卷积层神经网路学习conv2d

卷积层神经网路学习conv2d

torch.nn.functional.conv2d

torch.nn.functional.conv2d(input, weight, bias=None, stride=1, padding=0, dilation=1, groups=1) → Tensor

Applies a 2D convolution over an input image composed of several input planes.

This operator supports TensorFloat32.

See Conv2d for details and output shape.

NOTE

In some circumstances when given tensors on a CUDA device and using CuDNN, this operator may select a nondeterministic algorithm to increase performance. If this is undesirable, you can try to make the operation deterministic (potentially at a performance cost) by setting torch.backends.cudnn.deterministic = True. See Reproducibility for more information.

NOTE

This operator supports complex data types i.e. complex32, complex64, complex128.

Parameters

• input – input tensor of shape (minibatch,in_channels,𝑖𝐻,𝑖𝑊)(minibatch,in_channels,iH,iW)

• weight – filters of shape (out_channels,in_channelsgroups,𝑘𝐻,𝑘𝑊)(out_channels,groupsin_channels​,kH,kW)

• bias – optional bias tensor of shape (out_channels)(out_channels). Default: None

• stride – the stride of the convolving kernel. Can be a single number or a tuple (sH, sW). Default: 1

• padding –

  implicit paddings on both sides of the input. Can be a string {‘valid’, ‘same’}, single number or a tuple (padH, padW). Default: 0 padding='valid' is the same as no padding. padding='same' pads the input so the output has the same shape as the input. However, this mode doesn’t support any stride values other than 1.

  WARNING

  For padding='same', if the weight is even-length and dilation is odd in any dimension, a full pad() operation may be needed internally. Lowering performance.

• dilation – the spacing between kernel elements. Can be a single number or a tuple (dH, dW). Default: 1

• groups – split input into groups, both in_channelsin_channels and out_channelsout_channels should be divisible by the number of groups. Default: 1

Examples:

>>> # With square kernels and equal stride
>>> filters = torch.randn(8, 4, 3, 3)
>>> inputs = torch.randn(1, 4, 5, 5)
>>> F.conv2d(inputs, filters, padding=1)