Updated percentile.py to allow user defined percentiles, and separati…

…ng calculated values into separate columns (LLNL#103) * Changed percentiles.py to take in custom values, and split calulated values into seperate columns * Updated doc string for function * Updated percentile.py to pass black and flake8 * Updating percentile unit testing * Updated formatting for unit testing * code cleanup * percentiles: add test for percentiles=None --------- Co-authored-by: Stephanie Brink <[email protected]>
TauferLab · Mar 6, 2024 · a23f568 · a23f568
1 parent 22df8ef
commit a23f568
Show file tree

Hide file tree

Showing 2 changed files with 42 additions and 3 deletions.
diff --git a/thicket/stats/percentiles.py b/thicket/stats/percentiles.py
@@ -9,8 +9,7 @@
 
 
 def percentiles(thicket, columns=None, percentiles=[0.25, 0.50, 0.75]):
-    """
-    Calculate the q-th percentile for each node in the performance data table.
+    """Calculate the q-th percentile for each node in the performance data table.
 
     Designed to take in a thicket, and append one or more columns to the aggregated
     statistics table for the q-th percentile calculation for each node. Each percentile
@@ -76,7 +75,7 @@ def percentiles(thicket, columns=None, percentiles=[0.25, 0.50, 0.75]):
                     and column_to_append not in thicket.statsframe.exc_metrics
                 ):
                     thicket.statsframe.exc_metrics.append(column_to_append)
-                # check to see if inclusive metric
+                # check inclusive metrics
                 elif (
                     column in thicket.inc_metrics
                     and column_to_append not in thicket.statsframe.inc_metrics

diff --git a/thicket/tests/test_stats.py b/thicket/tests/test_stats.py
@@ -305,6 +305,46 @@ def test_percentiles_columnar_join(thicket_axis_columns):
         combined_th, columns=[(idx, "Min time/rank")], percentiles=[0.4]
     )
 
+    assert (
+        idx,
+        "Min time/rank_percentiles_25",
+    ) in combined_th.statsframe.dataframe.columns
+    assert (
+        idx,
+        "Min time/rank_percentiles_50",
+    ) in combined_th.statsframe.dataframe.columns
+    assert (
+        idx,
+        "Min time/rank_percentiles_75",
+    ) in combined_th.statsframe.dataframe.columns
+    assert (
+        idx,
+        "Min time/rank_percentiles_25",
+    ) in combined_th.statsframe.exc_metrics + combined_th.statsframe.inc_metrics
+    assert (
+        idx,
+        "Min time/rank_percentiles_50",
+    ) in combined_th.statsframe.exc_metrics + combined_th.statsframe.inc_metrics
+    assert (
+        idx,
+        "Min time/rank_percentiles_75",
+    ) in combined_th.statsframe.exc_metrics + combined_th.statsframe.inc_metrics
+
+    assert (
+        idx,
+        "Min time/rank_percentiles_25",
+    ) in combined_th.statsframe.show_metric_columns()
+    assert (
+        idx,
+        "Min time/rank_percentiles_50",
+    ) in combined_th.statsframe.show_metric_columns()
+    assert (
+        idx,
+        "Min time/rank_percentiles_75",
+    ) in combined_th.statsframe.show_metric_columns()
+
+    th.percentiles(combined_th, columns=[(idx, "Min time/rank")], percentiles=[0.4])
+
     assert (
         idx,
         "Min time/rank_percentiles_40",