tf.compat.v1.nn.conv3d_transpose

The transpose of conv3d.

This operation is sometimes called "deconvolution" after (Zeiler et al., 2010), but is really the transpose (gradient) of conv3d rather than an actual deconvolution.

valueA 5-D Tensor of type float and shape [batch, depth, height, width, in_channels].
filterA 5-D Tensor with the same type as value and shape [depth, height, width, output_channels, in_channels]. filter's in_channels dimension must match that of value.
output_shapeA 1-D Tensor representing the output shape of the deconvolution op.
stridesA list of ints. The stride of the sliding window for each dimension of the input tensor.
paddingA string, either 'VALID' or 'SAME'. The padding algorithm. See the "returns" section of tf.nn.convolution for details.
data_formatA string, either 'NDHWC' or 'NCDHW' specifying the layout of the input and output tensors. Defaults to 'NDHWC'.
nameOptional name for the returned tensor.
inputAlias of value.
filtersAlias of filter.
dilationsAn int or list of ints that has length 1, 3 or 5, defaults to 1. The dilation factor for each dimension ofinput. If a single value is given it is replicated in the D, H and W dimension. By default the N and C dimensions are set to 1. If set to k > 1, there will be k-1 skipped cells between each filter element on that dimension. The dimension order is determined by the value of data_format, see above for details. Dilations in the batch and depth dimensions if a 5-d tensor must be 1.

A Tensor with the same type as value.

ValueErrorIf input/output depth does not match filter's shape, or if padding is other than 'VALID' or 'SAME'.

Deconvolutional Networks: Zeiler et al., 2010 (pdf)