# Copyright 2022 ByteDance Ltd. and/or its affiliates.
#
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you 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.
from typing import Any, Tuple, List
import sys
matx = sys.modules['matx']
from .. import ASYNC, PadWithBorderOp
from ._base import BaseInterfaceClass, BatchBaseClass
from ._common import _torch_padding_mode, _assert
[文档]class Pad(BaseInterfaceClass):
[文档] def __init__(self,
padding: List[int],
fill: List[int] = [0],
padding_mode: str = "constant",
device_id: int = -2,
sync: int = ASYNC) -> None:
super().__init__(device_id=device_id, sync=sync)
self._padding: List[int] = padding
self._fill: List[int] = fill
self._padding_mode: str = padding_mode
[文档] def __call__(self, device: Any, device_str: str, sync: int) -> Any:
return PadImpl(device, device_str, self._padding, self._fill, self._padding_mode, sync)
class PadImpl(BatchBaseClass):
def __init__(self,
device: Any,
device_str: str,
padding: List[int],
fill: List[int] = [0],
padding_mode: str = "constant",
sync: int = ASYNC) -> None:
super().__init__()
self.device_str: str = device_str
if len(padding) == 1:
self.padding: List[int] = [padding[0]] * 4
elif len(padding) == 2:
self.padding: List[int] = [padding[0], padding[1], padding[0], padding[1]]
elif len(padding) == 4: # left, top, right, bottom
self.padding: List[int] = [padding[0], padding[1], padding[2], padding[3]]
else:
_assert(
False,
"Padding must be an int or a 1, 2, or 4 element tuple, not a {} element tuple".format(
len(padding)))
if len(fill) == 1:
self.fill: Tuple[int, int, int] = (fill[0], fill[0], fill[0])
else:
self.fill: Tuple[int, int, int] = (fill[0], fill[1], fill[2])
#_assert(padding_mode in _torch_padding_mode_list, "padding_mode not found")
self.padding_mode: str = _torch_padding_mode(padding_mode)
self.sync: int = sync
self.op: PadWithBorderOp = PadWithBorderOp(device, self.fill, self.padding_mode)
self.name: str = "Pad"
def _process(self, imgs: List[matx.NDArray]) -> List[matx.NDArray]:
return self.op(
imgs, [
self.padding[1]], [
self.padding[3]], [
self.padding[0]], [
self.padding[2]], sync=self.sync)
def __repr__(self) -> str:
return self.name + '(padding={0}, fill={1}, padding_mode={2}, device={3}, sync={4})'.\
format(self.padding, self.fill, self.padding_mode, self.device_str, self.sync)