neurodata
RandMatPoisson does not output $d$ new features.
@jovo @falkben @ttomita The below code chunk shows that when running RandMatPoisson the number of specified output features is not equal to the input. Should this be corrected via shifting the Poisson distribution's support to $N$ instead of $N \cup {0}$ ?
set.seed(42)
paramList <- list(p = 100, d = 25, lambda = 5/10)
(a <- do.call(RandMatPoisson, paramList))
nz.rows nz.cols
[1,] 52 1 1
[2,] 39 1 1
[3,] 91 2 1
[4,] 45 2 -1
[5,] 84 4 -1
[6,] 74 5 -1
[7,] 82 7 1
[8,] 39 9 -1
[9,] 69 10 1
[10,] 1 12 1
[11,] 84 13 -1
[12,] 1 13 -1
[13,] 21 16 1
[14,] 90 16 1
[15,] 62 17 1
[16,] 38 17 1
[17,] 44 21 1
[18,] 4 23 1
[19,] 97 23 -1
[20,] 43 23 -1
[21,] 96 24 1
[22,] 88 24 1
i don't fully understand, happy to talk in person when i get back. but seems like you found a bug and know how to fix it?
On Mon, Mar 11, 2019 at 4:03 PM Jesse Leigh Patsolic < [email protected]> wrote:
@jovo https://github.com/jovo @falkben https://github.com/falkben @ttomita https://github.com/ttomita The below code chunk shows that when running RandMatPoisson the number of specified output features is not equal to the input. Should this be corrected via shifting the Poisson distribution's support to $N$ instead of $N \cup {0}$ ?
set.seed(42) paramList <- list(p = 100, d = 25, lambda = 5/10) (a <- do.call(RandMatPoisson, paramList)) nz.rows nz.cols [1,] 52 1 1 [2,] 39 1 1 [3,] 91 2 1 [4,] 45 2 -1 [5,] 84 4 -1 [6,] 74 5 -1 [7,] 82 7 1 [8,] 39 9 -1 [9,] 69 10 1 [10,] 1 12 1 [11,] 84 13 -1 [12,] 1 13 -1 [13,] 21 16 1 [14,] 90 16 1 [15,] 62 17 1 [16,] 38 17 1 [17,] 44 21 1 [18,] 4 23 1 [19,] 97 23 -1 [20,] 43 23 -1 [21,] 96 24 1 [22,] 88 24 1
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/neurodata/RerF/issues/208, or mute the thread https://github.com/notifications/unsubscribe-auth/AACjckO0VEojn-tBrTRBEXfVVr_SVO9Tks5vVraFgaJpZM4bpcut .
-- the glass is all full: half water, half air. neurodata.io
