paralleldomain.data_lab.config.distribution¶

class Bucket(*, proto=None, string_value=None, float_value=None, int32_value=None, probability=None)¶

A discrete value with an associated probability.

Parameters:

string_value (Optional[str]) – string_value
float_value (Optional[float]) – float_value
int32_value (Optional[int]) – int32_value
probability (Optional[float]) – probability
proto (Optional[pd_distributions_pb2.Bucket]) –

string_value¶: String name of the bucket.

float_value¶: Float value of the bucket.

int32_value¶: Integer value of the bucket.

probability¶: The probability associated with the bucket.

class CategoricalDistribution(*, proto=None, buckets=None)¶

Discrete probability distribution that describes the results of a random variable that can take on one of K possible categories, with the probability of each category (bucket) separately specified.

Parameters:

buckets (Optional[List[Bucket]]) – buckets
proto (Optional[pd_distributions_pb2.CategoricalDistribution]) –

buckets¶: The buckets which make up the distribution.

sample(random_seed)¶

Sample from a categorical distribution with buckets.

Parameters:: random_seed (int) – Random seed.
Return type:: Bucket
Returns:: Sampled value.

class ConstantDistribution(*, proto=None, string_value=None, float_value=None, int32_value=None)¶

A single sample distribution.

Parameters:

string_value (Optional[str]) – string_value
float_value (Optional[float]) – float_value
int32_value (Optional[int]) – int32_value
proto (Optional[pd_distributions_pb2.ConstantDistribution]) –

string_value¶: The string value of the distribution.

float_value¶: The float value of the distribution.

int32_value¶: The integer value of the distribution.

class ContinousUniformDistribution(*, proto=None, min=None, max=None)¶

A uniform distribution with constant probability.

Parameters:

min (Optional[float]) – min
max (Optional[float]) – max
proto (Optional[pd_distributions_pb2.ContinousUniformDistribution]) –

min¶: Minimum value of the continuous normal distribution.

max¶: Maximum value of the continuous normal distribution.

sample(random_seed)¶

Sample from a continuous uniform distribution with min and max.

Parameters:: random_seed (int) – Random seed.
Return type:: Union[float, int]
Returns:: Sampled value.

class Distribution(*, proto=None, normal=None, truncated_normal=None, uniform_continous=None, constant=None, uniform_discrete=None, categorical=None)¶

A wrapper which contains a single distribution.

Parameters:

normal (Optional[NormalDistribution]) – normal
truncated_normal (Optional[TrucatedNormalDistribution]) – truncated_normal
uniform_continous (Optional[ContinousUniformDistribution]) – uniform_continous
constant (Optional[ConstantDistribution]) – constant
uniform_discrete (Optional[DiscreteUniformDistribution]) – uniform_discrete
categorical (Optional[CategoricalDistribution]) – categorical
proto (Optional[pd_distributions_pb2.Distribution]) –

normal¶: A normal distribution.

truncated_normal¶: A truncated normal distribution.

uniform_continous¶: A continuous uniform distribution.

constant¶: A constant distribution.

uniform_discrete¶: A discrete uniform distribution.

categorical¶: A categorical distribution.

sample(random_seed)¶

Sample from a distribution.

Parameters:: random_seed (int) – Random seed.
Return type:: Union[str, int, float, Bucket]
Returns:: Sampled value.

class NormalDistribution(*, proto=None, mean=None, variance=None)¶

Normal distribution.

Parameters:

mean (Optional[float]) – mean
variance (Optional[float]) – variance
proto (Optional[pd_distributions_pb2.NormalDistribution]) –

mean¶: Mean of the normal distribution.

variance¶: Variance of the normal distribution

sample(random_seed)¶

Sample from a normal distribution with mean and variance.

Parameters:: random_seed (int) – Random seed.
Return type:: Union[float, int]
Returns:: Sampled value.