Displaying 20 results from an estimated 20 matches for "coef0".
Did you mean:
coef
2006 Jul 07
1
Polynomial kernel in SVM in e1071 package
Dear list,
In some places (for example,
http://en.wikipedia.org/wiki/Support_vector_machine) , the polynomail
kernel in SVM is written as (u'*v + 1)^d, while in the document of
svm() in e1071 package, the polynomial kernel is written as
(gamma*u'*v + coef0)^d. I am a little confused here:
When doing parameter optimization (grid search or so) for polynomial
kernel, does it need to tune four parameters, gamma, coef0, C and
degree, or just two of them, C and degree (and fixing gamma to 1 and
coef0 = 1)?
Thanks,
Wuming
2008 Jan 02
1
Plot.svm error
...solve the problem and haven't been able yet to do it. So I'm doing this:
> dados<-read.table("b.txt",sep="",nrows=30000)
> model<-svm(x=dados[,-1],y=dados[,1],scale=TRUE,type="C-classification",kernel="radial",cost=1,degree=3,gamma=0.083,coef0=0,nu=0.5,epsilon=0.1)
Warning message:
In svm.default(x = dados[, -1], y = dados[, 1], scale = TRUE, type = "C-classification", :
Variable(s) ?V12? and ?V13? constant. Cannot scale data.
> summary(model)
Call:
svm.default(x = dados[, -1], y = dados[, 1], scale = TRUE, type = "...
2015 Dec 20
2
[Aarch64 v2 05/18] Add Neon intrinsics for Silk noise shape quantization.
Jonathan Lennox wrote:
> +opus_int32 silk_noise_shape_quantizer_short_prediction_neon(const opus_int32 *buf32, const opus_int32 *coef32)
> +{
> + int32x4_t coef0 = vld1q_s32(coef32);
> + int32x4_t coef1 = vld1q_s32(coef32 + 4);
> + int32x4_t coef2 = vld1q_s32(coef32 + 8);
> + int32x4_t coef3 = vld1q_s32(coef32 + 12);
> +
> + int32x4_t a0 = vld1q_s32(buf32 - 15);
> + int32x4_t a1 = vld1q_s32(buf32 - 11);
> + int32x4_t a...
2015 Nov 20
2
[Aarch64 00/11] Patches to enable Aarch64
> On Nov 19, 2015, at 5:47 PM, John Ridges <jridges at masque.com> wrote:
>
> Any speedup from the intrinsics may just be swamped by the rest of the encode/decode process. But I think you really want SIG2WORD16 to be (vqmovns_s32(PSHR32((x), SIG_SHIFT)))
Yes, you?re right. I forgot to run the vectors under qemu with my previous version (oh, the embarrassment!) Fixed forthcoming
2003 Nov 03
1
svm in e1071 package: polynomial vs linear kernel
I am trying to understand what is the difference between linear and
polynomial kernel:
linear: u'*v
polynomial: (gamma*u'*v + coef0)^degree
It would seem that polynomial kernel with gamma = 1; coef0 = 0 and degree
= 1
should be identical to linear kernel, however it gives me significantly
different results for very simple
data set, with linear kernel significantly outperforming polynomial
kernel.
*** mse, r2 = 0.5, 0.9 fo...
2004 Nov 29
1
tune()
Hi
I am trying to tune an svm by doing the following:
tune(svm, similarity ~., data = training, degree = 2^(1:2), gamma =
2^(-1:1), coef0 = 2^(-1:1), cost = 2^(2:4), type = "polynomial")
but I am getting
Error in svm.default(x, y, scale = scale, ...) :
wrong type specification!
>
I have to admit I am not sure what I am doing wrong. Could anyone tell
me why the parameters I am using are wrong?
Plus cou...
2015 Nov 23
1
[Aarch64 v2 05/18] Add Neon intrinsics for Silk noise shape quantization.
...iced that your NEON code doesn't use any of the "high" intrinsics for ARM64, e.g. instead of:
int32x4_t coef1 = vmovl_s16(vget_high_s16(coef16));
you could use:
int32x4_t coef1 = vmovl_high_s16(coef16);
and instead of:
int64x2_t b1 = vmlal_s32(b0, vget_high_s32(a0), vget_high_s32(coef0));
you could use:
int64x2_t b1 = vmlal_high_s32(b0, a0, coef0);
and instead of:
int64x1_t c = vadd_s64(vget_low_s64(b3), vget_high_s64(b3));
int64x1_t cS = vshr_n_s64(c, 16);
int32x2_t d = vreinterpret_s32_s64(cS);
out = vget_lane_s32(d, 0);
you could use:
out = (opus_int32)(vaddvq_s64(b3) &g...
