shimazaki

function [optN, optD, edges, C, N] = sshist(x,N)
% [optN, edges, C, N] = sshist(x,N)
%
% Function `sshist' returns the optimal number of bins in a histogram
% used for density estimation.
% Optimization principle is to minimize expected L2 loss function between 
% the histogram and an unknown underlying density function.
% An assumption made is merely that samples are drawn from the density
% independently each other.
%

shimazaki

# Gauss function
Gauss(x,w)=1/sqrt(2*pi)./w.*exp(-x.^2/2./w.^2) 

shimazaki

# Benchmark
# Julia
tic(); for i=1:1e5; pi; end; toc()

# Matlab
tic; for i=1:1e5, pi; end; toc;


x = randn(100,1);

shimazaki

% figure setting
figure; set(0,'DefaultAxesFontSize',14);
set(0,'DefaultAxesFontSize',12,'DefaultAxesFontName','Arial');

% adjust axis
axis tight;
tmp = get(gca,'XLim'); set(gca,'XLim',[tmp(1)-.2*diff(tmp),tmp(1)+1.2*diff(tmp)]);
tmp = get(gca,'YLim'); set(gca,'YLim',[tmp(1)-.2*diff(tmp),tmp(1)+1.2*diff(tmp)]);

% save figure

shimazaki

%% COMPARISON OF KERNEL DENSITY ESTIMATION ALGORITHMS
%  FFT METHOD IS >1000 FASTER THAN CONVOLUTION METHOD.

clear all; close all;

ts = 1+rand(1,1e5);     %spike time
dt = 0.001;
ts = round(ts/dt)*dt;   %make sampling resolution 1/dt;

shimazaki

function [y, t, optw, yconf, C, W] = sskernel(x,t,W)
% [optW, C, W] = sskernel(x,W,t)
%
% Function `sskernel' returns an optimal bandwidth (standard deviation) 
% of the Gauss density function used in kernel density estimation.
% Optimization principle is to minimize expected L2 loss function between 
% the kernel estimate and an unknown underlying density function. 
% An assumption made is merely that samples are drawn from the density
% independently each other.
%

shimazaki

Amazon S3
apt-get install s3cmd
s3cmd --configure

s3cmd get --recursive s3://bucketname ./


du -sh
du -sh ~/*

shimazaki

data %2d data

subplot(2,2,1);
[optW,C,W,Z,X,Y,dx,dy]=sskernel2d(data);
axis square;

fx = sum(Z'* dy);
fy = sum(Z * dx);

shimazaki

// JavaScript Document

function binary_sequence(N) {
	var x = new Array(N);
	
	for (var i=0;i<N-1;i++) {
		x[i] = (Math.random() < 0.2);
	}
	return x;
}