Taylor泰勒图，包含Excel直接出图版本、MATLAB代码出图版本，有测试数据。

% TAYLORDIAG Plot a Taylor Diagram % % [hp ht axl] = taylordiag(STDs,RMSs,CORs,['option',value]) % % Plot a Taylor diagram from statistics of different series. % % INPUTS: % STDs: Standard deviations % RMSs: Centered Root Mean Square Difference % CORs: Correlation % % Each of these inputs are one dimensional with same length. First % indice corresponds to the reference serie for the diagram. For exemple % STDs(1) is the standard deviation of the reference serie and STDs(2:N) % are the standard deviations of the other series. % % Note that by definition the following relation must be true for all series i: % RMSs(i) - sqrt(STDs(i).^2 + STDs(1)^2 - 2*STDs(i)*STDs(1).*CORs(i)) = 0 % This relation is checked and if not verified an error message is sent. Please see % Taylor's JGR article for more informations about this. % You can use the ALLSTATS function to avoid this to happen, I guess ;-). You can get % it somewhere from: http://codes.guillaumemaze.org/matlab % % OUTPUTS: % hp: returns handles of plotted points % ht: returns handles of the text legend of points % axl: returns a structure of handles of axis labels % % LIST OF OPTIONS: % For an exhaustive list of options to customize your diagram, please call the function % without arguments: % >> taylordiag % % SHORT TUTORIAL (see taylordiag_test.m for more informations): % An easy way to get compute inputs is to use the ALLSTATS function you can get from: % http://codes.guillaumemaze.org/matlab % Let's say you gathered all the series you want to put in the Taylor diagram in a % single matrix BUOY(N,nt) with N the number of series and nt their (similar) length. % If BUOY(1,:) is the serie of reference for the diagram: % for iserie = 2 : size(BUOY,1) % S = allstats(BUOY(1,:),BUOY(iserie,:)); % MYSTATS(iserie,:) = S(:,2); % We get stats versus reference % end%for iserie % MYSTATS(1,:) = S(:,1); % We assign reference stats to the first row % Note that the ALLSTATS function can handle NaNs, so be careful to compute statistics % with enough points ! % Then you're ready to simply run: % taylordiag(MYSTATS(:,2),MYSTATS(:,3),MYSTATS(:,4)); % % REF: K. Taylor % Summarizing multiple aspects of model performance in a single diagram % Journal of Geophysical Research-Atmospheres, 2001, V106, D7. % % Rev. by Guillaume Maze on 2010-02-10: Help more helpful ! Options now displayed by call. % Copyright (c) 2008 Guillaume Maze. % http://codes.guillaumemaze.org % All rights reserved. % % Redistribution and use in source and binary forms, with or without % modification, are permitted provided that the following conditions are met: % * Redistributions of source code must retain the above copyright notice, this list of % conditions and the following disclaimer. % * Redistributions in binary form must reproduce the above copyright notice, this list % of conditions and the following disclaimer in the documentation and/or other materials % provided with the distribution. % * Neither the name of the Laboratoire de Physique des Oceans nor the names of its contributors may be used % to endorse or promote products derived from this software without specific prior % written permission. % % THIS SOFTWARE IS PROVIDED BY Guillaume Maze ''AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, % INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A % PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL Guillaume Maze BE LIABLE FOR ANY % DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT % LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR % BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, % STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE % OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. % function varargout = taylordiag(varargin) %% if nargin == 0 disp_optionslist; return else narg = nargin - 3; if mod(narg,2) ~=0 error('taylordiag.m : Wrong number of arguments') end end STDs = varargin{1}; RMSs = varargin{2}; CORs = varargin{3}; %% CHECK THE INPUT FIELDS: apro = 100; di = fix(RMSs*apro)/apro - fix(sqrt(STDs.