DropoutMaskComponent Class Reference

#include <nnet-general-component.h>

Inheritance diagram for DropoutMaskComponent:
Collaboration diagram for DropoutMaskComponent:

Public Member Functions

virtual int32 InputDim () const
 Returns input-dimension of this component. More...
 
virtual int32 OutputDim () const
 Returns output-dimension of this component. More...
 
virtual std::string Info () const
 Returns some text-form information about this component, for diagnostics. More...
 
virtual void InitFromConfig (ConfigLine *cfl)
 Initialize, from a ConfigLine object. More...
 
 DropoutMaskComponent ()
 
 DropoutMaskComponent (const DropoutMaskComponent &other)
 
virtual std::string Type () const
 Returns a string such as "SigmoidComponent", describing the type of the object. More...
 
virtual int32 Properties () const
 Return bitmask of the component's properties. More...
 
virtual void * Propagate (const ComponentPrecomputedIndexes *indexes, const CuMatrixBase< BaseFloat > &in, CuMatrixBase< BaseFloat > *out) const
 Propagate function. More...
 
virtual void Backprop (const std::string &debug_info, const ComponentPrecomputedIndexes *indexes, const CuMatrixBase< BaseFloat > &, const CuMatrixBase< BaseFloat > &, const CuMatrixBase< BaseFloat > &out_deriv, void *memo, Component *to_update, CuMatrixBase< BaseFloat > *in_deriv) const
 Backprop function; depending on which of the arguments 'to_update' and 'in_deriv' are non-NULL, this can compute input-data derivatives and/or perform model update. More...
 
virtual void Read (std::istream &is, bool binary)
 Read function (used after we know the type of the Component); accepts input that is missing the token that describes the component type, in case it has already been consumed. More...
 
virtual void Write (std::ostream &os, bool binary) const
 Write component to stream. More...
 
virtual ComponentCopy () const
 Copies component (deep copy). More...
 
virtual void GetInputIndexes (const MiscComputationInfo &misc_info, const Index &output_index, std::vector< Index > *desired_indexes) const
 This function only does something interesting for non-simple Components. More...
 
virtual bool IsComputable (const MiscComputationInfo &misc_info, const Index &output_index, const IndexSet &input_index_set, std::vector< Index > *used_inputs) const
 This function only does something interesting for non-simple Components, and it exists to make it possible to manage optionally-required inputs. More...
 
void SetDropoutProportion (BaseFloat p)
 
- Public Member Functions inherited from RandomComponent
void ResetGenerator ()
 
void SetTestMode (bool test_mode)
 
 RandomComponent ()
 
 RandomComponent (const RandomComponent &other)
 
- Public Member Functions inherited from Component
virtual void StoreStats (const CuMatrixBase< BaseFloat > &in_value, const CuMatrixBase< BaseFloat > &out_value, void *memo)
 This function may store stats on average activation values, and for some component types, the average value of the derivative of the nonlinearity. More...
 
virtual void ZeroStats ()
 Components that provide an implementation of StoreStats should also provide an implementation of ZeroStats(), to set those stats to zero. More...
 
virtual void ReorderIndexes (std::vector< Index > *input_indexes, std::vector< Index > *output_indexes) const
 This function only does something interesting for non-simple Components. More...
 
virtual ComponentPrecomputedIndexesPrecomputeIndexes (const MiscComputationInfo &misc_info, const std::vector< Index > &input_indexes, const std::vector< Index > &output_indexes, bool need_backprop) const
 This function must return NULL for simple Components. More...
 
virtual void Scale (BaseFloat scale)
 This virtual function when called on – an UpdatableComponent scales the parameters by "scale" when called by an UpdatableComponent. More...
 
virtual void Add (BaseFloat alpha, const Component &other)
 This virtual function when called by – an UpdatableComponent adds the parameters of another updatable component, times some constant, to the current parameters. More...
 
virtual void DeleteMemo (void *memo) const
 This virtual function only needs to be overwritten by Components that return a non-NULL memo from their Propagate() function. More...
 
 Component ()
 
virtual ~Component ()
 

Private Member Functions

const DropoutMaskComponentoperator= (const DropoutMaskComponent &other)
 

Private Attributes

int32 output_dim_
 
BaseFloat dropout_proportion_
 
bool continuous_
 

Additional Inherited Members

- Static Public Member Functions inherited from Component
static ComponentReadNew (std::istream &is, bool binary)
 Read component from stream (works out its type). Dies on error. More...
 
static ComponentNewComponentOfType (const std::string &type)
 Returns a new Component of the given type e.g. More...
 
