3 #include <torch/expanding_array.h> 4 #include <torch/nn/cloneable.h> 5 #include <torch/nn/pimpl.h> 6 #include <torch/types.h> 18 int64_t input_channels,
19 int64_t output_channels,
21 input_channels_(input_channels),
22 output_channels_(output_channels),
23 kernel_size_(std::move(kernel_size)) {}
78 template <
size_t D,
typename Derived>
82 int64_t input_channels,
83 int64_t output_channels,
89 void reset()
override;
92 void pretty_print(std::ostream& stream)
const override;
122 TORCH_MODULE(Conv1d);
142 TORCH_MODULE(Conv2d);
162 TORCH_MODULE(Conv3d);
ConvOptions< D > options
The options with which this Module was constructed.
Tensor weight
The learned kernel (or "weight").
A utility class that accepts either a container of D-many values, or a single value, which is internally repeated D times.
The clone() method in the base Module class does not have knowledge of the concrete runtime type of i...
Applies convolution over a 1-D input.
Applies convolution over a 3-D input.
Tensor bias
The learned bias. Only defined if the with_bias option was true.
Applies convolution over a 2-D input.
Base class for all (dimension-specialized) convolution modules.
TORCH_ARG(int64_t, input_channels)
The number of channels the input volumes will have.
Options for a D-dimensional convolution module.