Skip to content

Commit a7c428d

Browse files
committed
Add notes
1 parent fe8c062 commit a7c428d

File tree

1 file changed

+15
-1
lines changed

1 file changed

+15
-1
lines changed

articles/machine-learning/algorithm-module-reference/summarize-data.md

Lines changed: 15 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ ms.topic: reference
99

1010
author: likebupt
1111
ms.author: keli19
12-
ms.date: 09/09/2019
12+
ms.date: 01/27/2020
1313
---
1414

1515
# Summarize Data
@@ -67,6 +67,20 @@ The report from the module can include the following statistics.
6767
|**P95**|95% percentile|
6868
|**P99.5**|99.5% percentile |
6969

70+
## Technical notes
71+
72+
- For non-numeric columns, only the values for Count, Unique value count, and Missing value count are computed. For other statistics, a null value is returned.
73+
74+
- Columns that contain Boolean values are processed using these rules:
75+
76+
- When calculating Min, a logical AND is applied.
77+
78+
- When calculating Max, a logical OR is applied
79+
80+
- When computing Range, the module first checks whether the number of unique values in the column equals 2.
81+
82+
- When computing any statistic that requires floating-point calculations, values of True are treated as 1.0, and values of False are treated as 0.0.
83+
7084
## Next steps
7185

7286
See the [set of modules available](module-reference.md) to Azure Machine Learning.

0 commit comments

Comments
 (0)