- Protected Attributes inherited from RandomComponent
CuRand< BaseFloatrandom_generator_
 
bool test_mode_
 

Detailed Description

Definition at line 707 of file nnet-general-component.h.

Constructor & Destructor Documentation

◆ DropoutMaskComponent() [1/2]

◆ DropoutMaskComponent() [2/2]

Definition at line 1403 of file nnet-general-component.cc.

1404  :
1405  output_dim_(other.output_dim_),
1406  dropout_proportion_(other.dropout_proportion_),
1407  continuous_(other.continuous_) { }

Member Function Documentation

◆ Backprop()

virtual void Backprop ( const std::string &  debug_info,
const ComponentPrecomputedIndexes indexes,
const CuMatrixBase< BaseFloat > &  in_value,
const CuMatrixBase< BaseFloat > &  out_value,
const CuMatrixBase< BaseFloat > &  out_deriv,
void *  memo,
Component to_update,
CuMatrixBase< BaseFloat > *  in_deriv 
) const
inlinevirtual

Backprop function; depending on which of the arguments 'to_update' and 'in_deriv' are non-NULL, this can compute input-data derivatives and/or perform model update.

Parameters
[in]debug_infoThe component name, to be printed out in any warning messages.
[in]indexesA pointer to some information output by this class's PrecomputeIndexes function (will be NULL for simple components, i.e. those that don't do things like splicing).
[in]in_valueThe matrix that was given as input to the Propagate function. Will be ignored (and may be empty) if Properties()&kBackpropNeedsInput == 0.
[in]out_valueThe matrix that was output from the Propagate function. Will be ignored (and may be empty) if Properties()&kBackpropNeedsOutput == 0
[in]out_derivThe derivative at the output of this component.
[in]memoThis will normally be NULL, but for component types that set the flag kUsesMemo, this will be the return value of the Propagate() function that corresponds to this Backprop() function. Ownership of any pointers is not transferred to the Backprop function; DeleteMemo() will be called to delete it.
[out]to_updateIf model update is desired, the Component to be updated, else NULL. Does not have to be identical to this. If supplied, you can assume that to_update->Properties() & kUpdatableComponent is nonzero.
[out]in_derivThe derivative at the input of this component, if needed (else NULL). If Properties()&kBackpropInPlace, may be the same matrix as out_deriv. If Properties()&kBackpropAdds, this is added to by the Backprop routine, else it is set. The component code chooses which mode to work in, based on convenience.

Implements Component.

Definition at line 738 of file nnet-general-component.h.

References DistributeComponent::Copy(), DistributeComponent::Read(), and DistributeComponent::Write().

745  { }

◆ Copy()

Component * Copy ( ) const
virtual

Copies component (deep copy).

Implements Component.

Definition at line 1500 of file nnet-general-component.cc.

References DropoutMaskComponent::DropoutMaskComponent().

1500  {
1501  return new DropoutMaskComponent(*this);
1502 }

◆ GetInputIndexes()

virtual void GetInputIndexes ( const MiscComputationInfo misc_info,
const Index output_index,
std::vector< Index > *  desired_indexes 
) const
inlinevirtual

This function only does something interesting for non-simple Components.

For a given index at the output of the component, tells us what indexes are required at its input (note: "required" encompasses also optionally-required things; it will enumerate all things that we'd like to have). See also IsComputable().

Parameters
[in]misc_infoThis argument is supplied to handle things that the framework can't very easily supply: information like which time indexes are needed for AggregateComponent, which time-indexes are available at the input of a recurrent network, and so on. We will add members to misc_info as needed.
[in]output_indexThe Index at the output of the component, for which we are requesting the list of indexes at the component's input.
[out]desired_indexesA list of indexes that are desired at the input. are to be written to here. By "desired" we mean required or optionally-required.

The default implementation of this function is suitable for any SimpleComponent; it just copies the output_index to a single identical element in input_indexes.

Reimplemented from Component.

Definition at line 753 of file nnet-general-component.h.

755  {
756  desired_indexes->clear(); // requires no inputs.
757  }

◆ Info()

std::string Info ( ) const
virtual

Returns some text-form information about this component, for diagnostics.

