BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
husseinmazaar
Quartz | Level 8

Dear Members,

I have used a Neural Networks in EM for classification problem with default options. I want to know the default options of:

Hidden/Target layer combination function.

Hidden/Target layer activation function.

Thanks

1 ACCEPTED SOLUTION

Accepted Solutions
RalphAbbey
SAS Employee

If you are using the Neural Network Node in Enterprise Miner with the defaults it will run the neural network with the Multilayer Perceptron architecture and 3 hidden units. You can check this in the Node Results, by looking at the log and seeing if the proc neural call uses "arch MLP Hidden = 3"

For Multilayer Perceptron (default setting in Enterprise Miner):

Hidden layer activation function     = tanh

Hidden layer combination function = linear

The default target layer activation and combination functions are a bit more complicated, as it depends on the level of the target variable itself:

Target layer activation function      = identity (if LEVEL = INTERVAL)

                                                 = logistic (if LEVEL = ORDINAL)

                                                 = mlogistic (if LEVEL = NOMINAL)

Target layer combination function  = linear

I hope this helps. Most of this information came from the proc neural documentation if you care for further reading.

View solution in original post

1 REPLY 1
RalphAbbey
SAS Employee

If you are using the Neural Network Node in Enterprise Miner with the defaults it will run the neural network with the Multilayer Perceptron architecture and 3 hidden units. You can check this in the Node Results, by looking at the log and seeing if the proc neural call uses "arch MLP Hidden = 3"

For Multilayer Perceptron (default setting in Enterprise Miner):

Hidden layer activation function     = tanh

Hidden layer combination function = linear

The default target layer activation and combination functions are a bit more complicated, as it depends on the level of the target variable itself:

Target layer activation function      = identity (if LEVEL = INTERVAL)

                                                 = logistic (if LEVEL = ORDINAL)

                                                 = mlogistic (if LEVEL = NOMINAL)

Target layer combination function  = linear

I hope this helps. Most of this information came from the proc neural documentation if you care for further reading.

sas-innovate-2024.png

Available on demand!

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

 

Register now!

How to choose a machine learning algorithm

Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 1 reply
  • 2504 views
  • 0 likes
  • 2 in conversation