The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

first time user and first time neural network learner clarification needed

bdlagnobdlagno Member Posts: 3 Contributor I
edited October 2019 in Help
hey guys,

this is my first time using Rapidminer, I been watching youtube videos on how to import data and etc.

I am having a hard time understanding ImprovedNeuralNet TAP.
So I ran the data and i got this window(see bellow)

My question is
===========
1)What is this ImprovedNeuralNet tap telling me??
2) what is output telling me??
3) what is node??
4 ) what is threshold??

5) ***Finally is the output regression written like this ??
Y= -2.544(BOOK) +1.648(Tea T-1) +1.425(Tea T-2)-1.794(Tea T-3)
==========================
==========================

ImprovedNeuralNet
Hidden 1
========

Node 1 (Sigmoid)
----------------
BOOK: -2.843
Tea T-1 : -0.918
Tea T-2 : -1.112
Tea T-3 : 2.088
Bias: -3.673

Node 2 (Sigmoid)
----------------
BOOK : 2.197
TEA T-1 : -1.435
TEA T-2 : -0.601
TEA T-3 : 0.797
Bias: -2.739

Node 3 (Sigmoid)
----------------
BOOK: -0.387
SPY T-1 : -2.403
SPY T-2 : 2.422
SPY T-3 : -4.690
Bias: -5.121

Node 4 (Sigmoid)
----------------
BOOK: -2.066
TEA T-1 : -0.799
TEA T-2 : 0.651
TEA T-3 : -1.014
Bias: -2.397


Output
======

Regression (Linear)
-------------------
Node 1: -2.544
Node 2: 1.648
Node 3: 1.425
Node 4: -1.794
Threshold: 0.196

Answers

  • mafern76mafern76 Member Posts: 45 Contributor II
    Hi bdlagno,

    You should read some bibliography about Neural Networks, and also maybe look into your results, you want to evaluate a model not just read the architecture and extract a formula.

    You should give more details about your project, what are you trying to predict, and how to implement the model. Did you do some data preparation on your attributes?

    On 5), no that's not what you need, a NN is more complex than a simple regression. That I believe is just the final regression of the NN.

    Cheers.
Sign In or Register to comment.