Starts with the type of the component. E.g. "SigmoidComponent dim=900", although most components will have much more info.

Reimplemented from Component.

Definition at line 1390 of file nnet-general-component.cc.

References ConstantComponent::Type().

1390  {
1391  std::ostringstream stream;
1392  stream << Type()
1393  << ", output-dim=" << output_dim_
1394  << ", dropout-proportion=" << dropout_proportion_;
1395  if (continuous_)
1396  stream << ", continuous=true";
1397  return stream.str();
1398 }
virtual std::string Type() const
Returns a string such as "SigmoidComponent", describing the type of the object.

◆ InitFromConfig()

void InitFromConfig ( ConfigLine cfl)
virtual

Initialize, from a ConfigLine object.

Parameters
[in]cflA ConfigLine containing any parameters that are needed for initialization. For example: "dim=100 param-stddev=0.1"

Implements Component.

Definition at line 1504 of file nnet-general-component.cc.

References DropoutMaskComponent::continuous_, DropoutMaskComponent::dropout_proportion_, ConfigLine::GetValue(), KALDI_ASSERT, DropoutMaskComponent::output_dim_, and RandomComponent::test_mode_.

1504  {
1505  output_dim_ = 0;
1506  bool ok = cfl->GetValue("output-dim", &output_dim_);
1507  KALDI_ASSERT(ok && output_dim_ > 0);
1508  dropout_proportion_ = 0.5;
1509  cfl->GetValue("dropout-proportion", &dropout_proportion_);
1510  continuous_ = false;
1511  cfl->GetValue("continuous", &continuous_);
1512  test_mode_ = false;
1513  cfl->GetValue("test-mode", &test_mode_);
1514 }
#define KALDI_ASSERT(cond)
Definition: kaldi-error.h:169

◆ InputDim()

virtual int32 InputDim ( ) const
inlinevirtual

Returns input-dimension of this component.

Implements Component.

Definition at line 711 of file nnet-general-component.h.

References DistributeComponent::output_dim_.

◆ IsComputable()

virtual bool IsComputable ( const MiscComputationInfo misc_info,
const Index output_index,
const IndexSet input_index_set,
std::vector< Index > *  used_inputs 
) const
inlinevirtual

This function only does something interesting for non-simple Components, and it exists to make it possible to manage optionally-required inputs.

It tells the user whether a given output index is computable from a given set of input indexes, and if so, says which input indexes will be used in the computation.

Implementations of this function are required to have the property that adding an element to "input_index_set" can only ever change IsComputable from false to true, never vice versa.

Parameters
[in]misc_infoSome information specific to the computation, such as minimum and maximum times for certain components to do adaptation on; it's a place to put things that don't easily fit in the framework.
[in]output_indexThe index that is to be computed at the output of this Component.
[in]input_index_setThe set of indexes that is available at the input of this Component.
[out]used_inputsIf this is non-NULL and the output is computable this will be set to the list of input indexes that will actually be used in the computation.
Returns
Returns true iff this output is computable from the provided inputs.

The default implementation of this function is suitable for any SimpleComponent: it just returns true if output_index is in input_index_set, and if so sets used_inputs to vector containing that one Index.

Reimplemented from Component.

Definition at line 762 of file nnet-general-component.h.

765  {
766  if (used_inputs) used_inputs->clear();
767  return true;
768  }

◆ operator=()

const DropoutMaskComponent& operator= ( const DropoutMaskComponent other)
private

◆ OutputDim()

virtual int32 OutputDim ( ) const
inlinevirtual

Returns output-dimension of this component.

Implements Component.

Definition at line 713 of file nnet-general-component.h.

References Component::Info(), DistributeComponent::InitFromConfig(), and DistributeComponent::output_dim_.

◆ Propagate()

void * Propagate ( const ComponentPrecomputedIndexes indexes,
const CuMatrixBase< BaseFloat > &  in,
CuMatrixBase< BaseFloat > *  out 
) const
virtual

Propagate function.

Parameters
[in]indexesA pointer to some information output by this class's PrecomputeIndexes function (will be NULL for simple components, i.e. those that don't do things like splicing).
[in]inThe input to this component. Num-columns == InputDim().
[out]outThe output of this component. Num-columns == OutputDim(). Note: output of this component will be added to the initial value of "out" if Properties()&kPropagateAdds != 0; otherwise the output will be set and the initial value ignored. Each Component chooses whether it is more convenient implementation-wise to add or set, and the calling code has to deal with it.
Returns
Normally returns NULL, but may return a non-NULL value for components which have the flag kUsesMemo set. This value will be passed into the corresponding Backprop routine.

