Closed Raphango closed 6 years ago
Thanks Angelo. I'm afraid I'm not sure why you are seeing those results. I haven't looked at this code for several years, but it was definitely working when the package was first created and should work with Python 2.
My only suggestion is to debug as the mechanics of ANFIS have not changed!
On Mon, 14 May 2018, 19:56 Ângelo Paulino, notifications@github.com wrote:
Hi!
Thanks for your work, but I'm afraid your example isn't working. The error just keeps going up and up, even raising the epochs number. There is no going down...
[image: figure_1] https://user-images.githubusercontent.com/6752809/40017274-3fb836c4-578f-11e8-8a4b-5919bc35725c.png [image: figure_2] https://user-images.githubusercontent.com/6752809/40017275-3fd98036-578f-11e8-9ab9-91bc78ec2808.png
Is there any way to fix this?
Thanks in advance!
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/twmeggs/anfis/issues/3, or mute the thread https://github.com/notifications/unsubscribe-auth/ABxrdIYpTlCvOhTsTSW3tGvcZB1Bys6Eks5tydM_gaJpZM4T-VAn .
I've tried running in Python 2.7 as well, but with the same results.
Do you have any other example code?
Thanks again for the prompt answer!
Em seg, 14 de mai de 2018 16:30, twmeggs notifications@github.com escreveu:
Thanks Angelo. I'm afraid I'm not sure why you are seeing those results. I haven't looked at this code for several years, but it was definitely working when the package was first created and should work with Python 2.
My only suggestion is to debug as the mechanics of ANFIS have not changed!
On Mon, 14 May 2018, 19:56 Ângelo Paulino, notifications@github.com wrote:
Hi!
Thanks for your work, but I'm afraid your example isn't working. The error just keeps going up and up, even raising the epochs number. There is no going down...
[image: figure_1] < https://user-images.githubusercontent.com/6752809/40017274-3fb836c4-578f-11e8-8a4b-5919bc35725c.png
[image: figure_2] < https://user-images.githubusercontent.com/6752809/40017275-3fd98036-578f-11e8-9ab9-91bc78ec2808.png
Is there any way to fix this?
Thanks in advance!
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/twmeggs/anfis/issues/3, or mute the thread < https://github.com/notifications/unsubscribe-auth/ABxrdIYpTlCvOhTsTSW3tGvcZB1Bys6Eks5tydM_gaJpZM4T-VAn
.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/twmeggs/anfis/issues/3#issuecomment-388935120, or mute the thread https://github.com/notifications/unsubscribe-auth/AGcKKW6jLPjFVwgEC2KEMiFtqKLmltrjks5tydtigaJpZM4T-VAn .
@twmeggs,
I got the same results mentioned by @Raphango. Please, run again the example and show us your results. Thanks!
I have the same issue
@Raphango @erralves @gcunhase Thanks for your patience. I've altered the initial state values for some of the parameters of the membership functions in the tests.py and the results converge now. In general, altering the initial parameters values has a large effect on how the algorithm then performs.
Hi!
Thanks for your work, but I'm afraid your example isn't working. The error just keeps going up and up, even raising the epochs number. There is no going down...
Is there any way to fix this?
Thanks in advance!