Distributions

class sweetpea.Distribution

Abstract class for the input distribution for ContinuousFactor. A subclass of Distribution must be instantiated to represent the distribution that the factor follows when generating values dynamically. When a Distribution is used to initialize a ContinuousFactor, a corresponding distribution object is created for the factor.

sample()

Generate values based on the input distribution

Returns:

a value sampled from the input distribution

Return type:

Any

class sweetpea.UniformDistribution(low, high)

Represents a uniform distribution for ContinuousFactor, where values are sampled evenly between the specified low and high bounds.

Parameters:

  • low (float) – The lower bound of the uniform distribution.

  • high (float) – The upper bound of the uniform distribution. Must be greater than lower bound.

Return type:

Distribution

class sweetpea.GaussianDistribution(mean, sigma)

Represents a Gaussian (normal) distribution for ContinuousFactor, where values are sampled according to the specified mean and standard deviation.

Parameters:

  • mean (float) – The mean of the normal distribution.

  • sigma (float) – The standard deviation of the distribution. Must be greater than zero.

Return type:

Distribution

class sweetpea.ExponentialDistribution(rate)

Represents an exponential distribution for ContinuousFactor, where values are sampled according to the specified rate parameter (the inverse of the mean) of the exponential distribution.

Parameters:

rate (float) – The rate parameter of the exponential distribution. Must be greater than zero.

Return type:

Distribution

class sweetpea.LogNormalDistribution(mean, sigma)

Represents a log-normal distribution for ContinuousFactor, where values are sampled according to the specified mean and sigma parameters.

Parameters:

  • mean (float) – The mean of the underlying normal distribution.

  • sigma (float) – The standard deviation of the underlying normal distribution. Must be greater than zero.

Return type:

Distribution

class sweetpea.CustomDistribution(func, dependents=[])

Represents a custom distribution for ContinuousFactor, where values are generated through calling a custom function, func.

CustomDistribution can also accept an optional argument dependents, which is the list of factors the current ContinuousFactor depends on. dependents can only contain DiscreteFactor or ContinuousFactor in the design.

When dependents is empty, the ContinuousFactor is a non-derived factor and func should not require additional inputs to generate values.

When dependents is not empty, the ContinuousFactor becomes a derived factor. In such cases, func requires additional inputs to generate values. Since the values of the dependent factors are not known when intializing CustomDistribution, the values of these factors, factor_values, need to be passed to func at runtime. See CustomDistribution.sample() for more details.

Parameters:

  • func (Callable) – A custom function to generate values.

  • dependents (List[Factor]) – A list of factors that ContinuousFactor depends on.

Return type:

Distribution

sample(factor_values: List[Any] = [])

Generate values using the custom func.

Parameters:

factor_values (List[Any]) – optional factor values when dependents is not empty for CustomDistribution. The size of factor_values must be the same as dependents. factor_values contains values for factors in dependents at runtime.

Returns:

a value generated by calling func

Return type:

Any