Pytorch上下采样函数--interpolate用法


Posted in Python onJuly 07, 2020

最近用到了上采样下采样操作,pytorch中使用interpolate可以很轻松的完成

def interpolate(input, size=None, scale_factor=None, mode='nearest', align_corners=None):
  r"""
  根据给定 size 或 scale_factor,上采样或下采样输入数据input.
  
  当前支持 temporal, spatial 和 volumetric 输入数据的上采样,其shape 分别为:3-D, 4-D 和 5-D.
  输入数据的形式为:mini-batch x channels x [optional depth] x [optional height] x width.

  上采样算法有:nearest, linear(3D-only), bilinear(4D-only), trilinear(5D-only).
  
  参数:
  - input (Tensor): input tensor
  - size (int or Tuple[int] or Tuple[int, int] or Tuple[int, int, int]):输出的 spatial 尺寸.
  - scale_factor (float or Tuple[float]): spatial 尺寸的缩放因子.
  - mode (string): 上采样算法:nearest, linear, bilinear, trilinear, area. 默认为 nearest.
  - align_corners (bool, optional): 如果 align_corners=True,则对齐 input 和 output 的角点像素(corner pixels),保持在角点像素的值. 只会对 mode=linear, bilinear 和 trilinear 有作用. 默认是 False.
  """
  from numbers import Integral
  from .modules.utils import _ntuple

  def _check_size_scale_factor(dim):
    if size is None and scale_factor is None:
      raise ValueError('either size or scale_factor should be defined')
    if size is not None and scale_factor is not None:
      raise ValueError('only one of size or scale_factor should be defined')
    if scale_factor is not None and isinstance(scale_factor, tuple)\
        and len(scale_factor) != dim:
      raise ValueError('scale_factor shape must match input shape. '
               'Input is {}D, scale_factor size is {}'.format(dim, len(scale_factor)))

  def _output_size(dim):
    _check_size_scale_factor(dim)
    if size is not None:
      return size
    scale_factors = _ntuple(dim)(scale_factor)
    # math.floor might return float in py2.7
    return [int(math.floor(input.size(i + 2) * scale_factors[i])) for i in range(dim)]

  if mode in ('nearest', 'area'):
    if align_corners is not None:
      raise ValueError("align_corners option can only be set with the "
               "interpolating modes: linear | bilinear | trilinear")
  else:
    if align_corners is None:
      warnings.warn("Default upsampling behavior when mode={} is changed "
             "to align_corners=False since 0.4.0. Please specify "
             "align_corners=True if the old behavior is desired. "
             "See the documentation of nn.Upsample for details.".format(mode))
      align_corners = False

  if input.dim() == 3 and mode == 'nearest':
    return torch._C._nn.upsample_nearest1d(input, _output_size(1))
  elif input.dim() == 4 and mode == 'nearest':
    return torch._C._nn.upsample_nearest2d(input, _output_size(2))
  elif input.dim() == 5 and mode == 'nearest':
    return torch._C._nn.upsample_nearest3d(input, _output_size(3))
  elif input.dim() == 3 and mode == 'area':
    return adaptive_avg_pool1d(input, _output_size(1))
  elif input.dim() == 4 and mode == 'area':
    return adaptive_avg_pool2d(input, _output_size(2))
  elif input.dim() == 5 and mode == 'area':
    return adaptive_avg_pool3d(input, _output_size(3))
  elif input.dim() == 3 and mode == 'linear':
    return torch._C._nn.upsample_linear1d(input, _output_size(1), align_corners)
  elif input.dim() == 3 and mode == 'bilinear':
    raise NotImplementedError("Got 3D input, but bilinear mode needs 4D input")
  elif input.dim() == 3 and mode == 'trilinear':
    raise NotImplementedError("Got 3D input, but trilinear mode needs 5D input")
  elif input.dim() == 4 and mode == 'linear':
    raise NotImplementedError("Got 4D input, but linear mode needs 3D input")
  elif input.dim() == 4 and mode == 'bilinear':
    return torch._C._nn.upsample_bilinear2d(input, _output_size(2), align_corners)
  elif input.dim() == 4 and mode == 'trilinear':
    raise NotImplementedError("Got 4D input, but trilinear mode needs 5D input")
  elif input.dim() == 5 and mode == 'linear':
    raise NotImplementedError("Got 5D input, but linear mode needs 3D input")
  elif input.dim() == 5 and mode == 'bilinear':
    raise NotImplementedError("Got 5D input, but bilinear mode needs 4D input")
  elif input.dim() == 5 and mode == 'trilinear':
    return torch._C._nn.upsample_trilinear3d(input, _output_size(3), align_corners)
  else:
    raise NotImplementedError("Input Error: Only 3D, 4D and 5D input Tensors supported"
                 " (got {}D) for the modes: nearest | linear | bilinear | trilinear"
                 " (got {})".format(input.dim(), mode))

