Caffe2 - C++ API
A deep learning, cross platform ML framework
Public Member Functions
caffe2::GradientNotImplementedYet Struct Reference

A helper class to indicate that the gradient mechanism is not ready. More...

#include <operator_gradient.h>

Inheritance diagram for caffe2::GradientNotImplementedYet:

Public Member Functions

GradientOpsMeta Get () override
 Returns the gradient ops meta. More...
- Public Member Functions inherited from caffe2::GradientMakerBase
 GradientMakerBase (const OperatorDef &def, const vector< GradientWrapper > &g_output)
virtual bool CopyDeviceOption () const
virtual bool CopyEngine () const
virtual bool CopyArguments () const
virtual void VerifyOp () const
const OperatorDef & Def () const

Additional Inherited Members

- Static Public Member Functions inherited from caffe2::GradientMakerBase
static CaffeMap< string, string > MatchGradsToParams (const OperatorDef &op)
 Returns map that returns the parameters that the gradients are for.
- Protected Member Functions inherited from caffe2::GradientMakerBase
virtual vector< OperatorDef > GetGradientDefs ()
string I (const int i)
string O (const int i)
string GI (const int i)
string GI_I (const int i)
string GI_V (const int i)
string GO (const int i)
string GO_I (const int i)
string GO_V (const int i)
const GradientWrapperGradOut (int i)
void SetDense (const int i, const string &name)
void SetSparse (const int i, const string &indices, const string &values)
- Static Protected Member Functions inherited from caffe2::GradientMakerBase
template<class... Args>
static vector< OperatorDef > SingleGradientDef (const Args &...args)
 a helper function to allow one to create one single operator def, which is usually the case for many simple operators.
- Protected Attributes inherited from caffe2::GradientMakerBase
const OperatorDef & def_
const vector< GradientWrapper > & g_output_
vector< GradientWrapperg_input_

Detailed Description

A helper class to indicate that the gradient mechanism is not ready.

This should only be used sparsely when the gradient does exist, but we have not implemented it yet and are using this as a lazy excuse. Eventually, a gradient operator should be implemented.

Definition at line 287 of file operator_gradient.h.

Member Function Documentation

GradientOpsMeta caffe2::GradientNotImplementedYet::Get ( )

Returns the gradient ops meta.

If your gradient op generator only use standard input and output manipulations, you can simply implement GetGradientDefs() that returns vector<OperatorDef>. In that, you can call GI, GI_V and GI_I that will automatically create the gradient registration for you.

If you need to do custom gradient name registration, overload this function directly.

Reimplemented from caffe2::GradientMakerBase.

Definition at line 289 of file operator_gradient.h.

The documentation for this struct was generated from the following file: