25.10.2016 Views

SAP HANA Predictive Analysis Library (PAL)

sap_hana_predictive_analysis_library_pal_en

sap_hana_predictive_analysis_library_pal_en

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Name Data Type Default Value Description<br />

TRY_NUM_MAX Integer Variables number Upper bound of try number<br />

where the search for optimal<br />

try number will stop. The<br />

value must be greater than<br />

or equal to the value of<br />

TRY_NUM_MIN and less<br />

than or equal to variable<br />

number of the data.<br />

Output Tables<br />

Table 153:<br />

Table Column Column Data Type Description<br />

Evaluation Results 1st column Integer ID<br />

2nd column Varchar or nvarchar Evaluation measure name<br />

3rd column Varchar or nvarchar Evaluation measure content<br />

Selected Parameters 1st column Integer ID<br />

2nd column Varchar or nvarchar Parameter name<br />

3rd column Varchar or nvarchar Parameter value<br />

4th column Integer Topic ID<br />

Function Specific Output Table(s)<br />

Function specific Function specific Function specific<br />

Note<br />

Serialized confusion matrix is output in the evaluation results table in JSON format as shown below. The<br />

"Count"array is filled by row. It may be split into more rows if the length exceeds 5000.<br />

{<br />

}<br />

ConfusionMatrix:{<br />

ActualCategories:[],<br />

Count:[],<br />

PredictedCategories:[]<br />

}<br />

For example,<br />

{"ConfusionMatrix":{"ActualCategories":<br />

["setosa","versicolor","virginica"],"Count":<br />

[50,0,0,0,47,3,0,4,46],"PredictedCategories":<br />

["setosa","versicolor","virginica"]}} represents<br />

<strong>SAP</strong> <strong>HANA</strong> <strong>Predictive</strong> <strong>Analysis</strong> <strong>Library</strong> (<strong>PAL</strong>)<br />

<strong>PAL</strong> Functions P U B L I C 207

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!