How to create an output from trained model using "nnstart" with new set of input data with unknown output?

Question

Ahmad Sedaghat 2024-3-15

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2094991-how-to-create-an-output-from-trained-model-using-nnstart-with-new-set-of-input-data-with-unknown-o

评论： Tejas 2024-4-2

Hello experts in MATLAB neural network. In nntool I introduce 3 files, "input", "target", and "sample". I use "input and target" to train my nn model. Then I use a completely new set of data (either larger or smaller number of data sets but with the same input variables) called "sample" and I extract my output called "predict" which I never given to nntool. How can I do the same with nnstart?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Tejas 2024-3-28

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2094991-how-to-create-an-output-from-trained-model-using-nnstart-with-new-set-of-input-data-with-unknown-o#answer_1432431

在 MATLAB Online 中打开

Hello Ahmad,

To train a neural network and obtain predictions using the ‘nnstart’ tool, simply type nnstart in the MATLAB command line. This action will launch the Neural Network Start GUI. Here is a step-by-step guide to navigate through the process:

First, select the type of neural network you wish to train.
At the top bar, choose the 'Import' option and add your ‘input’ and ‘target’ files.
Next, click on the 'Train' option in the top bar of the GUI. This will initiate the training of the model with your input data.
After the training is complete, click on the 'Export Model' option and then select 'Export to Workspace'.
You will be prompted to provide a name for the struct variable where this neural network will be saved. The default name is ‘results’.
The ‘results’ will contain the network weights under the name ‘Network’ and ‘TrainingResults’ in the form of a structure.
To feed a sample dataset into this model, use the following syntax:

predict = results.Network(sample);

This will store the prediction in the 'predict' variable.

For additional details on the ‘nnstart’ tool, please refer to this documentation:

https://www.mathworks.com/help/releases/R2022b/deeplearning/ref/nnstart.html .

Hope it helps!

4 个评论
显示 2更早的评论隐藏 2更早的评论

Ahmad Sedaghat 2024-3-30

I get either a function or a script. For function I get the following. How do I use it to run for a sample?

function [Y,Xf,Af] = trainedModel(X,Xi,~)

%MYNEURALNETWORKFUNCTION neural network simulation function.

%

% Auto-generated by MATLAB, 30-Mar-2024 00:48:57.

%

% [Y,Xf,Af] = myNeuralNetworkFunction(X,Xi,~) takes these arguments:

%

% X = 2xTS cell, 2 inputs over TS timesteps

% Each X{1,ts} = 8xQ matrix, input #1 at timestep ts.

% Each X{2,ts} = 8xQ matrix, input #2 at timestep ts.

%

% Xi = 2x2 cell 2, initial 2 input delay states.

% Each Xi{1,ts} = 8xQ matrix, initial states for input #1.

% Each Xi{2,ts} = 8xQ matrix, initial states for input #2.

%

% Ai = 2x0 cell 2, initial 2 layer delay states.

% Each Ai{1,ts} = 10xQ matrix, initial states for layer #1.

% Each Ai{2,ts} = 8xQ matrix, initial states for layer #2.

%

% and returns:

% Y = 1xTS cell of 2 outputs over TS timesteps.

% Each Y{1,ts} = 8xQ matrix, output #1 at timestep ts.

%

% Xf = 2x2 cell 2, final 2 input delay states.

% Each Xf{1,ts} = 8xQ matrix, final states for input #1.

% Each Xf{2,ts} = 8xQ matrix, final states for input #2.

%

% Af = 2x0 cell 2, final 0 layer delay states.

% Each Af{1ts} = 10xQ matrix, final states for layer #1.

% Each Af{2ts} = 8xQ matrix, final states for layer #2.

%

% where Q is number of samples (or series) and TS is the number of timesteps.

