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

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

In the training set, let N be the number of samples, X∈R N×P be the features where X i,p is the p-th feature of the<br />

i-th sample, and y∈R N be the labels where y i ∈{1, 2, ..., K} is the label of the i-th sample. The output of<br />

the training phase is denoted as W * ∈R (P+1)×K , where<br />

(p ≤ P) corresponds the weight of the p-th feature<br />

for the k-th class, and<br />

following optimization problem:<br />

corresponds the constant for the k-th class. W * is obtained by solving the<br />

Testing Phase<br />

In the testing set, let be the number of samples, ∈R N×P be the features where is the p-th feature of<br />

the i-th sample. Let be the unknown labels where is the label of the i-th<br />

sample, and be the prediction confidences of the prediction where is the confidence (likelihood) of the i-<br />

th sample. , are computed as follows,<br />

LRMCTR<br />

This is the algorithm for the training phase.<br />

Procedure Generation<br />

CALL SYS.AFLLANG_WRAPPER_PROCEDURE_CREATE (‘AFL<strong>PAL</strong>’, ‘LRMCTR’, ‘’,<br />

'', );<br />

180 P U B L I C<br />

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

<strong>PAL</strong> Functions

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

Saved successfully!

Ooh no, something went wrong!