0
$\begingroup$

i am trying to use neuralnet function on my data, i have two dataset one for train and one for test with 5 attributes and one output , the question is after i scale my data if i use 75% of my training data with backprop it works fine but if i use whole my training data it does not converge and gives me the famous error: algorithm did not converge in 1 of 1 repetition(s) within the stepmax

can someone advise me what should i do ? i already tried all possible ways like changing the learning rate and stepmax and even rep , but still if i use the whole train data it wont converge thank you

$\endgroup$
1
$\begingroup$

i was able to solve this problem by defining the rep number as small as 10 and then define the step max as large as 2000, and finally the problem solved. actually the reason for this problem is it means the function did not converge in one of your rep but if you define different ranges (not small ) the function will converge. sorry for not explaining enough or as expert

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.