举个例子:

x = Variable(torch.randn([1, 3, 64, 64]))
y0 = F.interpolate(x, scale_factor=0.5)
y1 = F.interpolate(x, size=[32, 32])

y2 = F.interpolate(x, size=[128, 128], mode="bilinear")

print(y0.shape)
print(y1.shape)
print(y2.shape)

这里注意上采样的时候mode默认是“nearest”,这里指定双线性插值“bilinear”

得到结果

torch.Size([1, 3, 32, 32])
torch.Size([1, 3, 32, 32])
torch.Size([1, 3, 128, 128])

补充知识:pytorch插值函数interpolate——图像上采样-下采样,scipy插值函数zoom

在训练过程中,需要对图像数据进行插值,如果此时数据是numpy数据,那么可以使用scipy中的zoom函数:

from scipy.ndimage.interpolation import zoom

def zoom(input, zoom, output=None, order=3, mode='constant', cval=0.0,
     prefilter=True):
  """
  Zoom an array.
  The array is zoomed using spline interpolation of the requested order.
  Parameters
  ----------
  %(input)s
  zoom : float or sequence
    The zoom factor along the axes. If a float, `zoom` is the same for each
    axis. If a sequence, `zoom` should contain one value for each axis.
  %(output)s
  order : int, optional
    The order of the spline interpolation, default is 3.
    The order has to be in the range 0-5.
  %(mode)s
  %(cval)s
  %(prefilter)s
  Returns
  -------
  zoom : ndarray
    The zoomed input.
  Examples
  --------
  >>> from scipy import ndimage, misc
  >>> import matplotlib.pyplot as plt
  >>> fig = plt.figure()
  >>> ax1 = fig.add_subplot(121) # left side
  >>> ax2 = fig.add_subplot(122) # right side
  >>> ascent = misc.ascent()
  >>> result = ndimage.zoom(ascent, 3.0)
  >>> ax1.imshow(ascent)
  >>> ax2.imshow(result)
  >>> plt.show()
  >>> print(ascent.shape)
  (512, 512)
  >>> print(result.shape)
  (1536, 1536)
  """
  if order < 0 or order > 5:
    raise RuntimeError('spline order not supported')
  input = numpy.asarray(input)
  if numpy.iscomplexobj(input):
    raise TypeError('Complex type not supported')
  if input.ndim < 1:
    raise RuntimeError('input and output rank must be > 0')
  mode = _ni_support._extend_mode_to_code(mode)
  if prefilter and order > 1:
    filtered = spline_filter(input, order, output=numpy.float64)
  else:
    filtered = input
  zoom = _ni_support._normalize_sequence(zoom, input.ndim)
  output_shape = tuple(
      [int(round(ii * jj)) for ii, jj in zip(input.shape, zoom)])
 
  output_shape_old = tuple(
      [int(ii * jj) for ii, jj in zip(input.shape, zoom)])
  if output_shape != output_shape_old:
    warnings.warn(
        "From scipy 0.13.0, the output shape of zoom() is calculated "
        "with round() instead of int() - for these inputs the size of "
        "the returned array has changed.", UserWarning)
 
  zoom_div = numpy.array(output_shape, float) - 1
  # Zooming to infinite values is unpredictable, so just choose
  # zoom factor 1 instead
  zoom = numpy.divide(numpy.array(input.shape) - 1, zoom_div,
            out=numpy.ones_like(input.shape, dtype=numpy.float64),
            where=zoom_div != 0)
 
  output = _ni_support._get_output(output, input,
                          shape=output_shape)
  zoom = numpy.ascontiguousarray(zoom)
  _nd_image.zoom_shift(filtered, zoom, None, output, order, mode, cval)
  return output

中的zoom函数进行插值,

但是,如果此时的数据是tensor(张量)的时候,使用zoom函数的时候需要将tensor数据转为numpy,将GPU数据转换为CPU数据等,过程比较繁琐,可以使用pytorch自带的函数进行插值操作,interpolate函数有几个参数:size表示输出大小,scale_factor表示缩放倍数,mode表示插值方式,align_corners是bool类型,表示输入和输出中心是否对齐:

