Ignored Distribution
The IgnoredDistribution and IgnoredEstimator classes provide a way of either ignoring particular features or specifying distributions that are known in advance. This allows users to fix a certain component of the model (i.e. fix a component of a composite distribution).
IgnoredDistribution
- class pysp.stats.ignored.IgnoredDistribution(dist, name=None, keys=None)
IgnoredDistribution object for using IgnoredDistributions in estimation.
- dist
Distribution to be ignored.
- name
Set name for object instance.
- Type:
Optional[str]
- keys
Keys for distribution (just a place holder).
- Type:
Optional[str]
- __init__(dist, name=None, keys=None)
IgnoredDistribution object.
- Parameters:
dist (Optional[SequenceEncodableProbabilityDistribution]) – Distribution to be ignored.
name (Optional[str]) – Set name for object instance.
keys (Optional[str]) – Keys for distribution (just a place holder).
- density(x)
Evaluate the density of the IgnoredDistribution at x.
- Parameters:
x (T) – Type corresponding to attribute ‘dist’.
- Returns:
Density of attribute ‘dist’ at x
- Return type:
float
- dist_to_encoder()
Create DataSequenceEncoder object for SequenceEncodableProbabilityDistribution instance.
- Return type:
IgnoredDataEncoder- Returns:
DataSequenceEncoder
- estimator(pseudo_count=None)
Create a ParameterEstimator for corresponding SequenceEncodableProbabilityDistribution.
- Parameters:
pseudo_count (Optional[float]) – Regularize sufficient statistics in estimation step.
- Return type:
- Returns:
ParameterEstimator
- log_density(x)
Evaluate the log-density of the IgnoredDistribution at x.
- Parameters:
x (T) – Type corresponding to attribute ‘dist’.
- Returns:
log-density of attribute ‘dist’ at x.
- Return type:
float
- sampler(seed=None)
Create a DistributionSampler object for a given ProbabilityDistribution.
- Parameters:
seed (Optional[int]) – Set seed for drawing samples from distribution.
- Return type:
- seq_log_density(x)
Vectorized evaluation of the log density.
- Parameters:
x (EncodedDataSequence) – EncodedDataSequence for corresponding SequenceEncodedProbabilityDistribution.
- Return type:
ndarray- Returns:
np.ndarray
IgnoredEstimator
- class pysp.stats.ignored.IgnoredEstimator(dist=NullDistribution(name=None), pseudo_count=None, suff_stat=None, keys=None, name=None)
IgnoredEstimator object for consistency in estimation step.
- dist
Distribution to be ignored.
- pseudo_count
Place holder for consistency.
- Type:
Optional[float]
- suff_stat
Place holder for consistency.
- Type:
Optional[Any]
- keys
Place holder for consistency.
- Type:
Optional[str]
- name
Set name for object instance.
- Type:
Optional[str]
- __init__(dist=NullDistribution(name=None), pseudo_count=None, suff_stat=None, keys=None, name=None)
IgnoredEstimator object.
- Parameters:
dist (Optional[SequenceEncodableProbabilityDistribution]) – Distribution to be ignored.
pseudo_count (Optional[float]) – Place holder for consistency.
suff_stat (Optional[Any]) – Place holder for consistency.
keys (Optional[str]) – Place holder for consistency.
name (Optional[str]) – Set name for object instance.
- accumulator_factory()
Create SequenceEncodableStatisticAccumulator object.
- estimate(nobs, suff_stat)
Estimate SequenceEncodableProbabilityDistribution for sufficient statistics.
- Parameters:
nobs (Optional[float]) – Weighted number of observations.
suff_stat (Tuple[int, np.ndarray, np.ndarray, np.ndarray]) – Sufficient statistics for dirichlet distribution.
- Return type:
- Returns:
SequenceEncodableProbabilityDistribution
IgnoredSampler
- class pysp.stats.ignored.IgnoredSampler(dist, seed=None)
IgnoredSampler object for generating samples from Ignored distribution.
- dist_sampler
DistributionSampler for ignored distribution.
- Type:
- null_sampler
True if IgnoredDistribution is the NullDistribution.
- Type:
bool
- sample(size=None)
Generate samples from distribution.
- Parameters:
size (Optional[int]) – Number of samples to generate.
- Returns:
Samples from distribution.