naywhayare changed the topic of #mlpack to: http://www.mlpack.org/ -- We don't respond instantly... but we will respond. Give it a few minutes. Or hours. -- Channel logs: http://www.mlpack.org/irc/
andrewmw94 has left #mlpack []
govg has joined #mlpack
govg has quit [Changing host]
govg has joined #mlpack
andrewmw94 has joined #mlpack
< marcus_zoq>
naywhayare: I think there is a bug in the hmm viterbi executable...
< marcus_zoq>
naywhayare: or my model file changed lets check
< marcus_zoq>
naywhayare: nevermind the model file changed
< naywhayare>
I'm looking forward to Michael Fox
< naywhayare>
*Michael Fox's changes to the SaveRestoreUtility
< naywhayare>
they should make model save/load possible and easy for more than just HMMs and GMMs...
< marcus_zoq>
sounds great, he works for/with you?
sumedhghaisas has joined #mlpack
Anand has joined #mlpack
< Anand>
Marcus : I found a possible bug in mean predictive information code and fixed it. But, I still dont see how the precision value can be nan.
< marcus_zoq>
Anand: okay, can you reproduce the nan bug?
< Anand>
For MPI, I found a case when it can be nan. Not for precision though!
< marcus_zoq>
Anand: You can produce the bug with the iris dataset and the scikit method.
< Anand>
Ok. I need a smaller example to figure the bug out. I was working on it. But we never got this nan error before for other libraries, right?
< marcus_zoq>
Anand: Yeah, I think so.
< Anand>
Marcus : Yeah. Let me think more. Btw, what will we do next?
< marcus_zoq>
Anand: I guess, we can start working on the representation of the results?
< Anand>
Yes, I guess so.
< Anand>
I will start it next week then
< marcus_zoq>
Anand: Okay, great!
Anand has quit [Ping timeout: 246 seconds]
sumedhghaisas has quit [Remote host closed the connection]
govg has quit [Ping timeout: 245 seconds]
Anand has joined #mlpack
< Anand>
Marcus : You are making changes to the master?
< marcus_zoq>
Anand: Yeah, I'm nearly done.
< Anand>
Ok. Is there a problem?
< marcus_zoq>
Anand: We've updated the shogun version, the new version uses a different import structure, so I've updated all shogun methods. Beside that I've update the linear regression and logistic regression weka methods because we've include the saving time into the classification process.
< Anand>
Marcus : Alright. No changes in the interface right?