Implements Component.

Definition at line 1409 of file nnet-general-component.cc.

References CuVectorBase< Real >::Add(), CuMatrixBase< Real >::Add(), CuMatrixBase< Real >::ApplyHeaviside(), DropoutMaskComponent::continuous_, CuMatrixBase< Real >::CopyColFromVec(), DropoutMaskComponent::dropout_proportion_, KALDI_ASSERT, kaldi::kUndefined, CuMatrixBase< Real >::NumCols(), CuMatrixBase< Real >::NumRows(), DropoutMaskComponent::output_dim_, RandomComponent::random_generator_, kaldi::RandUniform(), CuVectorBase< Real >::Scale(), CuMatrixBase< Real >::Scale(), CuMatrixBase< Real >::Set(), and RandomComponent::test_mode_.

1412  {
1413  KALDI_ASSERT(in.NumRows() == 0 && out->NumCols() == output_dim_);
1414  BaseFloat dropout_proportion = dropout_proportion_;
1415  KALDI_ASSERT(dropout_proportion >= 0.0 && dropout_proportion <= 1.0);
1416 
1417  if (dropout_proportion == 0) {
1418  out->Set(1.0);
1419  return NULL;
1420  }
1421 
1422  if (continuous_) {
1423  if (test_mode_) {
1424  out->Set(1.0);
1425  } else {
1426  const_cast<CuRand<BaseFloat>&>(random_generator_).RandUniform(out);
1427  out->Scale(dropout_proportion * 4.0);
1428  // make the expected value 1.0.
1429  out->Add(1.0 - (2.0 * dropout_proportion));
1430  }
1431  return NULL;
1432  }
1433 
1434  if (test_mode_) {
1435  out->Set(1.0 - dropout_proportion);
1436  return NULL;
1437  }
1438 
1439  const_cast<CuRand<BaseFloat>&>(random_generator_).RandUniform(out);
1440  out->Add(-dropout_proportion);
1441  out->ApplyHeaviside();
1442 
1443  if (out->NumCols() == 2 || out->NumCols() == 3) {
1444  // This is a kind of special case relevant to LSTms.
1445  // To generate data where it's never the case that both of the dimensions
1446  // for a row are zero, we generate uniformly distributed data (call this u_i),
1447  // and for row i, set (*out)(i, 0) = (0 if u_i < dropout_proportion else 1)
1448  // and (*out)(i, 1) = (0 if u_i > 1-dropout_proportion else 1)
1449  int32 num_rows = out->NumRows();
1450  // later we may make this a bit more efficient.
1451  CuVector<BaseFloat> temp(num_rows, kUndefined);
1452  const_cast<CuRand<BaseFloat>&>(random_generator_).RandUniform(&temp);
1453  temp.Add(-dropout_proportion);
1454  out->CopyColFromVec(temp, 0);
1455  temp.Add(-1.0 + (2.0 * dropout_proportion));
1456  // Now, 'temp' contains the original uniformly-distributed data plus
1457  // -(1 - dropout_proportion).
1458  temp.Scale(-1.0);
1459  out->CopyColFromVec(temp, 1);
1460  out->ApplyHeaviside();
1461  }
1462  return NULL;
1463 }
float RandUniform(struct RandomState *state=NULL)
Returns a random number strictly between 0 and 1.
Definition: kaldi-math.h:151
void ApplyHeaviside()
For each element, sets x = (x > 0 ? 1.0 : 0.0).
Definition: cu-matrix.cc:2445
void CopyColFromVec(const CuVectorBase< Real > &v, const MatrixIndexT col)
Copy vector into specific column of matrix.
Definition: cu-matrix.cc:2388
void Scale(Real value)
Definition: cu-matrix.cc:610
float BaseFloat
Definition: kaldi-types.h:29
void Add(Real value)
Definition: cu-matrix.cc:548
MatrixIndexT NumCols() const
Definition: cu-matrix.h:215
#define KALDI_ASSERT(cond)
Definition: kaldi-error.h:169
MatrixIndexT NumRows() const
Dimensions.
Definition: cu-matrix.h:214
void Set(Real value)
Definition: cu-matrix.cc:497
CuRand< BaseFloat > random_generator_

◆ Properties()

virtual int32 Properties ( ) const
inlinevirtual

Return bitmask of the component's properties.

