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=None, bias_init=None, dtype=mstype.float32)

Calculates a 1D transposed convolution, which can be regarded as Conv1d for the gradient of the input, also called deconvolution (although it is not an actual deconvolution).

The input is typically of shape \((N, C_{in}, L_{in})\), where \(N\) is batch size, \(C_{in}\) is a number of channels and \(L_{in}\) is a length of sequence.

When Conv1d and Conv1dTranspose are initialized with the same parameters, and pad_mode is set to 'pad', \(dilation * (kernel\_size - 1) - padding\) amount of zero will be padded to both sides of input, they are inverses of each other in regard to the input and output shapes in this case. However, when stride > 1, Conv1d maps multiple input shapes to the same output shape. Deconvolutional network can refer to Deconvolutional Networks.

Warning

Starting after version 2.9.0, the constructor signature will be changed to Conv1dTranspose(in_channels, out_channels, kernel_size, stride=1, padding=0, output_padding=0, groups=1, bias=True, dilation=1, padding_mode='zeros', dtype=None). Parameter group will be renamed to groups, has_bias will be renamed to bias with its default value changed from False to True, pad_mode, weight_init, and bias_init will be removed, and parameters output_padding and padding_mode will be added.

Parameters:

• in_channels (int) – The number of channels of the input tensor of the Conv1dTranspose layer.

• out_channels (int) – The number of channels of the output tensor of the Conv1dTranspose layer.

• kernel_size (int) – Specifies the width of the 1D convolution kernel.

• stride (int) – The movement stride of the 1D convolution kernel. Default: 1 .

• pad_mode (str, optional) –

  Specifies the padding mode with a padding value of 0. It can be set to: "same" , "valid" or "pad" . Default: "same" .

  • "same": Pad the input at the begin and end so that the shape of input and output are the same when stride is set to 1. The amount of padding is calculated by the operator internally. If the amount is even, it is uniformly distributed around the input, if it is odd, the excess padding goes to the right side. If this mode is set, padding must be 0.

  • "valid": No padding is applied to the input, and the output returns the maximum possible length. Extra pixels that could not complete a full stride will be discarded. If this mode is set, padding must be 0.

  • "pad": Pad the input with a specified amount. In this mode, the amount of padding at the beginning and end is determined by the padding parameter. If this mode is set, padding must be greater than or equal to 0.

• padding (int) – The number of padding on both sides of input. The value should be greater than or equal to 0. Default: 0 .

• dilation (int) – Dilation size of 1D convolution kernel. If \(k > 1\), the kernel is sampled every k elements. The value of k is in range of [1, L]. Default: 1 .

• group (int) – Splits filter into groups, in_channels and out_channels must be divisible by group. When group > 1, the Ascend platform is not supported yet. Default: 1 .

• has_bias (bool) – Whether the Conv1dTranspose layer has a bias parameter. Default: False.

• weight_init (Union[Tensor, str, Initializer, numbers.Number]) – Initialization method of weight parameter. It can be a Tensor, a string, an Initializer or a numbers.Number. When a string is specified, values from 'TruncatedNormal' , 'Normal' , 'Uniform' , 'HeUniform' and 'XavierUniform' distributions as well as constant 'One' and 'Zero' distributions are possible. Alias 'xavier_uniform' , 'he_uniform', 'ones' and 'zeros' are acceptable. Uppercase and lowercase are both acceptable. Refer to the values of Initializer for more details. Default: None , weight will be initialized using HeUniform.

• bias_init (Union[Tensor, str, Initializer, numbers.Number]) – Initialization method of bias parameter. Available initialization methods are the same as 'weight_init'. Refer to the values of Initializer for more details. Default: None , bias will be initialized using Uniform.

• dtype (mindspore.dtype) – Dtype of Parameters. Default: mstype.float32 .

Inputs:

• x (Tensor) - Tensor of shape \((N, C_{in}, L_{in})\).

Outputs:

Tensor of shape \((N, C_{out}, L_{out})\).

pad_mode is 'same': \(L_{out} = \frac{ L_{in} + \text{stride} - 1 }{ \text{stride} }\)

pad_mode is 'valid': \(L_{out} = (L_{in} - 1) \times \text{stride} + \text{dilation} \times (\text{kernel_size} - 1) + 1\)

pad_mode is 'pad': \(L_{out} = (L_{in} - 1) \times \text{stride} - 2 \times \text{padding} + \text{dilation} \times (\text{kernel_size} - 1) + 1\)

Raises:

• TypeError – If in_channels, out_channels, kernel_size, stride, padding or dilation is not an int.

• ValueError – If in_channels, out_channels, kernel_size, stride or dilation is less than 1.

• ValueError – If padding is less than 0.

• ValueError – If pad_mode is not one of 'same', 'valid', 'pad'.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> from mindspore import Tensor, nn
>>> import numpy as np
>>> net = nn.Conv1dTranspose(3, 64, 4, has_bias=False, weight_init='normal', pad_mode='pad')
>>> x = Tensor(np.ones([1, 3, 50]), mindspore.float32)
>>> output = net(x).shape
>>> print(output)
(1, 64, 53)