^2 + STDs(1)^2 - 2*STDs*STDs(1).*CORs)*apro)/apro; if find(di~=0) % help taylordiag.m ii = find(di~=0); if length(ii) == length(di) error(sprintf('taylordiag.m : Something''s wrong with ALL the datas\nYou must have:\nRMSs - sqrt(STDs.^2 + STDs(1)^2 - 2*STDs*STDs(1).*CORs) = 0 !')) else error(sprintf('taylordiag.m : Something''s wrong with data indice(s): [%i]\nYou must have:\nRMSs - sqrt(STDs.^2 + STDs(1)^2 - 2*STDs*STDs(1).*CORs) = 0 !',ii)) end end %% IN POLAR COORDINATES: rho = STDs; theta = real(acos(CORs)); dx = rho(1); % Observed STD %% %% BEGIN THE PLOT HERE TO GET AXIS VALUES: hold off cax = gca; tc = get(cax,'xcolor'); %ls = get(cax,'gridlinestyle'); ls = '-'; % DEFINE HERE THE GRID STYLE next = lower(get(cax,'NextPlot')); %% LOAD CUSTOM OPTION OF AXE LIMIT: nopt = narg/2; foundrmax = 0; for iopt = 4 : 2 : narg+3 optvalue = varargin{iopt+1}; switch lower(varargin{iopt}), case 'limstd', rmax = optvalue; foundrmax=1; end end % make a radial grid hold(cax,'on'); if foundrmax==0 maxrho = max(abs(rho(:))); else maxrho = rmax; end hhh = line([-maxrho -maxrho maxrho maxrho],[-maxrho maxrho maxrho -maxrho],'parent',cax); set(cax,'dataaspectratio',[1 1 1],'plotboxaspectratiomode','auto') v = [get(cax,'xlim') get(cax,'ylim')]; ticks = sum(get(cax,'ytick')>=0); delete(hhh); % check radial limits and ticks rmin = 0; if foundrmax == 0; rmax = v(4); end rticks = max(ticks-1,2); if rticks > 5 % see if we can reduce the number if rem(rticks,2) == 0 rticks = rticks/2; elseif rem(rticks,3) == 0 rticks = rticks/3; end end rinc = (rmax-rmin)/rticks; tick = (rmin+rinc):rinc:rmax; %% LOAD DEFAULT PARAMETERS: if find(CORs<0) Npan = 2; % double panel else Npan = 1; end tickRMSangle = 135; showlabelsRMS = 1; showlabelsSTD = 1; showlabelsCOR = 1; colSTD = [0 0 0]; colRMS = [0 .6 0]; colCOR = [0 0 1]; tickCOR(1).val = [1 .99 .95 .9:-.1:0]; tickCOR(2).val = [1 .99 .95 .9:-.1:0 -.1:-.1:-.9 -.95 -.99 -1]; widthCOR = .8; widthRMS = .8; widthSTD = .8; styleCOR = '-.'; styleRMS = '--'; styleSTD = ':'; titleRMS = 1; titleCOR = 1; titleSTD = 1; tickRMS = tick; rincRMS = rinc; tickSTD = tick; rincSTD = rinc; %% LOAD CUSTOM OPTIONS: nopt = narg/2; for iopt = 4 : 2 : narg+3 optname = varargin{iopt}; optvalue = varargin{iopt+1}; switch lower(optname) case 'tickrms' tickRMS = sort(optvalue); rincRMS = (max(tickRMS)-min(tickRMS))/length(tickRMS); case 'showlabelsrms' showlabelsRMS = optvalue; case 'tickrmsangle' tickRMSangle = optvalue; case 'colrms' colRMS = optvalue; case 'widthrms' widthRMS = optvalue; case 'stylerms' styleRMS = optvalue; case 'titlerms' titleRMS = optvalue; case 'tickstd' tickSTD = sort(optvalue); rincSTD = (max(tickSTD)-min(tickSTD))/length(tickSTD); case 'showlabelsstd' showlabelsSTD = optvalue; case 'colstd' colstd = optvalue; case 'widthstd' widthSTD = optvalue; case 'stylestd' styleSTD = optvalue; case 'titlestd' titleSTD = optvalue; case 'npan' Npan = optvalue; case 'tickcor' tickCOR(Npan).val = optvalue; case 'colcor' colCOR = optvalue; case 'widthcor' widthCOR = optvalue; case 'stylecor' styleCOR = optvalue; case 'titlecor' titleCOR = optvalue; case 'showlabelscor' showlabelsCOR = optvalue; end end %% CONTINUE THE PLOT WITH UPDATED OPTIONS: % define a circle th = 0:pi/150:2*pi; xunit = cos(th); yunit = sin(th); % now really force points on x/y axes to lie on them exactly inds = 1:(length(th)-1)/4:length(th); xunit(inds(2:2:4)) = zeros(2,1); yunit(inds(1:2:5)) = zeros(3,1); % plot background if necessary if ~ischa