Merge pull request #428 from UCL-CCS/feature_enseble_boot_multiple

Added EnsebleBootMultiple class to enable analysis of many statistics…

easyvvuq/analysis/__init__.py

@@ -1,5 +1,5 @@
 from .basic_stats import BasicStats
-from .ensemble_boot import EnsembleBoot
+from .ensemble_boot import EnsembleBoot, EnsembleBootMultiple
 from .sc_analysis import SCAnalysis
 from .ssc_analysis import SSCAnalysis
 from .pce_analysis import PCEAnalysis

easyvvuq/analysis/ensemble_boot.py

@@ -289,3 +289,96 @@ def analyse(self, data_frame=None):
             stat_name=self.stat_name)
 
         return results
+
+class EnsembleBootMultiple(BaseAnalysisElement):
+
+    def __init__(self, groupby=[], qoi_cols=[],
+                 stat_func=[np.mean], alpha=0.05,
+                 sample_size=None, n_boot_samples=1000,
+                 pivotal=False, stat_name=None):
+        """
+        Element to perform bootstrapping on collated simulation output.
+
+        Parameters
+        ----------
+        groupby : list or None
+            Columns to use to group the data in `analyse` method before
+            calculating stats.
+        qoi_cols : list or None
+            Columns of quantities of interest (for which stats will be
+            calculated).
+        stat_func : list[function]
+            List of statistical functions to be applied to data for bootstrapping.
+        alpha : float, default=0.05
+            Produce estimate of 100.0*(1-`alpha`) confidence interval.
+        sample_size : int
+            Size of the sample to be drawn from the input data.
+        n_boot_samples : int, default=1000
+            Number of times samples are to be drawn from the input data.
+        pivotal : bool, default=False
+            Use the pivotal method? Default to percentile method.
+        stat_name : str, default=None
+            Name to use to describe columns containing output statistic (for example
+            'mean'). If not provided, then attr '__name__' from each func is used.
+        """
+
+        if not stat_func or stat_func is None:
+            raise ValueError('stat_func cannot be empty or None')
+
+        self.groupby = groupby
+        self.qoi_cols = qoi_cols
+
+        self.stat_func = stat_func
+        self.alpha = alpha
+        self.sample_size = sample_size
+        self.n_boot_samples = n_boot_samples
+        self.pivotal = pivotal
+        self.stat_name = stat_name if stat_name is not None else [func.__name__ for func in stat_func]
+
+        self.output_type = OutputType.SUMMARY
+
+    def element_name(self):
+        """Name for this element for logging purposes"""
+        return "ensemble_boot_multiple"
+
+    def element_version(self):
+        """Version of this element for logging purposes"""
+        return "0.1"
+
+    def analyse(self, data_frame=None):
+        """Perform bootstrapping analysis on the input `data_frame`.
+
+        The data_frame is grouped according to `self.groupby` if specified and
+        analysis is performed on the columns selected in `self.qoi_cols` if set.
+
+        Parameters
+        ----------
+        data_frame : :obj:`pandas.DataFrame`
+            Summary data produced through collation of simulation output.
+
+        Returns
+        -------
+        :obj:`pandas.DataFrame`
+            Basic statistic for selected columns and groupings of data.
+        """
+
+        if data_frame is None:
+            raise RuntimeError(
+                    "This VVUQ element needs a data frame to analyse")
+        elif data_frame.empty:
+            raise RuntimeError(
+                    "No data in data frame passed to analyse element")
+        frames = []
+        for stat_func, stat_name in zip(self.stat_func, self.stat_name):
+            results = ensemble_bootstrap(
+                    data_frame,
+                    groupby=self.groupby,
+                    qoi_cols=self.qoi_cols,
+                    stat_func=stat_func,
+                    alpha=self.alpha,
+                    sample_size=self.sample_size,
+                    n_samples=self.n_boot_samples,
+                    pivotal=self.pivotal,
+                    stat_name=stat_name)
+            frames.append(results)
+        return pd.concat(frames, axis=1, keys=self.stat_name).swaplevel(0, 1, axis=1)

tests/test_ensemble_boot.py

@@ -1,5 +1,5 @@
 from easyvvuq.analysis.ensemble_boot import confidence_interval, bootstrap
-from easyvvuq.analysis.ensemble_boot import ensemble_bootstrap, EnsembleBoot
+from easyvvuq.analysis.ensemble_boot import ensemble_bootstrap, EnsembleBoot, EnsembleBootMultiple
 import os
 import numpy as np
 import pandas as pd
@@ -82,3 +82,19 @@ def test_ensemble_boot():
         'b': ['group1'] * VALUES.shape[0] + ['group2'] * VALUES.shape[0]})
     results = analysis.analyse(df)
     assert (not results.empty)
+
+def test_ensemble_boot_multiple():
+    analysis = EnsembleBootMultiple()
+    assert (analysis.element_name() == 'ensemble_boot_multiple')
+    assert (analysis.element_version() == '0.1')
+    with pytest.raises(RuntimeError):
+        analysis.analyse()
+    with pytest.raises(RuntimeError):
+        analysis.analyse(pd.DataFrame({}))
+    analysis = EnsembleBootMultiple(groupby=['b'], qoi_cols=['a'], stat_func=[np.mean, np.var, np.median])
+    df = pd.DataFrame({
+        'a': np.concatenate((VALUES, VALUES)),
+        'b': ['group1'] * VALUES.shape[0] + ['group2'] * VALUES.shape[0]})
+    results = analysis.analyse(df)
+    assert (not results.empty)
+    assert (results.values.shape == (2, 9))