|
|
|
|
def pool3d_api(
|
|
x,
|
|
ksize=[],
|
|
strides=[],
|
|
paddings=[],
|
|
ceil_mode=False,
|
|
exclusive=True,
|
|
data_format="NCHW",
|
|
pooling_type="max",
|
|
global_pooling=False,
|
|
adaptive=False,
|
|
padding_algorithm="EXPLICIT",
|
|
):
|
|
return paddle._C_ops.pool3d(
|
|
x,
|
|
ksize,
|
|
strides,
|
|
paddings,
|
|
ceil_mode,
|
|
exclusive,
|
|
data_format,
|
|
pooling_type,
|
|
global_pooling,
|
|
adaptive,
|
|
padding_algorithm,
|
|
)
|
|
|
|
|
|
class TestPooling3dTRTCase1Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool3d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 3, 5, 5, 5).astype("float32"),
|
|
"ksize": [1, 1, 1],
|
|
"strides": [1, 1, 1],
|
|
"paddings": [0, 0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": False,
|
|
"adaptive": False,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.opt_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.max_shape = {"x": [2, 3, 5, 5, 5]}
|
|
|
|
def test_fp32_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
def test_fp16_trt_result(self):
|
|
self.check_trt_result(precision_mode="fp16")
|
|
|
|
|
|
class TestPooling3dTRTCase2Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool3d_api
|
|
self.api_args = {
|
|
"x": np.ones([1, 3, 5, 5, 5]).astype("float32"),
|
|
"ksize": [1, 1, 1],
|
|
"strides": [1, 1, 1],
|
|
"paddings": [0, 0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": False,
|
|
"adaptive": True,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.opt_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.max_shape = {"x": [2, 3, 5, 5, 5]}
|
|
|
|
def test_fp32_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
def test_fp16_trt_result(self):
|
|
self.check_trt_result(precision_mode="fp16")
|
|
|
|
|
|
class TestPooling3dTRTCase3Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool3d_api
|
|
self.api_args = {
|
|
"x": np.ones([1, 3, 5, 5, 5]).astype("float32"),
|
|
"ksize": [1, 1, 1],
|
|
"strides": [1, 1, 1],
|
|
"paddings": [0, 0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": True,
|
|
"adaptive": False,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.opt_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.max_shape = {"x": [2, 3, 5, 5, 5]}
|
|
|
|
def test_fp32_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
def test_fp16_trt_result(self):
|
|
self.check_trt_result(precision_mode="fp16")
|
|
|
|
|
|
class TestPooling3dTRTCase4Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool3d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 3, 5, 5, 5).astype("float32"),
|
|
"ksize": [1, 1, 1],
|
|
"strides": [1, 1, 1],
|
|
"paddings": [0, 0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": False,
|
|
"adaptive": False,
|
|
"padding_algorithm": "SAME",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.opt_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.max_shape = {"x": [2, 3, 5, 5, 5]}
|
|
|
|
def test_fp32_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
def test_fp16_trt_result(self):
|
|
self.check_trt_result(precision_mode="fp16")
|
|
|
|
|
|
class TestPooling3dTRTCase5Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool3d_api
|
|
self.api_args = {
|
|
"x": np.random.randn(1, 3, 5, 5, 5).astype("float32"),
|
|
"ksize": [1, 1, 1],
|
|
"strides": [1, 1, 1],
|
|
"paddings": [0, 0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "avg",
|
|
"global_pooling": False,
|
|
"adaptive": False,
|
|
"padding_algorithm": "VALID",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.opt_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.max_shape = {"x": [2, 3, 5, 5, 5]}
|
|
|
|
def test_fp32_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
def test_fp16_trt_result(self):
|
|
self.check_trt_result(precision_mode="fp16")
|
|
|
|
|
|
class TestPooling3dTRTCase6Pattern(TensorRTBaseTest):
|
|
def setUp(self):
|
|
self.python_api = pool3d_api
|
|
self.api_args = {
|
|
"x": np.ones([1, 3, 5, 5, 5]).astype("float32"),
|
|
"ksize": [1, 1, 1],
|
|
"strides": [1, 1, 1],
|
|
"paddings": [0, 0, 0],
|
|
"ceil_mode": False,
|
|
"exclusive": True,
|
|
"data_format": "NCHW",
|
|
"pooling_type": "max",
|
|
"global_pooling": True,
|
|
"adaptive": False,
|
|
"padding_algorithm": "EXPLICIT",
|
|
}
|
|
self.program_config = {"feed_list": ["x"]}
|
|
self.min_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.opt_shape = {"x": [1, 3, 5, 5, 5]}
|
|
self.max_shape = {"x": [2, 3, 5, 5, 5]}
|
|
|
|
def test_fp32_trt_result(self):
|
|
self.check_trt_result()
|
|
|
|
def test_fp16_trt_result(self):
|
|
self.check_trt_result(precision_mode="fp16")
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|