2015 Dec 21
0
[Aarch64 v2 05/18] Add Neon intrinsics for Silk noise shape quantization.
> On Dec 19, 2015, at 10:07 PM, Timothy B. Terriberry <tterribe at xiph.org> wrote:
>
> Jonathan Lennox wrote:
>> +opus_int32 silk_noise_shape_quantizer_short_prediction_neon(const opus_int32 *buf32, const opus_int32 *coef32)
>> +{
>> + int32x4_t coef0 = vld1q_s32(coef32);
>> + int32x4_t coef1 = vld1q_s32(coef32 + 4);
>> + int32x4_t coef2 = vld1q_s32(coef32 + 8);
>> + int32x4_t coef3 = vld1q_s32(coef32 + 12);
>> +
>> + int32x4_t a0 = vld1q_s32(buf32 - 15);
>> + int32x4_t a1 = vld1q_s32(buf32 - 11...
2015 Aug 05
0
[PATCH 7/8] Add Neon intrinsics for Silk noise shape feedback loop.
...vdupq_n_s32(data0[0]);
+ int32x4_t a01 = vld1q_s32(data1); // data1[0] ... [3]
+
+ int32x4_t a0 = vextq_s32 (a00, a01, 3); // data0[0] data1[0] ...[2]
+ int32x4_t a1 = vld1q_s32(data1 + 3); // data1[3] ... [6]
+
+ int16x8_t coef16 = vld1q_s16(coef);
+ int32x4_t coef0 = vmovl_s16(vget_low_s16(coef16));
+ int32x4_t coef1 = vmovl_s16(vget_high_s16(coef16));
+
+ int64x2_t b0 = vmull_s32(vget_low_s32(a0), vget_low_s32(coef0));
+ int64x2_t b1 = vmlal_s32(b0, vget_high_s32(a0), vget_high_s32(coef0));
+ int64x2_t b2 = vmlal_s32(b1, vget_low_...
2015 Nov 21
0
[Aarch64 v2 06/18] Add Neon intrinsics for Silk noise shape feedback loop.
...vdupq_n_s32(data0[0]);
+ int32x4_t a01 = vld1q_s32(data1); // data1[0] ... [3]
+
+ int32x4_t a0 = vextq_s32 (a00, a01, 3); // data0[0] data1[0] ...[2]
+ int32x4_t a1 = vld1q_s32(data1 + 3); // data1[3] ... [6]
+
+ int16x8_t coef16 = vld1q_s16(coef);
+ int32x4_t coef0 = vmovl_s16(vget_low_s16(coef16));
+ int32x4_t coef1 = vmovl_s16(vget_high_s16(coef16));
+
+ int64x2_t b0 = vmull_s32(vget_low_s32(a0), vget_low_s32(coef0));
+ int64x2_t b1 = vmlal_s32(b0, vget_high_s32(a0), vget_high_s32(coef0));
+ int64x2_t b2 = vmlal_s32(b1, vget_low_...
2008 Jan 03
0
Svm formula
...;m having trouble. I should be very grateful if someone could help me on this..
> dados<-read.table("b.txt",sep="",nrows=30000)
> model<-svm(x=dados[,-1],y=dados[,1],scale=TRUE,type="C-classification",kernel="radial",cost=1,degree=3,gamma=0.083,coef0=0,nu=0.5,epsilon=0.1)
Warning message:
In svm.default(x = dados[, -1], y = dados[, 1], scale = TRUE, type = "C-classification", :
Variable(s) ?V12? and ?V13? constant. Cannot scale data.
> summary(model)
Call:
svm.default(x = dados[, -1], y = dados[, 1], scale = TRUE, type = "...
2015 Nov 23
0
[Aarch64 v2 05/18] Add Neon intrinsics for Silk noise shape quantization.
...ced that your NEON code doesn't use any of the "high" intrinsics for
ARM64, e.g. instead of:
int32x4_t coef1 = vmovl_s16(vget_high_s16(coef16));
you could use:
int32x4_t coef1 = vmovl_high_s16(coef16);
and instead of:
int64x2_t b1 = vmlal_s32(b0, vget_high_s32(a0), vget_high_s32(coef0));
you could use:
int64x2_t b1 = vmlal_high_s32(b0, a0, coef0);
and instead of:
int64x1_t c = vadd_s64(vget_low_s64(b3), vget_high_s64(b3));
int64x1_t cS = vshr_n_s64(c, 16);
int32x2_t d = vreinterpret_s32_s64(cS);
out = vget_lane_s32(d, 0);
you could use:
out = (opus_int32)(vaddvq_s64(b3) &g...
