On 2017/11/14 11:04:56, wissemchouk wrote: > On 2017/11/14 08:56:04, liulee.chn wrote: > > Using kernal ...
6 years, 5 months ago
(2017-11-14 11:05:49 UTC)
#6
On 2017/11/14 11:04:56, wissemchouk wrote:
> On 2017/11/14 08:56:04, liulee.chn wrote:
> > Using kernal RBF can have a better result
>
> Yes I will see what I can do about that ;)
Yup A LOT more, otherwise it's not really meaningful. Let's discuss on the next
meeting
On 2017/11/14 08:56:04, liulee.chn wrote: > Using kernal RBF can have a better result Even ...
6 years, 5 months ago
(2017-11-14 11:25:19 UTC)
#7
On 2017/11/14 08:56:04, liulee.chn wrote:
> Using kernal RBF can have a better result
Even though, I'm not sure it's going to be better because RBF basically
approximates the result based on the sum of functions which means we need a huge
amount of data . It could work if we have a small input space but as the
workload may vary a lot, I am afraid it's not going to be very accurate but
let's see
Issue 334050043: machine learning tool (svm knn)
Created 6 years, 5 months ago by wissemchouk
Modified 6 years, 5 months ago
Reviewers: Jiangfeng, liulee.chn, cuteliaoliao, tsumak.dmitri_gmail.com
Base URL:
Comments: 2