Caffe2 - C++ API
A deep learning, cross platform ML framework
Public Member Functions
torch::serialize::InputArchive Class Referencefinal

A recursive representation of tensors that can be deserialized from a file or stream. More...

#include <input-archive.h>

Public Member Functions

 InputArchive ()
 Default-constructs the InputArchive.
 
 InputArchive (InputArchive &&)=default
 
InputArchiveoperator= (InputArchive &&)=default
 
 InputArchive (InputArchive &)=delete
 
InputArchiveoperator= (InputArchive &)=delete
 
void read (const std::string &key, Tensor &tensor, bool is_buffer=false)
 Reads a tensor associated with a given key. More...
 
void read (const std::string &key, InputArchive &archive)
 Reads an InputArchive associated with a given key. More...
 
void load_from (const std::string &filename, c10::optional< torch::Device > device=c10::nullopt)
 Loads the InputArchive from a serialized representation stored in the file at filename. More...
 
void load_from (std::istream &stream, c10::optional< torch::Device > device=c10::nullopt)
 Loads the InputArchive from a serialized representation stored in the given stream. More...
 
template<typename... Ts>
void operator() (Ts &&...ts)
 Forwards all arguments to read(). More...
 

Detailed Description

A recursive representation of tensors that can be deserialized from a file or stream.

In most cases, users should not have to interact with this class, and should instead use torch::load.

Definition at line 32 of file input-archive.h.

Member Function Documentation

void torch::serialize::InputArchive::load_from ( const std::string &  filename,
c10::optional< torch::Device device = c10::nullopt 
)

Loads the InputArchive from a serialized representation stored in the file at filename.

Storage are remapped using device option. If device is not specified, the module is loaded to the original device.

Definition at line 62 of file input-archive.cpp.

void torch::serialize::InputArchive::load_from ( std::istream &  stream,
c10::optional< torch::Device device = c10::nullopt 
)

Loads the InputArchive from a serialized representation stored in the given stream.

Storage are remapped using device option. If device is not specified, the module is loaded to the original device.

Definition at line 67 of file input-archive.cpp.

template<typename... Ts>
void torch::serialize::InputArchive::operator() ( Ts &&...  ts)
inline

Forwards all arguments to read().

Useful for generic code that can be re-used for both InputArchive and OutputArchive (where operator() forwards to write()).

Definition at line 73 of file input-archive.h.

void torch::serialize::InputArchive::read ( const std::string &  key,
Tensor tensor,
bool  is_buffer = false 
)

Reads a tensor associated with a given key.

If the tensor is expected to be a buffer (not differentiable), is_buffer must be true.

Definition at line 22 of file input-archive.cpp.

void torch::serialize::InputArchive::read ( const std::string &  key,
InputArchive archive 
)

Reads an InputArchive associated with a given key.

The archive can thereafter be used for further deserialization of the nested data.

Definition at line 53 of file input-archive.cpp.


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