File weight_sampling.hpp

Functions

template<typename Iterator>
static inline void sampleWeightsWithoutReplacementViaTrackingSelection(BitWeightVector &weightVector, Iterator iterator, uint32 numTotal, uint32 numSamples, RNG &rng)

Randomly selects numSamples out of numTotal elements and sets their weights to 1, while the remaining weights are set to 0, by using a set to keep track of the elements that have already been selected. This method is suitable if numSamples is much smaller than numTotal.

Template Parameters:

Iterator – The type of the iterator that provides random access to the indices of the available elements to sample from

Parameters:

  • weightVector – A reference to an object of type BitWeightVector the weights should be written to

  • iterator – An iterator that provides random access to the indices of the available elements to sample from

  • numTotal – The total number of available elements to sample from

  • numSamples – The number of weights to be set to 1

  • rng – A reference to an object of type RNG, implementing the random number generator to be used

template<typename Iterator>
static inline void sampleWeightsWithoutReplacementViaPool(BitWeightVector &weightVector, Iterator iterator, uint32 numTotal, uint32 numSamples, RNG &rng)

Randomly selects numSamples out of numTotal elements and sets their weights to 1, while the remaining weights are set to 0, by using a pool, i.e., an array, to keep track of the elements that have not been selected yet.

Template Parameters:

Iterator – The type of the iterator that provides random access to the indices of the available elements to sample from

Parameters:

  • weightVector – A reference to an object of type BitWeightVector the weights should be written to

  • iterator – An iterator that provides random access to the indices of the available elements to sample from

  • numTotal – The total number of available elements to sample from

  • numSamples – The number of weights to be set to 1

  • rng – A reference to an object of type RNG, implementing the random number generator to be used

template<typename Iterator>
static inline void sampleWeightsWithoutReplacement(BitWeightVector &weightVector, Iterator iterator, uint32 numTotal, uint32 numSamples, RNG &rng)

Randomly selects numSamples out of numTotal elements and sets their weights to 1, while the remaining weights are set to 0. The method that is used internally is chosen automatically, depending on the ratio numSamples / numTotal.

Template Parameters:

Iterator – The type of the iterator that provides random access to the indices of the available elements to sample from

Parameters:

  • weightVector – A reference to an object of type BitWeightVector the weights should be written to

  • iterator – An iterator that provides random access to the indices of the available elements to sample from

  • numTotal – The total number of available elements to sample from

  • numSamples – The number of weights to be set to 1

  • rng – A reference to an object of type RNG, implementing the random number generator to be used