Caffe2 - C++ API
A deep learning, cross platform ML framework
Public Member Functions
torch::nn::Conv1dImpl Class Reference

Applies convolution over a 1-D input. More...

#include <conv.h>

Inheritance diagram for torch::nn::Conv1dImpl:
torch::nn::ConvImpl< 1, Conv1dImpl > torch::nn::Cloneable< Conv1dImpl > torch::nn::Module

Public Member Functions

Tensor forward (const Tensor &input)
 
- Public Member Functions inherited from torch::nn::ConvImpl< 1, Conv1dImpl >
 ConvImpl (int64_t input_channels, int64_t output_channels, ExpandingArray< D > kernel_size)
 
 ConvImpl (ConvOptions< D > options)
 
void reset () override
 reset() must perform initialization of all members with reference semantics, most importantly parameters, buffers and submodules. More...
 
void pretty_print (std::ostream &stream) const override
 Pretty prints the Conv{1,2,3}d module into the given stream.
 
- Public Member Functions inherited from torch::nn::Cloneable< Conv1dImpl >
std::shared_ptr< Moduleclone (const optional< Device > &device=nullopt) const override
 Performs a recursive "deep copy" of the Module, such that all parameters and submodules in the cloned module are different from those in the original module. More...
 
- Public Member Functions inherited from torch::nn::Module
 Module (std::string name)
 Tells the base Module about the name of the submodule.
 
 Module ()
 Constructs the module without immediate knowledge of the submodule's name. More...
 
const std::string & name () const noexcept
 Returns the name of the Module. More...
 
void apply (const ModuleApplyFunction &function)
 Applies the function to the Module and recursively to every submodule. More...
 
void apply (const ConstModuleApplyFunction &function) const
 Applies the function to the Module and recursively to every submodule. More...
 
void apply (const NamedModuleApplyFunction &function, const std::string &name_prefix=std::string())
 Applies the function to the Module and recursively to every submodule. More...
 
void apply (const ConstNamedModuleApplyFunction &function, const std::string &name_prefix=std::string()) const
 Applies the function to the Module and recursively to every submodule. More...
 
void apply (const ModulePointerApplyFunction &function) const
 Applies the function to the Module and recursively to every submodule. More...
 
void apply (const NamedModulePointerApplyFunction &function, const std::string &name_prefix=std::string()) const
 Applies the function to the Module and recursively to every submodule. More...
 
std::vector< Tensorparameters (bool recurse=true) const
 Returns the parameters of this Module and if recurse is true, also recursively of every submodule. More...
 
OrderedDict< std::string, Tensornamed_parameters (bool recurse=true) const
 Returns an OrderedDict with the parameters of this Module along with their keys, and if recurse is true also recursively of every submodule. More...
 
std::vector< Tensorbuffers (bool recurse=true) const
 Returns the buffers of this Module and if recurse is true, also recursively of every submodule. More...
 
OrderedDict< std::string, Tensornamed_buffers (bool recurse=true) const
 Returns an OrderedDict with the buffers of this Module along with their keys, and if recurse is true also recursively of every submodule. More...
 
std::vector< std::shared_ptr< Module > > modules (bool include_self=true) const
 Returns the submodules of this Module (the entire submodule hierarchy) and if include_self is true, also inserts a shared_ptr to this module in the first position. More...
 
OrderedDict< std::string, std::shared_ptr< Module > > named_modules (const std::string &name_prefix=std::string(), bool include_self=true) const
 Returns an OrderedDict of he submodules of this Module (the entire submodule hierarchy) and thei keys, and if include_self is true, also inserts a shared_ptr to this module in the first position. More...
 
std::vector< std::shared_ptr< Module > > children () const
 Returns the direct submodules of this Module.
 
OrderedDict< std::string, std::shared_ptr< Module > > named_children () const
 Returns an OrderedDict of the direct submodules of this Module and their keys. More...
 
virtual void train (bool on=true)
 Enables "training" mode.
 
void eval ()
 Calls train(false) to enable "eval" mode. More...
 
virtual bool is_training () const noexcept
 True if the module is in training mode. More...
 
virtual void to (torch::Device device, torch::Dtype dtype, bool non_blocking=false)
 Recursively casts all parameters to the given dtype and device. More...
 
virtual void to (torch::Dtype dtype, bool non_blocking=false)
 Recursively casts all parameters to the given dtype. More...
 
virtual void to (torch::Device device, bool non_blocking=false)
 Recursively moves all parameters to the given device. More...
 
virtual void zero_grad ()
 Recursively zeros out the grad value of each registered parameter.
 
template<typename ModuleType >
ModuleType::ContainedType * as () noexcept
 Attempts to cast this Module to the given ModuleType. More...
 
template<typename ModuleType >
const ModuleType::ContainedType * as () const noexcept
 Attempts to cast this Module to the given ModuleType. More...
 
template<typename ModuleType , typename = torch::detail::disable_if_module_holder_t<ModuleType>>
ModuleType * as () noexcept
 Attempts to cast this Module to the given ModuleType. More...
 
template<typename ModuleType , typename = torch::detail::disable_if_module_holder_t<ModuleType>>
const ModuleType * as () const noexcept
 Attempts to cast this Module to the given ModuleType. More...
 
virtual void save (serialize::OutputArchive &archive) const
 Serializes the Module into the given OutputArchive.
 
virtual void load (serialize::InputArchive &archive)
 Deserializes the Module from the given InputArchive.
 

Additional Inherited Members

- Public Types inherited from torch::nn::Module
using ModuleApplyFunction = std::function< void(Module &)>
 
using ConstModuleApplyFunction = std::function< void(const Module &)>
 
using NamedModuleApplyFunction = std::function< void(const std::string &, Module &)>
 
using ConstNamedModuleApplyFunction = std::function< void(const std::string &, const Module &)>
 
using ModulePointerApplyFunction = std::function< void(const std::shared_ptr< Module > &)>
 
using NamedModulePointerApplyFunction = std::function< void(const std::string &, const std::shared_ptr< Module > &)>
 
- Data Fields inherited from torch::nn::ConvImpl< 1, Conv1dImpl >
ConvOptions< Doptions
 The options with which this Module was constructed.
 
Tensor weight
 The learned kernel (or "weight").
 
Tensor bias
 The learned bias. Only defined if the with_bias option was true.
 
- Protected Member Functions inherited from torch::nn::Module
Tensorregister_parameter (std::string name, Tensor tensor, bool requires_grad=true)
 Registers a parameter with this Module. More...
 
Tensorregister_buffer (std::string name, Tensor tensor)
 Registers a buffer with this Module. More...
 
template<typename ModuleType >
std::shared_ptr< ModuleType > register_module (std::string name, std::shared_ptr< ModuleType > module)
 Registers a submodule with this Module. More...
 
template<typename ModuleType >
std::shared_ptr< ModuleType > register_module (std::string name, ModuleHolder< ModuleType > module_holder)
 Registers a submodule with this Module. More...
 

Detailed Description

Applies convolution over a 1-D input.

See https://pytorch.org/docs/master/nn.html#torch.nn.Conv1d to learn about the exact behavior of this module.

Definition at line 109 of file conv.h.


The documentation for this class was generated from the following files: