From 84858f9a416e8a07a39654c6227d2529fbb84fbf Mon Sep 17 00:00:00 2001
From: Philipp Rudiger <prudiger@anaconda.com>
Date: Wed, 24 Jul 2024 16:40:59 +0200
Subject: [PATCH] Add correlation transform (#614)

---
 lumen/transforms/base.py | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/lumen/transforms/base.py b/lumen/transforms/base.py
index 8603b613c..41f53c50a 100644
--- a/lumen/transforms/base.py
+++ b/lumen/transforms/base.py
@@ -841,6 +841,31 @@ def apply(self, table: DataFrame) -> DataFrame:
         return table.dropna(**kwargs)
 
 
+class Corr(Transform):
+    """
+    ``Corr`` computes pairwise correlation of columns, excluding NA/null values.
+    """
+
+    method = param.Selector(default='pearson', objects=[
+        'pearson', 'kendall', 'spearman'], doc="""
+        Method of correlation.""")
+
+    min_periods = param.Integer(default=1, doc="""
+        Minimum number of observations required per pair of columns
+        to have a valid result. Currently only available for Pearson
+        and Spearman correlation.""")
+
+    numeric_only = param.Boolean(default=False, doc="""
+        Include only `float`, `int` or `boolean` data.""")
+
+    transform_type: ClassVar[str] = 'corr'
+
+    def apply(self, table: DataFrame) -> DataFrame:
+        return table.corr(
+            method=self.method, min_periods=self.min_periods, numeric_only=self.numeric_only
+        )
+
+
 class project_lnglat(Transform):
     """
     `project_lnglat` projects the given longitude/latitude columns to Web Mercator.