cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
changqi
changqi
Level I

Why JMP Neural Networks Don't Need Data Normalization

Created: Feb 8, 2022 01:52 AM  |  Last Modified: Jun 10, 2023 1:48 PM (1491 views)

When the data value vary widely, the data standardization process must be done first. Why does the JMP neural network do not need data standardization process?

0 Kudos
3 REPLIES 3
Mark_Bailey
Staff Mark_Bailey
Staff

Re: Why JMP Neural Networks Don't Need Data Normalization

Feb 8, 2022 10:19 AM (1460 views)  |  Posted in reply to message from changqi 02-08-2022

Here is a JMP white paper about the current Neural platform. It provides a lot of information that might help.

Preview file
189 KB
0 Kudos
changqi
changqi
Level I

Re: Why JMP Neural Networks Don't Need Data Normalization

Feb 8, 2022 08:00 PM (1438 views)  |  Posted in reply to message from Mark_Bailey 02-08-2022
Get it,i read it first,thanks
0 Kudos
changqi
changqi
Level I

Re: Why JMP Neural Networks Don't Need Data Normalization

Feb 23, 2022 03:15 AM (1393 views)  |  Posted in reply to message from Mark_Bailey 02-08-2022

Hi markbailey 

i use JMP16 and read the PDF your give me before,  i can not understand why don't need to normalization or do Johnson transformation but we get the model that fit new data. 

0 Kudos