from torch.nn.functional import interpolate

def interpolate(input, size=None, scale_factor=None, mode='nearest', align_corners=None):
  r"""Down/up samples the input to either the given :attr:`size` or the given
  :attr:`scale_factor`
  The algorithm used for interpolation is determined by :attr:`mode`.
  Currently temporal, spatial and volumetric sampling are supported, i.e.
  expected inputs are 3-D, 4-D or 5-D in shape.
  The input dimensions are interpreted in the form:
  `mini-batch x channels x [optional depth] x [optional height] x width`.
  The modes available for resizing are: `nearest`, `linear` (3D-only),
  `bilinear`, `bicubic` (4D-only), `trilinear` (5D-only), `area`
  Args:
    input (Tensor): the input tensor
    size (int or Tuple[int] or Tuple[int, int] or Tuple[int, int, int]):
      output spatial size.
    scale_factor (float or Tuple[float]): multiplier for spatial size. Has to match input size if it is a tuple.
    mode (str): algorithm used for upsampling:
      ``'nearest'`` | ``'linear'`` | ``'bilinear'`` | ``'bicubic'`` |
      ``'trilinear'`` | ``'area'``. Default: ``'nearest'``
    align_corners (bool, optional): Geometrically, we consider the pixels of the
      input and output as squares rather than points.
      If set to ``True``, the input and output tensors are aligned by the
      center points of their corner pixels. If set to ``False``, the input and
      output tensors are aligned by the corner points of their corner
      pixels, and the interpolation uses edge value padding for out-of-boundary values.
      This only has effect when :attr:`mode` is ``'linear'``,
      ``'bilinear'``, ``'bicubic'``, or ``'trilinear'``.
      Default: ``False``
  .. warning::
    With ``align_corners = True``, the linearly interpolating modes
    (`linear`, `bilinear`, and `trilinear`) don't proportionally align the
    output and input pixels, and thus the output values can depend on the
    input size. This was the default behavior for these modes up to version
    0.3.1. Since then, the default behavior is ``align_corners = False``.
    See :class:`~torch.nn.Upsample` for concrete examples on how this
    affects the outputs.
  .. include:: cuda_deterministic_backward.rst
  """
  from .modules.utils import _ntuple
 
  def _check_size_scale_factor(dim):
    if size is None and scale_factor is None:
      raise ValueError('either size or scale_factor should be defined')
    if size is not None and scale_factor is not None:
      raise ValueError('only one of size or scale_factor should be defined')
    if scale_factor is not None and isinstance(scale_factor, tuple)\
        and len(scale_factor) != dim:
      raise ValueError('scale_factor shape must match input shape. '
               'Input is {}D, scale_factor size is {}'.format(dim, len(scale_factor)))
 
  def _output_size(dim):
    _check_size_scale_factor(dim)
    if size is not None:
      return size
    scale_factors = _ntuple(dim)(scale_factor)
    # math.floor might return float in py2.7
 
    # make scale_factor a tensor in tracing so constant doesn't get baked in
    if torch._C._get_tracing_state():
      return [(torch.floor(input.size(i + 2) * torch.tensor(float(scale_factors[i])))) for i in range(dim)]
    else:
      return [int(math.floor(int(input.size(i + 2)) * scale_factors[i])) for i in range(dim)]
 
  if mode in ('nearest', 'area'):
    if align_corners is not None:
      raise ValueError("align_corners option can only be set with the "
               "interpolating modes: linear | bilinear | bicubic | trilinear")
  else:
    if align_corners is None:
      warnings.warn("Default upsampling behavior when mode={} is changed "
             "to align_corners=False since 0.4.0. Please specify "
             "align_corners=True if the old behavior is desired. "
             "See the documentation of nn.Upsample for details.".format(mode))
      align_corners = False
 
  if input.dim() == 3 and mode == 'nearest':
    return torch._C._nn.upsample_nearest1d(input, _output_size(1))
  elif input.dim() == 4 and mode == 'nearest':
    return torch._C._nn.upsample_nearest2d(input, _output_size(2))
  elif input.dim() == 5 and mode == 'nearest':
    return torch._C._nn.upsample_nearest3d(input, _output_size(3))
  elif input.dim() == 3 and mode == 'area':
    return adaptive_avg_pool1d(input, _output_size(1))
  elif input.dim() == 4 and mode == 'area':
    return adaptive_avg_pool2d(input, _output_size(2))
  elif input.dim() == 5 and mode == 'area':
    return adaptive_avg_pool3d(input, _output_size(3))
  elif input.dim() == 3 and mode == 'linear':
    return torch._C._nn.upsample_linear1d(input, _output_size(1), align_corners)
  elif input.dim() == 3 and mode == 'bilinear':
    raise NotImplementedError("Got 3D input, but bilinear mode needs 4D input")
  elif input.dim() == 3 and mode == 'trilinear':
    raise NotImplementedError("Got 3D input, but trilinear mode needs 5D input")
  elif input.dim() == 4 and mode == 'linear':
    raise NotImplementedError("Got 4D input, but linear mode needs 3D input")
  elif input.dim() == 4 and mode == 'bilinear':
    return torch._C._nn.upsample_bilinear2d(input, _output_size(2), align_corners)
  elif input.dim() == 4 and mode == 'trilinear':
    raise NotImplementedError("Got 4D input, but trilinear mode needs 5D input")
  elif input.dim() == 5 and mode == 'linear':
    raise NotImplementedError("Got 5D input, but linear mode needs 3D input")
  elif input.dim() == 5 and mode == 'bilinear':
    raise NotImplementedError("Got 5D input, but bilinear mode needs 4D input")
  elif input.dim() == 5 and mode == 'trilinear':
    return torch._C._nn.upsample_trilinear3d(input, _output_size(3), align_corners)
  elif input.dim() == 4 and mode == 'bicubic':
    return torch._C._nn.upsample_bicubic2d(input, _output_size(2), align_corners)
  else:
    raise NotImplementedError("Input Error: Only 3D, 4D and 5D input Tensors supported"
                 " (got {}D) for the modes: nearest | linear | bilinear | bicubic | trilinear"
                 " (got {})".format(input.dim(), mode))

