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
Neural Network Toplogy & Optimize Neural Network
Learner III
Hello, i would like to optimize a neural net using Genetic Algorithms, there is a operator named Optimize Parameters (Evolucionary) where i can use genetic algorithms approaches to optimize the neural net, but it wont let me use genetic algorithms to determine the best number of hidden layers and their respective values. Is there a way I can optimize my neural net with genetic algorithms approaches but optimizing the number of hidden layers and values?
Thanks.
Tagged:
0
Best Answer
-
IngoRM
Employee-RapidMiner, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,751 
RM Founder
This has been discussed and successfully closed here: http://community.rapidminer.com/t5/RapidMiner-Studio/Neural-Network-Toplogy-amp-Optimize-Parameters-Grid/m-p/31407#U31407
3
Answers
With the standard RapidMiner Neural Network I haven't yet found a way to do this. You can download our free DL4J extension which will work with the optimize parameter operators.
DL4J Extension Direct Link
Please note however, that it is still early Beta so it might throw an error for some parameter settings so make sure you keep an eye on it. We're working on this.
John
RapidMiner China - RapidMiner sales & consulting in China.
www.rapidminerchina.com
I knew that i found a way together with @MariusHelf to do this. But i do not remember how anymore..
~Martin
Dortmund, Germany