liesel.goose.summary_m.summary
liesel.goose.summary_m.summary#
- liesel.goose.summary_m.summary(results, per_chain=True, params=None, param_indices=None, chain_indices=None, quantiles=(0.05, 0.5, 0.95), hdi_prob=0.9, round_digits=3, as_dataframe=True)[source]#
Compute summary statistics and diagnostic measures of posterior samples.
- Parameters
results (
SamplingResults
) – Result object of the sampling process. Must have a methodget_posterior_samples()
which extracts all samples from the posterior distribution.per_chain (
bool
) – IfTrue
, all statistics and diagnostics (except the Rhat value) are (default:True
) computed for each chain separately. IfFalse
, one metric is computed for each subparameter after concatenating all chains.params (
Union
[str
,list
[str
],None
]) – Names of the model parameters that are contained in the summary output. Must (default:None
) coincide with the dictionary keys of thePosition
with the posterior samples. IfNone
, all parameters are included.param_indices (
Union
[int
,Sequence
[int
],None
]) – Indices of each model parameter that are contained in the summary output. (default:None
) Selects e.g.beta_0
out of abeta
parameter vector.A single index can be specified as an integer or a sequence containing one integer. IfNone
, all subparameters are included.chain_indices (
Union
[int
,Sequence
[int
],None
]) – Indices of chains for each model subparameter that are contained in the summary (default:None
) output. Selects e.g. chain 0 and chain 2 out of multiple chains. A single index can be specified as an integer or a sequence containing one integer. IfNone
, all chains are included.quantiles (
Sequence
[float
]) – Quantiles of the posterior distribution that are contained in the summary (default:(0.05, 0.5, 0.95)
) output.hdi_prob (
float
) – Coverage level of the Highest Density Interval of the posterior distribution. (default:0.9
) Summary output contains lower and upper bound of this interval.round_digits (
int
) – Number of decimals for each float value within the summary output. (default:3
)as_dataframe (
bool
) – IfTrue
, all statistics and diagnostics are embedded into apandas
data (default:True
) frame. IfFalse
, a dictionary with the same keys asPosition
with the posterior samples is returned.
- Return type
- Returns
Dictionary if
as_dataframe
is False,pandas.DataFrame
ifas_dataframe
is True.