TD_OrdinalEncodingFit InputTable: OrdEnc_titanic_train
passenger |
survived |
pclass |
name |
gender |
age |
sibsp |
parch |
ticket |
fare |
cabin |
embarked |
873 |
0 |
1 |
Carlsson; Mr. Frans Olof |
male |
33 |
0 |
0 |
695 |
5.0000 |
B51 B53 B55 |
S |
631 |
1 |
1 |
Barksworth; Mr. Algernon Henry Wilson |
male |
80 |
0 |
0 |
27042 |
30.0000 |
A23 |
S |
97 |
0 |
1 |
Goldschmidt; Mr. George B |
male |
71 |
0 |
0 |
PC 17754 |
34.6542 |
A5 |
C |
1000 |
0 |
1 |
|
|
71 |
0 |
0 |
|
34.6542 |
|
|
488 |
0 |
1 |
Kent; Mr. Edward Austin |
male |
58 |
0 |
0 |
11771 |
29.7000 |
B37 |
C |
505 |
1 |
1 |
Maioni; Miss. Roberta |
female |
16 |
0 |
0 |
110152 |
86.5000 |
B79 |
S |
Example: TD_OrdinalEncodingFit SQL Call Using Auto Approach
SELECT * FROM TD_OrdinalEncodingFit (
ON ordinal_titanic_dataset AS InputTable
OUT PERMANENT TABLE OutputTable (ordinal_titanic_fit_output)
USING
TargetColumn('name','gender','ticket','cabin','embarked')
Approach ('AUTO')
StartValue (5, 10, 15, 0, -5)
DefaultValue (-1, -10, -15, 20, 0)
) AS dt ORDER BY 1,3;
Multiple column support for TD_OrdinalEncoding, TD_OneHotEncoding, and TD_Histogram is available in release 17.20.03.07 and later. If you are using an older version, the TargetColumn argument accepts only one column.
TD_OrdinalEncodingTransform Output Table Using Auto Approach
TD_ColumnName_ORDFIT |
TD_Category_ORDFIT |
TD_Value_ORDFIT |
TD_Index_ORDFIT |
name |
gender |
ticket |
cabin |
embarked |
cabin |
|
0 |
3 |
None |
None |
None |
None |
None |
cabin |
A23 |
1 |
3 |
None |
None |
None |
None |
None |
cabin |
A5 |
2 |
3 |
None |
None |
None |
None |
None |
cabin |
B37 |
3 |
3 |
None |
None |
None |
None |
None |
cabin |
B51 B53 B55 |
4 |
3 |
None |
None |
None |
None |
None |
cabin |
B79 |
5 |
3 |
None |
None |
None |
None |
None |
cabin |
TD_CATEGORY_COUNT |
6 |
-1 |
None |
None |
None |
None |
None |
cabin |
TD_OTHER_CATEGORY |
20 |
-2 |
None |
None |
None |
None |
None |
embarked |
|
-5 |
4 |
None |
None |
None |
None |
None |
embarked |
C |
-4 |
4 |
None |
None |
None |
None |
None |
embarked |
S |
-3 |
4 |
None |
None |
None |
None |
None |
embarked |
TD_OTHER_CATEGORY |
0 |
-2 |
None |
None |
None |
None |
None |
embarked |
TD_CATEGORY_COUNT |
3 |
-1 |
None |
None |
None |
None |
None |
name |
TD_OTHER_CATEGORY |
-1 |
-2 |
None |
None |
None |
None |
None |
name |
|
5 |
0 |
None |
None |
None |
None |
None |
name |
Barksworth; Mr. Algernon Henry Wilson |
6 |
0 |
None |
None |
None |
None |
None |
name |
TD_CATEGORY_COUNT |
6 |
-1 |
None |
None |
None |
None |
None |
name |
Carlsson; Mr. Frans Olof |
7 |
0 |
None |
None |
None |
None |
None |
name |
Goldschmidt; Mr. George B |
8 |
0 |
None |
None |
None |
None |
None |
name |
Kent; Mr. Edward Austin |
9 |
0 |
None |
None |
None |
None |
None |
name |
Maioni; Miss. Roberta |
10 |
0 |
None |
None |
None |
None |
None |
gender |
TD_OTHER_CATEGORY |
-10 |
-2 |
None |
None |
None |
None |
None |
gender |
TD_CATEGORY_COUNT |
3 |
-1 |
None |
None |
None |
None |
None |
gender |
|
10 |
1 |
None |
None |
None |
None |
None |
gender |
female |
11 |
1 |
None |
None |
None |
None |
None |
gender |
male |
12 |
1 |
None |
None |
None |
None |
None |
ticket |
TD_OTHER_CATEGORY |
-15 |
-2 |
None |
None |
None |
None |
None |
ticket |
TD_CATEGORY_COUNT |
6 |
-1 |
None |
None |
None |
None |
None |
ticket |
|
15 |
2 |
None |
None |
None |
None |
None |
ticket |
110152 |
16 |
2 |
None |
None |
None |
None |
None |
ticket |
11771 |
17 |
2 |
None |
None |
None |
None |
None |
ticket |
27042 |
18 |
2 |
None |
None |
None |
None |
None |
ticket |
695 |
19 |
2 |
None |
None |
None |
None |
None |
ticket |
PC 17754 |
20 |
2 |
None |
None |
None |
None |
None |
Example: TD_OrdinalEncodingTransform SQL Call Using Output from TD_OrdinalEncodingFit
SELECT * FROM TD_OrdinalEncodingTransform (
ON ordinal_titanic_dataset AS InputTable
ON ordinal_titanic_fit_output as FitTable Dimension
USING
Accumulate ('passenger')
) AS dt ORDER BY 1;
TD_OrdinalEncodingTransform Output
passenger name gender ticket cabin embarked
--------- ---- ------ ------ ----- --------
97 8 12 20 2 -4
488 9 12 17 3 -4
505 10 11 16 5 -3
631 6 12 18 1 -3
873 7 12 19 4 -3
1000 5 10 15 0 -5