conversion

Perceval provides helper methods to convert the three types of results of the Sampler (namely the BSDistribution, BSCount and BSSamples) into each other.

>>> import perceval as pcvl
>>> distribution = pcvl.BSDistribution({pcvl.BasicState([1, 0]): 0.4, pcvl.BasicState([0, 1]): 0.6})
>>> print(pcvl.probs_to_sample_count(distribution, count=1000))  # Sampling noise is applied; results may vary
{
  |1,0>: 392
  |0,1>: 608
}

Note that for methods converting from probabilities, passing a kwarg with a count is mandatory. This can either be:

count, in which case this is the number of resulting samples.
max_shots and/or max_samples, in which case the one defined or the minimum of the two is the number of resulting samples.

Conversion code reference

class perceval.utils.conversion.ConversionHelper

Helper to makes the conversion between probs, samples and sample_count easy

static convert_to(target, original, **kwargs)

Parameters:

target (str) – the target name (probs, samples, sample_count)
original (BSDistribution | BSSamples | BSCount) – The BSSamples, BSCount or BSDistribution to convert
kwargs – The parameters to pass to the conversion (shots, samples)

Return type:

BSDistribution | BSSamples | BSCount

Returns:

perceval.utils.conversion.probs_to_sample_count(probs, **kwargs)

Convert a measured state probability distribution to a state count.

This conversion artificially adds random sampling noise, following a normal law, to the result.

Parameters:

probs (BSDistribution) – the distribution to convert
count – (int) – The final number of samples to generate. Can be None if either of the remaining kwargs is defined.
max_shots – (int) – If both max_shots and max_samples are given, then the minimum of the two will be used. Else, the one defined will be used if count is not given.
max_samples – (int) – See max_shots.

Return type:

BSCount

Returns:

the state count

perceval.utils.conversion.probs_to_samples(probs, **kwargs)

Convert a measured state probability distribution to a chronological list of samples

Parameters:

probs (BSDistribution) – the distribution to convert
count – (int) – The final number of samples to generate. Can be None if either of the remaining kwargs is defined.
max_shots – (int) – If both max_shots and max_samples are given, then the minimum of the two will be used. Else, the one defined will be used if count is not given.
max_samples – (int) – See max_shots.

Return type:

BSSamples

Returns:

the sample list

perceval.utils.conversion.sample_count_to_probs(sample_count)

Convert a state count to a state probability distribution

Parameters:: sample_count (BSCount) – the state count
Return type:: BSDistribution
Returns:: the state probability distribution

perceval.utils.conversion.sample_count_to_samples(sample_count, **kwargs)

Convert a state count to a chronological list of samples, by randomly sampling on the count

Parameters:

sample_count (BSCount) – the state count
count – (int) – The final number of samples to generate. Can be None to deduce it from the number of samples in the BSCount.
max_shots – (int) – If both max_shots and max_samples are given, then the minimum of the two will be used. Else, the one defined will be used if count is not given.
max_samples – (int) – See max_shots.

Return type:

BSSamples

Returns:

the sample list

perceval.utils.conversion.samples_to_probs(sample_list)

Convert a chronological measured sample list to a state distribution

Parameters:: sample_list (BSSamples) – the list to convert
Return type:: BSDistribution
Returns:: the state distribution

perceval.utils.conversion.samples_to_sample_count(sample_list)

Convert a chronological measured sample list to a state count

Parameters:: sample_list (BSSamples) – the list to convert
Return type:: BSCount
Returns:: the state count