These properties depend only on the component's type. See enum ComponentProperties.

Implements Component.

Definition at line 731 of file nnet-general-component.h.

References kaldi::nnet3::kRandomComponent, and DistributeComponent::Propagate().

◆ Read()

void Read ( std::istream &  is,
bool  binary 
)
virtual

Read function (used after we know the type of the Component); accepts input that is missing the token that describes the component type, in case it has already been consumed.

Implements Component.

Definition at line 1466 of file nnet-general-component.cc.

References DropoutMaskComponent::continuous_, DropoutMaskComponent::dropout_proportion_, kaldi::nnet3::ExpectOneOrTwoTokens(), kaldi::nnet3::ExpectToken(), DropoutMaskComponent::output_dim_, kaldi::PeekToken(), kaldi::ReadBasicType(), and RandomComponent::test_mode_.

1466  {
1467  ExpectOneOrTwoTokens(is, binary, "<DropoutMaskComponent>", "<OutputDim>");
1468  ReadBasicType(is, binary, &output_dim_);
1469  ExpectToken(is, binary, "<DropoutProportion>");
1470  ReadBasicType(is, binary, &dropout_proportion_);
1471  if (PeekToken(is, binary) == 'T') {
1472  ExpectToken(is, binary, "<TestMode>");
1473  ReadBasicType(is, binary, &test_mode_); // read test mode
1474  } else {
1475  test_mode_ = false;
1476  }
1477  if (PeekToken(is, binary) == 'C') {
1478  ExpectToken(is, binary, "<Continuous>");
1479  continuous_ = true;
1480  } else {
1481  continuous_ = false;
1482  }
1483  ExpectToken(is, binary, "</DropoutMaskComponent>");
1484 }
void ReadBasicType(std::istream &is, bool binary, T *t)
ReadBasicType is the name of the read function for bool, integer types, and floating-point types...
Definition: io-funcs-inl.h:55
void ExpectOneOrTwoTokens(std::istream &is, bool binary, const std::string &token1, const std::string &token2)
This function is like ExpectToken but for two tokens, and it will either accept token1 and then token...
Definition: nnet-parse.cc:224
static void ExpectToken(const std::string &token, const std::string &what_we_are_parsing, const std::string **next_token)
int PeekToken(std::istream &is, bool binary)
PeekToken will return the first character of the next token, or -1 if end of file.
Definition: io-funcs.cc:170

◆ SetDropoutProportion()

void SetDropoutProportion ( BaseFloat  p)
inline

◆ Type()

virtual std::string Type ( ) const
inlinevirtual

Returns a string such as "SigmoidComponent", describing the type of the object.

Implements Component.

Definition at line 730 of file nnet-general-component.h.

Referenced by GeneralDropoutComponent::Info().

730 { return "DropoutMaskComponent"; }

◆ Write()

void Write ( std::ostream &  os,
bool  binary 
) const
virtual

Write component to stream.

Implements Component.

Definition at line 1487 of file nnet-general-component.cc.

References DropoutMaskComponent::continuous_, DropoutMaskComponent::dropout_proportion_, DropoutMaskComponent::output_dim_, RandomComponent::test_mode_, kaldi::WriteBasicType(), and kaldi::WriteToken().

1487  {
1488  WriteToken(os, binary, "<DropoutMaskComponent>");
1489  WriteToken(os, binary, "<OutputDim>");
1490  WriteBasicType(os, binary, output_dim_);
1491  WriteToken(os, binary, "<DropoutProportion>");
1492  WriteBasicType(os, binary, dropout_proportion_);
1493  WriteToken(os, binary, "<TestMode>");
1494  WriteBasicType(os, binary, test_mode_);
1495  if (continuous_)
1496  WriteToken(os, binary, "<Continuous>");
1497  WriteToken(os, binary, "</DropoutMaskComponent>");
1498 }
void WriteToken(std::ostream &os, bool binary, const char *token)
The WriteToken functions are for writing nonempty sequences of non-space characters.
Definition: io-funcs.cc:134
void WriteBasicType(std::ostream &os, bool binary, T t)
WriteBasicType is the name of the write function for bool, integer types, and floating-point types...
Definition: io-funcs-inl.h:34

Member Data Documentation

◆ continuous_

◆ dropout_proportion_

◆ output_dim_


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