Yeah, you might not be able to go much faster here unless A has some
specialized structure that you can take advantage of (e.g., sparsity)?
On 8/10/18, 11:22 AM, "Ravi Varadhan" wrote:
>Hi,
>
>I would like to compute: A %*% B %*% t(A)
>
>
>
>A is a mxn matrix and B is an nxn symmetric,
Duncan,
Since you asked, here is an updated version of my function.
# This method gets the Data.
getReturns1 <- function(symbol, norm = FALSE)
{
library(quantmod)
series = getSymbols(symbol, src = "yahoo", from = start, to = end,
auto.assign = FALSE)
length <- nrow( series )
Marco,
The error message indicates that nlon*nlat is 420 and that 1378620/420 has a
remainder. For the matrix to form, all rows have to be complete.
I am guessing you have at least one value incorrect among nlon, nlat, t or the
length of fulldatav.
Mark
R. Mark Sharp, Ph.D.
Data Scientist
Hi Ravi,
You can achieve substantial speed up by using a faster BLAS (e.g.,
OpenBLAS or MKL), especially on systems with multiple CPUs. On my (6
year old, but 8 core) system your example takes 3.9 seconds with using
the reference BLAS and only 0.9 seconds using OpenBLAS.
Best,
Ista
On Fri, Aug
Dear R-list users,
I have 10 data frames (called df1, df2, ... df10), where each of them contains
snow data from an automatic meteorological station (obviously each station has
a different station code).
Here is an example of df1:
station_code date_factor date_POSIX snow
217 1999-12-15
I am trying to write a function to make a matrix of precipitation with
this secuencie;
PRIMER PERIODO
cordex1 <-
nc_open("pr_CAM-44i_ICHEC-EC-EARTH_rcp45_r12i1p1_SMHI-RCA4_v1_mon_200601-201012.nc")
fullmon1<-ncvar_get(cordex1,"pr")
lat<-ncvar_get(cordex1,"lat", start=c(35.5),count=c(20))
Hi,
I would like to compute: A %*% B %*% t(A)
A is a mxn matrix and B is an nxn symmetric, positive-definite matrix, where m
is large relative to n (e.g., m=50,000 and n=100).
Here is a sample code.
M <- 1
N <- 100
A <- matrix(rnorm(M*N), M, N)
B <- crossprod(matrix(rnorm(N*N),
On Fri, 10 Aug 2018, MacQueen, Don wrote:
I would start by trying to install rgdal by itself, rather than as part of
a "batch" update. As in
Install.packages('rgdal')
Don,
rgdal was supposed to be installed, but you have a valid point.
My expectation is that you will see a more
I would start by trying to install rgdal by itself, rather than as part of a
"batch" update. As in
Install.packages('rgdal')
My expectation is that you will see a more complete error message specific to
rgdal, which presumably will provide a clue or pointer.
-Don
--
Don MacQueen
Lawrence
On Fri, 10 Aug 2018, MacQueen, Don wrote:
C++11 is a programming language.
Don,
That's what I assumed; a version of c++, which is installed here by
default. Perhaps not that verision, but ...
configure: C++11 support available
and then later says
/usr/include/cpl_port.h:187:6: error:
What's C++11?
Google it!
-- Bert
Bert Gunter
"The trouble with having an open mind is that people keep coming along and
sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Fri, Aug 10, 2018 at 9:53 AM, Rich Shepard
wrote:
> On Fri, 10 Aug
Rich,
C++11 is a programming language.
Interestingly, someone else asked for help yesterday with exactly the same
error message for rgdal. The subject line was
"Help reinstalling rgdal (Ubuntu 16.04)"
The suggestions were to visit the mailing lists R-sig-geo and/or R-sig-debian.
It's
Hi,
I am wondering if I am doing correctly, but my auto.arima usually (if
not always) give me (0,1,0), whatever portion of the series I take. In the
following instances, only the last one yields ARIMA(0,1,1), and all the
other cases yield ARIMA(0,1,0). I would like to do forecast based on
Updating installed packages ends with a warning:
Warning message:
In install.packages(update[instlib == l, "Package"], l, contriburl =
contriburl, :
installation of package ‘rgdal’ had non-zero exit status
What steps should I take to correct this?
Rich
14 matches
Mail list logo
