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.

3.2.2 Back Propagation Neural Network<br />

Neural network is a calculation model inspired by biological nervous system. The functionality of neural<br />

network is determined by its network structure and connection weights between neurons. Back propagation<br />

neural network (BPNN) is one of the very popular types for its training method called back propagation.<br />

Single Neuron<br />

A neuron takes signals from outside and transforms them to a single value as output:<br />

●<br />

The function F is called activate function.<br />

● x i ( i = 1, 2, …, n ) is the outside signal and b is a constant bias usually set to 1.<br />

●<br />

w i ( i = 0, 1, …, n ) is the weight value on each connection.<br />

Neural Network Structure<br />

A neural network consists of three parts: input layer, hidden layer(s), and output layer. Each layer owns several<br />

neurons and there are connections between layers. Signals are given in input layer and transmit through<br />

connections and layers. Finally output layer gives transformed signals.<br />

Steps<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 125

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

Saved successfully!

Ooh no, something went wrong!