以上这篇Pytorch上下采样函数--interpolate用法就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持三水点靠木。

Python 相关文章推荐
Python学习资料
Feb 08 Python
Python和Ruby中each循环引用变量问题(一个隐秘BUG?)
Jun 04 Python
python实现ipsec开权限实例
Nov 11 Python
python爬取各类文档方法归类汇总
Mar 22 Python
使用python爬虫获取黄金价格的核心代码
Jun 13 Python
详解Django 中是否使用时区的区别
Jun 14 Python
对python For 循环的三种遍历方式解析
Feb 01 Python
三步实现Django Paginator分页的方法
Jun 11 Python
python网络编程socket实现服务端、客户端操作详解
Mar 24 Python
解决Python 写文件报错TypeError的问题
Oct 23 Python
史上最详细的Python打包成exe文件教程
Jan 17 Python
python爬虫爬取某网站视频的示例代码
Feb 20 Python
pytorch随机采样操作SubsetRandomSampler()
Jul 07 #Python
pytorch加载自己的图像数据集实例
Jul 07 #Python
keras实现VGG16 CIFAR10数据集方式
Jul 07 #Python
使用darknet框架的imagenet数据分类预训练操作
Jul 07 #Python
Python调用C语言程序方法解析
Jul 07 #Python
keras实现VGG16方式(预测一张图片)
Jul 07 #Python
通过实例解析Python RPC实现原理及方法
Jul 07 #Python
You might like
解密ThinkPHP3.1.2版本之模板继承
2014/06/19 PHP
php树型类实例
2014/12/05 PHP
启用Csrf后POST数据时出现的400错误
2015/07/05 PHP
php 生成签名及验证签名详解
2016/10/26 PHP
thinkphp项目如何自定义微信分享描述内容
2017/02/20 PHP
js left,right,mid函数
2008/06/10 Javascript
JavaScript 判断判断某个对象是Object还是一个Array
2010/01/28 Javascript
jQuery中focus事件用法实例
2014/12/26 Javascript
Javascript在IE和Firefox浏览器常见兼容性问题总结
2016/08/03 Javascript
jQuery Validate插件实现表单验证
2016/08/19 Javascript
详解Jquery Easyui的验证扩展
2017/01/09 Javascript
微信小程序技巧之show内容展示,上传文件编码问题
2017/01/23 Javascript
Jquery uploadify 多余的Get请求(404错误)的解决方法
2017/01/26 Javascript
JS闭包可被利用的常见场景小结
2017/04/09 Javascript
angularjs实现猜大小功能
2017/10/23 Javascript
bootstrap paginator分页插件的两种使用方式实例详解
2017/11/14 Javascript
简单的编程0基础下Python入门指引
2015/04/01 Python
Python中的字符串类型基本知识学习教程
2016/02/04 Python
基于Linux系统中python matplotlib画图的中文显示问题的解决方法
2017/06/15 Python
Python实现matplotlib显示中文的方法详解
2018/02/06 Python
关于python写入文件自动换行的问题
2018/06/23 Python
Python 使用 attrs 和 cattrs 实现面向对象编程的实践
2019/06/12 Python
Python Process多进程实现过程
2019/10/22 Python
使用Python的Turtle库绘制森林的实例
2019/12/18 Python
编译 pycaffe时报错:fatal error: numpy/arrayobject.h没有那个文件或目录
2020/11/29 Python
Pop In A Box英国:Funko POP搪胶公仔
2019/05/27 全球购物
全球性的女装店:storets
2019/06/12 全球购物
HSRP的含义以及如何工作
2014/09/10 面试题
导游的职业规划书范文
2013/12/27 职场文书
销售职业生涯规划范文
2014/03/14 职场文书
《青海高原一株柳》教学反思
2014/04/25 职场文书
销售员试用期自我评价
2014/09/15 职场文书
幼儿园老师新年寄语2015
2014/12/08 职场文书
如何计划开一家便利店?
2019/07/31 职场文书
Go 自定义package包设置与导入操作
2021/05/06 Golang
JUnit5常用注解的使用
2021/07/02 Java/Android