verne.freenode.net 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/
rohitshinde has joined #mlpack
rohitshinde has quit [Quit: Leaving]
Jayjaychicago has joined #mlpack
< Jayjaychicago>
Hi, I'm Julien. I'm completely new to doing kmeans using mlpack and I have a few questions
< Jayjaychicago>
My input cvs contains 5 columns (say: age, city, number of phone used, purchase_or_no_purchase, #visits to website) as you can see strings and booleans. My ideal output would be: 5 clusters with the importance of each variable within that cluster and the median for that each variable within the cluster. Is that what the output file is representing? What about the centroid file? How does it deal with string, boolean inputs (or
< Jayjaychicago>
Ideally, it would also tell me how many of my cases fall within that cluster
rohitshinde has joined #mlpack
< rohitshinde>
I am looking at the BSP implementation
< rohitshinde>
Where is the SplitNode function defined?
< rohitshinde>
For a quadtree, what type of data would be input? Would they be co ordinates?
< rohitshinde>
Like (1,2) (3,4), etc?
< rohitshinde>
Once I get data I think I would be able to implement it
< rohitshinde>
*the type of data
< rohitshinde>
Sorry! My bad. I did not see the complete implementation of the BSP
< rohitshinde>
I got the function
< rohitshinde>
So my final question is this: What type of data will be input to a quadtree?
witness has joined #mlpack
rohitshinde has quit [Quit: Leaving]
Jayjaychicago has quit [Ping timeout: 246 seconds]
rohitshinde has joined #mlpack
rohitshinde has quit [Client Quit]
JaraxussTong has quit [Quit: No Ping reply in 180 seconds.]
JaraxussTong has joined #mlpack
govg has joined #mlpack
govg has quit [Quit: leaving]
rohitshinde has joined #mlpack
rohitshinde has quit [Ping timeout: 250 seconds]
rohitshinde has joined #mlpack
rohitshinde has quit [Ping timeout: 264 seconds]
rohitshinde has joined #mlpack
< naywhayare>
rohitshinde: the data will be an armadillo matrix
< naywhayare>
where each point in the dataset is a column
< naywhayare>
i.e. point i can be accessed via dataset.col(i)
< rohitshinde>
Could you give me an example point?
< naywhayare>
sure, each element in the point would correspond to a coordinate... so a 3-dimensional point at the origin would just be [0, 0, 0]
< naywhayare>
(you can create this point as 'arma::vec point("0 0 0");')
< rohitshinde>
Okay
< rohitshinde>
Now, what will be the bounds of the quadtree?
< rohitshinde>
That is, what will be the bounding co-ordinates of the quadtree?
< rohitshinde>
I will need to know this so that I can split it into four parts evenly
< naywhayare>
you'll have to calculate this from the data that you're given
< rohitshinde>
Okay!
udit_s has joined #mlpack
udit_s has quit [Ping timeout: 264 seconds]
rohitshinde has quit [Quit: Leaving]
udit_s has joined #mlpack
udit_s has quit [Quit: leaving]
udit_s has joined #mlpack
decltypeme has joined #mlpack
udit_s has quit [Quit: leaving]
travis-ci has joined #mlpack
< travis-ci>
mlpack/mlpack#69 (master - 861ac83 : Ryan Curtin): The build passed.