2003 Dec 10
3
e1071:svm - default epsilon = 0.1 (NOT 0.5) (PR#5671)
In e1071 package/svm default epsilon value is set to 0.1 and not 0.5
as documentation says.
R
2015 Nov 21
12
[Aarch64 v2 00/18] Patches to enable Aarch64 (version 2)
As promised, here's a re-send of all my Aarch64 patches, following
comments by John Ridges.
Note that they actually affect more than just Aarch64 -- other than
the ones specifically guarded by AARCH64_NEON defines, the Neon
intrinsics all also apply on armv7; and the OPUS_FAST_INT64 patches
apply on any 64-bit machine.
The patches should largely be independent and independently useful,
other
2015 Aug 05
0
[PATCH 6/8] Add Neon intrinsics for Silk noise shape quantization.
...ude "main.h"
+#include "stack_alloc.h"
+#include "NSQ.h"
+#include "celt/cpu_support.h"
+#include "celt/arm/armcpu.h"
+
+opus_int32 silk_noise_shape_quantizer_short_prediction_neon(const opus_int32 *buf32, const opus_int32 *coef32)
+{
+ int32x4_t coef0 = vld1q_s32(coef32);
+ int32x4_t coef1 = vld1q_s32(coef32 + 4);
+ int32x4_t coef2 = vld1q_s32(coef32 + 8);
+ int32x4_t coef3 = vld1q_s32(coef32 + 12);
+
+ int32x4_t a0 = vld1q_s32(buf32 - 15);
+ int32x4_t a1 = vld1q_s32(buf32 - 11);
+ int32x4_t a2 = vld1q_s32(buf32 - 7);
+ int3...
2015 Nov 21
0
[Aarch64 v2 05/18] Add Neon intrinsics for Silk noise shape quantization.
...ude "main.h"
+#include "stack_alloc.h"
+#include "NSQ.h"
+#include "celt/cpu_support.h"
+#include "celt/arm/armcpu.h"
+
+opus_int32 silk_noise_shape_quantizer_short_prediction_neon(const opus_int32 *buf32, const opus_int32 *coef32)
+{
+ int32x4_t coef0 = vld1q_s32(coef32);
+ int32x4_t coef1 = vld1q_s32(coef32 + 4);
+ int32x4_t coef2 = vld1q_s32(coef32 + 8);
+ int32x4_t coef3 = vld1q_s32(coef32 + 12);
+
+ int32x4_t a0 = vld1q_s32(buf32 - 15);
+ int32x4_t a1 = vld1q_s32(buf32 - 11);
+ int32x4_t a2 = vld1q_s32(buf32 - 7);
+ int3...
2015 Dec 23
6
[AArch64 neon intrinsics v4 0/5] Rework Neon intrinsic code for Aarch64 patchset
Following Tim's comments, here are my reworked patches for the Neon intrinsic function patches of
of my Aarch64 patchset, i.e. replacing patches 5-8 of the v2 series. Patches 1-4 and 9-18 of the
old series still apply unmodified.
The one new (as opposed to changed) patch is the first one in this series, to add named constants
for the ARM architecture variants.
There are also some minor code
2015 Aug 05
8
[PATCH 0/8] Patches for arm64 (aarch64) support
This sequence of patches provides arm64 support for Opus. Tested on
iOS, Android, and Ubuntu 14.04.
The patch sequence was written on top of Viswanath Puttagunta's Ne10
patches, but all but the second ("Reorganize pitch_arm.h") should, I
think, apply independently of it. It does depends on my previous
intrinsics configury reorganization, however.
Comments welcome.
With this and
2015 Nov 07
12
[Aarch64 00/11] Patches to enable Aarch64 (arm64) optimizations, rebased to current master.
Here are my aarch64 patches rebased to the current tip of Opus master.
They're largely the same as my previous patch set, with the addition
of the final one (the Neon fixed-point implementation of
xcorr_kernel). This replaces Viswanath's Neon fixed-point
celt_pitch_xcorr, since xcorr_kernel is used in celt_fir and celt_iir
as well.
These have been tested for correctness under qemu
2005 Apr 26
3
Error using e1071 svm: NA/NaN/Inf in foreign function call
...29, 29,
29, 29, 29, 29, 29, 29, 30, 30, 30, 30, 30, 30, 30, 30, 30,
30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30)), scale = c(TRUE,
TRUE, FALSE, FALSE, TRUE), type = "nu-regression", kernel = "polynomial",
cost = 1, nu = 0, degree = 2, gamma = 0.0009765625, coef0 = -1)
debug: UseMethod("svm")
Browse[1]>
Error in predict.svm(ret, xhold) : NA/NaN/Inf in foreign function call (arg 9)
>
[[alternative HTML version deleted]]