MLP with backpropogation

hira.ishtiaq · 07-13-2008, 11:32 PM

i am designing MLP with back propogation algo with rhoe as a learning factor i need information regarding it how it helps in learning m using it for classifying heart sounds should i give all wts by myself aur take it from data randomly

hira.ishtiaq · 07-13-2008, 11:34 PM

do reply me urgently as i have to appy in ma final year project

**Salem** · 07-13-2008, 11:45 PM

http://www.catb.org/~esr/faqs/smart-questions.html
Bzzzt - urgency
Bzzzt - not using google

abachler · 07-22-2008, 02:42 PM

not familiar with rhoe and a google didnt turn up much on the first page, sorry not my project not my responsibility to do the research, so perhaps you could elaborate and/or spell out what it stands for. Never assume that Everyone, even experts, are familiar with every acronym you use in your little niche of the field. I have worked for 10 different companies and all 10 used different terms for the same things.

As for the initial weights, it is prefered to use random initial weights.

As for training, that depends on your training methodology. Do you use localized training methods, or apply the global error . Using global error, there is no need to sequence the trainign examples. If you use localized training, i.e. test then apply the error, you need to randomly shuffle the examples after every epoch

07-13-2008, 11:32 PM	#1
hira.ishtiaq Registered User Join Date: Jul 2008 Posts: 2	MLP with backpropogation i am designing MLP with back propogation algo with rhoe as a learning factor i need information regarding it how it helps in learning m using it for classifying heart sounds should i give all wts by myself aur take it from data randomly
hira.ishtiaq is offline

07-13-2008, 11:34 PM	#2
hira.ishtiaq Registered User Join Date: Jul 2008 Posts: 2	do reply me urgently as i have to appy in ma final year project
hira.ishtiaq is offline

07-13-2008, 11:45 PM	#3
Salem and the hat of vanishing Join Date: Aug 2001 Location: The edge of the known universe Posts: 21,214	http://www.catb.org/~esr/faqs/smart-questions.html Bzzzt - urgency Bzzzt - not using google __________________ If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut. Up to 8Mb PlusNet broadband from only £5.99 a month!
Salem is offline

07-22-2008, 02:42 PM	#4
abachler Rampaging 35 Stone Welsh Join Date: Apr 2007 Posts: 2,927	not familiar with rhoe and a google didnt turn up much on the first page, sorry not my project not my responsibility to do the research, so perhaps you could elaborate and/or spell out what it stands for. Never assume that Everyone, even experts, are familiar with every acronym you use in your little niche of the field. I have worked for 10 different companies and all 10 used different terms for the same things. As for the initial weights, it is prefered to use random initial weights. As for training, that depends on your training methodology. Do you use localized training methods, or apply the global error . Using global error, there is no need to sequence the trainign examples. If you use localized training, i.e. test then apply the error, you need to randomly shuffle the examples after every epoch __________________ He is free, you say. Ah! That is his misfortune… These men… [have] the most terrible, the most imperious of masters, that is, need. … They must therefore find someone to hire them, or die of hunger. Is that to be free? - Simon Linguet Last edited by abachler; 07-22-2008 at 02:45 PM.
abachler is offline

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode