|
From: | Kevin Martin |
Subject: | Re: [Varnamproject-discuss] [bug #40401] Implement a better frequency calculation algorithm |
Date: | Sat, 5 Mar 2016 12:22:06 +0530 |
Thanks for the reply!Like i said i'm new to Open Source and would appreciate it if you could help me through the first few steps so that i could start contributing to your project in large. I'm a computational linguistics student at IIIT-Hyderabad and was actually very impressed and excited by your project and would love to contribute and help build it!BTW , do you have any suggestions as to the normalised range , and could i take this bug up for debugging?Waiting for your reply!Pranav NairOn Thu, Mar 3, 2016 at 2:34 PM, Kevin Martin <address@hidden> wrote:Hi Pranav,Glad to see that you are interested. But capping the max value would not work in the long term. Eventually, all the words would reach the max value and then the suggestion will not work as expected.Navneeth was suggesting something like normalizing the confidence values of the entire range of words once a word hits a max value. How to normalize them, howerver, is another question.thanks,Kevin Martin JoseOn Wed, Mar 2, 2016 at 7:21 PM, Pranav Nair <address@hidden> wrote:Follow-up Comment #1, bug #40401 (project varnamproject):
Would like to take it up! But i'm new to Open Source so if someone could guide
me around that would be nice. One suggestion i do have though is that there
could be a max value for the confidence value , and we could stop updating
when the value is reached
_______________________________________________________
Reply to this item at:
<http://savannah.nongnu.org/bugs/?40401>
_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/
[Prev in Thread] | Current Thread | [Next in Thread] |