比较与torch.nn.ConvTranspose1d的功能差异

torch.nn.ConvTranspose1d

class torch.nn.ConvTranspose1d(
    in_channels,
    out_channels,
    kernel_size,
    stride=1,
    padding=0,
    output_padding=0,
    groups=1,
    bias=True,
    dilation=1,
    padding_mode='zeros'
)(input) -> Tensor

更多内容详见torch.nn.ConvTranspose1d。

mindspore.nn.Conv1dTranspose

class mindspore.nn.Conv1dTranspose(
    in_channels,
    out_channels,
    kernel_size,
    stride=1,
    pad_mode='same',
    padding=0,
    dilation=1,
    group=1,
    has_bias=False,
    weight_init='normal',
    bias_init='zeros'
)(x) -> Tensor

更多内容详见mindspore.nn.Conv1dTranspose。

差异对比

PyTorch：计算一维转置卷积，可以视为Conv1d对输入求梯度，也称为反卷积（实际不是真正的反卷积）。输入的shape通常是\((N,C_{in}, L_{in})\)，其中\(N\)是batch size，\(C\)是空间维度，\(L\)是序列的长度。输出的shape为\((N,C_{out},L_{out})\)，其中\(L_{out}=(L_{in}−1)×stride−2×padding+dilation×(kernel\_size−1)+output\_padding+1\)

MindSpore：MindSpore此API实现功能与PyTorch基本一致，新增了填充模式参数”pad_mode”，当”pad_mode” = “pad”时与PyTorch默认方式相同，利用weight_init 和bias_init 参数可以配置初始化方式。

分类	子类	PyTorch	MindSpore	差异
参数	参数1	in_channels	in_channels	-
	参数2	out_channels	out_channels	-
	参数3	kernel_size	kernel_size	-
	参数4	stride	stride	-
	参数5	padding	padding	-
	参数6	output_padding	-	通常在stride > 1的前提下使用，用来调整output shapes。例如，通常将padding设置为(kernel_size - 1)/2，此时设置output_padding = (stride - 1)可确保input shapes/output shapes = stride，MindSpore无此参数
	参数7	groups	group	功能一致，参数名不同
	参数8	bias	has_bias	PyTorch默认为True，MindSpore默认为False
	参数9	dilation	dilation	-
	参数10	padding_mode	-	数值填充模式，只支持”zeros”即填充0。MindSpore无此参数，但默认填充0
	参数11	-	pad_mode	指定填充模式。可选值为”same”、”valid”、”pad”，在”same”和”valid”模式下，padding必须设置为0，默认为”same”，PyTorch无此参数
	参数12	-	weight_init	权重参数的初始化方法。可为Tensor，str，Initializer或numbers.Number。当使用str时，可选”TruncatedNormal”，”Normal”，”Uniform”，”HeUniform”和”XavierUniform”分布以及常量”One”和”Zero”分布的值，默认为”normal”，PyTorch无此参数
	参数13	-	bias_init	偏置参数的初始化方法。初始化方法与”weight_init”相同，默认为”zeros”，PyTorch无此参数
输入	单输入	input	x	功能一致，参数名不同

代码示例1

两API都是实现一维转置卷积运算，使用时需先进行实例化。当PyTorch将output_padding设为0，MindSpore将pad_mode设为”pad”时，输出宽度\(L_{out}=(L_{in}−1)×stride−2×padding+dilation×(kernel\_size−1)+1\)。PyTorch中利用net.weight.data = torch.ones()的方式将权重初始化为1，MindSpore直接设置参数weight_init = “ones”。

# PyTorch
import torch
from torch import tensor
import torch.nn as nn
import numpy as np

k = 4
x_ = np.ones([1, 3, 50])
x = tensor(x_, dtype=torch.float32)
net = nn.ConvTranspose1d(3, 64, kernel_size=k, stride=1, padding=0, output_padding=0, bias=False)
net.weight.data = torch.ones(3, 64, k)
output = net(x).detach().numpy()
print(output.shape)
# (1, 64, 53)


# MindSpore
import mindspore as ms
import mindspore.nn as nn
import numpy as np

k = 4
x_ = np.ones([1, 3, 50])
x = ms.Tensor(x_, ms.float32)
net = nn.Conv1dTranspose(3, 64, kernel_size=k, weight_init='ones', pad_mode='pad')
output = net(x)
print(output.shape)
# (1, 64, 53)

代码示例2

为使输出的宽度与输入整除stride后的值相同，PyTorch中设置output_padding = stride - 1，padding设置为(kernel_size - 1)/2。MindSpore则设置pad_mode = “same”，同时padding = 0。

# PyTorch
import torch
from torch import tensor
import torch.nn as nn
import numpy as np

k = 5
s = 3
x_ = np.ones([1, 3, 50])
x = tensor(x_, dtype=torch.float32)
net = nn.ConvTranspose1d(3, 64, kernel_size=k, stride=s, padding=(k-1)//2, output_padding=s-1, bias=False)
net.weight.data = torch.ones(3, 64, k)
output = net(x).detach().numpy()
print(output.shape)
# (1, 64, 150)


# MindSpore
import mindspore as ms
import mindspore.nn as nn
import numpy as np

k = 5
s = 3
x_ = np.ones([1, 3, 50])
x = ms.Tensor(x_, ms.float32)
net = nn.Conv1dTranspose(3, 64, kernel_size=k, stride=s, weight_init='ones', pad_mode='same', padding=0)
output = net(x)
print(output.shape)
# (1, 64, 150)

代码示例3

若不在原有图像上做任何填充，在stride>1的情况下可能舍弃一部分数据，输出宽度\(L_{out}=(L_{in}−1)×stride+dilation×(kernel\_size−1)+1\)。在PyTorch中将padding和output_padding设为0，MindSpore中设置pad_mode = “valid”。

# PyTorch
import torch
from torch import tensor
import torch.nn as nn
import numpy as np

k = 5
s = 3
x_ = np.ones([1, 3, 50])
x = tensor(x_, dtype=torch.float32)
net = nn.ConvTranspose1d(3, 64, kernel_size=k, stride=s, padding=0, output_padding=0, bias=False)
net.weight.data = torch.ones(3, 64, k)
output = net(x).detach().numpy()
print(output.shape)
# (1, 64, 152)


# MindSpore
import mindspore as ms
import mindspore.nn as nn
import numpy as np

k = 5
s = 3
x_ = np.ones([1, 3, 50])
x = ms.Tensor(x_, ms.float32)
net = nn.Conv1dTranspose(3, 64, kernel_size=k, stride=s, weight_init='ones', pad_mode='valid', padding=0)
output = net(x)
print(output.shape)
# (1, 64, 152)