User Community
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: 

  • JMP Security Notice: CVE-2026-109618 and CVE-2026-109619

    JMP promptly initiated a review of its products, services, and relevant infrastructure for exposure to CVE-2026-109618 and CVE-2026-109619.

    Our assessment of JMP products has not identified exposure to these vulnerabilities. Where remediation was required within supporting environments, appropriate security updates and controls have been applied and validated.

    We have identified no evidence of successful exploitation or unauthorized access to JMP systems or customer information associated with these vulnerabilities.

    JMP continues to monitor the issue and relevant vendor guidance. No action is required by JMP customers at this time.

  • Don’t miss your chance to experience Discovery Summit Europe at our best available rate! Early bird registration through 31 Oct.
  • Graph Builder Toolbar streamlines interactive graphing. Creates shortcuts on Report and GraphBuilder Helpers toolbars. Download and install the extension.

JMP Wish List

We want to hear your ideas for improving JMP. Share them here.
Choose Language Hide Translation Bar

Neural Network third layer

Status: Acknowledged Submitted by on ‎04-14-2020 04:18 PM
5 Comments (5 New)

I know there is a lot that can be done with two layers, but being limited with two layers even in JMP Pro is not a good idea. As a customer, I can get any layer in a free package of Python and I don't know why I should not be able to get the same with JMP Pro. Number of layers in the best option would be selective, otherwise, at least provide three layers (Input, middle, output layers).

5 Comments
craigwb
Level I
‎10-13-2020 03:48 PM
‎10-13-2020 03:48 PM

I'm actually advocating for a minimum of 5 layers, though 3 layers would be a good start!  I want to do nonlinear data compression,  in a like form of autoencoders-decoders.  The middle layer would have the lowest count, and I think in general 5 layers for this kind of data compression makes sense.

hnasiri
Level II
‎10-13-2020 04:02 PM
‎10-13-2020 04:02 PM

Probably they can include a drop down menu to select how many layers are required and leave it to the user. 

Sarah-Sylvestre
Level V
‎12-08-2022 02:31 PM
‎12-08-2022 02:31 PM
Status changed to: Acknowledged

Hi @hnasiri, thank you for your suggestion! We have captured your request and will take it under consideration.

FN
Level VI
‎02-18-2023 08:39 AM
‎02-18-2023 08:39 AM

What's the status of this? I agree we should have 5 rather than 3.

JMP 17 broke the possibility to use the same input as output, so we are out of luck even using one layer. In principle, they will revert it.


MikeD_Anderson
Staff
Staff
‎06-27-2024 11:10 AM
‎06-27-2024 11:10 AM

JMP 18 brings the Torch Add-in that provides access to these larger NN's: https://community.jmp.com/t5/JMP-Add-Ins/Torch-Deep-Learning-Add-In-for-JMP-Pro/ta-p/762296