search for: sums

From: Thorvald Natvig <slicer at users.sourceforge.net> Some optimizing compilers miscompile the current SSE optimizations when full optimizations are enabled. By using output value pointer instead of a return value, we can bypass this misbehaviour. --- libspeex/resample.c | 8 ++++---- libspeex/resample_sse.h | 24 ++++++++---------------- 2 files changed, 12 insertions(+), 20

Dear List, Suppose we have a variable K.JUN defined as (with 1=wet, 0=dry): K.JUN1984 = c(1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) K.JUN1985 = c(0, 1, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1) K.JUN1986 = c(0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1)

Hello I'm testing SPEEX on embedded board using ARM7 (Atmel). ARM7 don't have floating point so I'm using FIXED_POINT. Unfortunately the encoding speed is about 5 times slower then necessary for real time. ARM7 is slow for 16/8 bits operations. The sequence: static inline spx_word32_t compute_pitch_error(spx_word16_t *C, spx_word16_t *g, spx_word16_t pitch_control) {

...or very large files the block count would grow linearly with file size to outrageous numbers (65536/GB). An iso image would need over 40,000 blocks and with the fixed sum size of 4+2 bytes had a measurable sum collision rate. With the sum collisions the file would have to be redone with 4+16 byte sums killing efficiency. Additionally all those blocks meant a good deal of processing to find matching blocks and a sizable memory and network load. varsumlen2.patch calculates the block size to be a rounded square root of the file size with a 700byte minimum. In this way block sizes and counts grow...

I have a data frame called test shown below that i would like to summarize in a particular way : I want to show the column sums (columns y ,f) grouped by country (column e1). However, I'm looking for the data to be split according to column e2. In other words, two tables of sum by country. One table for "con" and one table for "std" shown in column e2. Finally at the bottom of the two tables, I would...

Jean-Marc, At least VS2005 (what I'm using) won't compile resample_sse.h with _USE_SSE2 defined because it refuses to cast __m128 to __m128d and vice versa. While there are intrinsics to do the casts, I thought it would be simpler to just use an intrinsic that accomplishes the same thing without all the casting. Thanks, --John @@ -91,7 +91,7 @@ static inline double

Full_Name: Nick Efthymiou Version: R1.5.0 and above OS: Red Hat Linux Submission from: (NULL) (162.93.14.73) With the introduction of the functions rowSums(), colSums(), rowMeans() and colMeans() in R1.5.0, function "SEXP do_colsum(SEXP call, SEXP op, SEXP args, SEXP rho)" was added to perform the fast summations. We have an excellent opportunity to improve the accuracy by implementing Kahan summation here. Kahan summation is described in...

I have two problems where I've come up with some code that will do the analysis that I want, but it looks pretty clumsy. In the first case, I calculate the variance on five different columns for each of 14 clusters and get them into one matrix. I get the job done, but I would have thought that it could be done in one or two lines, not six, and be generalized so that it didn't matter how

Can someone please tell me what is up with na.action in aggregate? My (somewhat) reproducible example: (I say somewhat because some lines wouldn't run in a separate session, more below) set.seed(100) dat=data.frame( x1=sample(c(NA,'m','f'), 100, replace=TRUE), x2=sample(c(NA, 1:10), 100, replace=TRUE), x3=sample(c(NA,letters[1:5]), 100, replace=TRUE),

Hi all, I committed a first cut at blockless Givens/Whens/Thens to RSpec''s trunk: cd /path/to/rspec/project svn up cd rspec bin/spec examples/story/calculator.rb Take a look at examples/story/calculator.rb to see what''s going on. Needs docs!!!! Thoughts welcome. I''ve also got a cut at the plain text parser checked in, but it''s not hooked up to anything

My simple functions were to compare the result with the gfortran compiler sum() function. I thought that the Fortran sum could not be less precise than R. I was wrong. I am impressed. The R sum does in fact match the result if we use the Kahan algorithm. P. I am glad to see that R sum() is more accurate than the gfortran compiler sum. On 16/03/18 11:37 AM, luke-tierney at uiowa.edu wrote:

Hi all, I found a discrepancy between the sum() in R and either a sum done in C or Fortran for vector of just 5 elements. The difference is very small, but this is a very small part of a much larger numerical problem in which first and second derivatives are computed numerically. This is part of a numerical method course I am teaching in which I want to compare speeds of R versus Fortran (We

On Fri, 7 Feb 2014, Timothy B. Terriberry wrote: > Martin Storsjo wrote: >> This is required in order to build using the built-in assembler >> in clang. > > These patches break the gcc build (with "Error: bad instruction"). Ah, right, sorry about that. > Documentation I've seen is contradictory on which order ({cond}{size} or > {size}{cond}) is correct.

It looks like the 'seq' variable to 'for' can be altered from within the loop, leading to incorrect answers. E.g., in the following I'd expect 'sum' to be 1+2=3, but R 2.10.0 (svn 48686) gives 44.5. > x = c(1,2); sum = 0; for (i in x) { x[i+1] = i + 42.5; sum = sum + i }; sum [1] 44.5 or, with a debugging cat()s, > x = c(1,2); sum = 0; for (i in x) {

We want to sum many vectors and numbers together as a vector if there is at least one vector in the arguments. For example, 1 + c(2,3) = c(3,4). Since we are not sure arguments to sum, we are using sum function: sum(v1,v2,...,n1,n2,..). The problem is that sum returns the sum of all the values present in its arguments: sum(1,c(2,3))=6 sum(1,2,3)=6 We do not want to turn sum(v1,v2,...,n1,n2,..) to

...le at the time. The two attached patches implement per-file dynamic checksum lengths. The dynsumlen patch lays the groundwork. It makes the checksum length a per-file attribute alongside the block size. This required a protocol bump and reaching into almost every routine that touches block checksums. I found that there was a great deal of code duplication involved with reading and writing the sum_struct fields. I have eliminated the duplication by consolidating that transmission in the new (read|write)_sum_head functions. A added bonus is that write_sum_head replaces the send_null_sums func...

> a <- 0 ; for(i in (1:200000000)) a <- a + 1/i > b <- 0 ; for(i in (200000000:1)) b <- b + 1/i > c <- sum(1/(1:200000000)) > d <- sum(1/(200000000:1)) > order(c(a,b,c,d)) [1] 1 2 4 3 > b<c [1] TRUE > c==d [1] FALSE I'd expected b being the largest, since we sum up the smallest numbers first. Instead, c is the largest, which is

Hi All, I am attempting to simulation an inventory model on R and I am having some problems. I believe the problem is when I define my demand rate is stays constant throughout so when I do need to reorder the model does not recognise it as I have the initial supply arrival time set to infinity at the start as no order has been played but I have an if statement saying, if the level falls below a

Hello, I have being trying to estimate the parameters of the?generalized?exponential distribution. The random number generation for the GE distribution is?x<-(-log(1-U^(1/p1))/b), where U stands for uniform dist. The data i have generated to estimate the parameters is right censored and the code is given below; The problem is that, the newton-Raphson approach isnt working and i do not know what

Hello, A strange problem appears when upgrading from release_34 to testing. Some transformations to bitcode files cause registered global_ctors to not be called. Here's an example (I've also attached the complete example and pasted it below): This works: clang -fsanitize=address -flto -c -o sum.o sum.c clang -fsanitize=address -o sum sum.o This doesn't work: clang