%#ok<*RPMT0>

% ===== NEURAL NETWORK CONSTANTS =====

% Input 1

x1_step1.xoffset = [1;6.193857405;6.594200472;1.4;10;0;0.1;0];

x1_step1.gain = [0.000228336568101381;0.0417105921738372;0.0242836806711481;0.0429184549356223;0.023121387283237;0.000556669762247836;0.149253731343284;0.00557103064066852];

x1_step1.ymin = -1;

% Input 2

x2_step1.xoffset = [1;6.193857405;6.594200472;1.4;10;0;0.1;0];

x2_step1.gain = [0.000228336568101381;0.0417105921738372;0.0242836806711481;0.0429184549356223;0.023121387283237;0.000556669762247836;0.149253731343284;0.00557103064066852];

x2_step1.ymin = -1;

% Layer 1

b1 = [-0.74564701442649261587;-3.3807811051649609979;12.696997215926753455;5.124594815958134042;-0.31217788928837814089;0.021269942789951909456;2.2280955432889344436;-0.23267974583429620883;-0.0087410257326856043852;3.2655375612259320839];

IW1_1 = [0.40957574758027120998 0.301012017255419706 0.012760664581249381647 -0.63234786046405044857 -0.36731524358903300254 -0.1374720264247825674 0.043624682519432057826 0.098111130922637415641 0.44084716900365700232 -0.081460072601820324167 0.059278810794876804491 0.50342646117271505624 -0.041512572897196370336 0.28231149194836535621 -0.12361881807838726033 -0.068584359128335894984;-0.87266344800095774215 4.3404130642184517441 6.9289446298660957169 12.543469224859304489 -0.90983886528274204863 -4.3213271614014860944 -0.61309663941533243392 -0.022079901213139511984 -0.56720557879541866342 -2.4782426172534774977 -5.9868747489897060632 -13.351304994699644269 0.31524178176367056992 0.97083852718641372448 0.39413629838130304073 0.39282383673591125817;-0.12614272401191783946 0.077970456226401224642 0.24293753444437274358 -0.86402311954972998986 0.82545516485622627556 4.0198833251171537384 -0.51491284679306048577 -0.0015418707036749599703 0.046264035438771500264 1.811532751660775542 -0.073396774372224554694 -0.20411523675238493758 -0.78490746343002104624 1.2896396885164709278 0.17124936426806808276 -0.13783459320409707782;0.094305449030925248 12.810989320847479433 8.3520341509207689512 -1.2987129469455918063 -0.22301496123794517512 4.9792146782880415401 -0.45763135493911055729 0.074854597547909282329 0.3606159670984981358 4.3006181576209563389 -3.2139539993047723243 -13.528362776324982519 -3.6960878279692419568 -4.8355688290575624677 0.98537725230338124582 -0.027361740989907325228;0.0018034753830968198052 -0.68408718698591242102 -0.3829312439320466388 0.82073374166721513312 0.29792857308188280818 -0.046384322711338243916 -0.60035760403244264882 1.2534490142989465067 -0.082492501081115235384 0.61237967419217897991 -0.020937882082933283179 -0.74036932037841285403 0.31616932073470394027 0.31956206487199190214 -0.12226530818383027355 -0.78857836663683078537;-0.29383479523330280347 -0.29886816860932419182 -0.34488732181761205808 0.029208537590726901118 0.15004137939504152977 -0.30427007388724902626 -0.34688454008194485878 0.19723641379588380262 -0.33579852196779541584 0.18515200022545885705 -0.00012800596895965488131 0.29969429909817207891 -0.09681206486383786558 -0.053483663429865498096 0.23124503648993793847 0.14034536937389183175;0.044582652715479675898 1.6066784139101577544 0.20279357691274169095 -5.2442259473930148417 0.88243000467528365238 -0.61309217487678213665 -0.41876580803647378559 -0.21618665991145180616 -0.25758810335795129243 2.075832290544280756 1.1234018148878739396 2.1233552629662657552 -1.9036058618092415173 0.84213855345363797245 0.29394426852626948099 0.20480082887435974159;0.52480924136866113905 -5.4809559718769254033 4.5194335281092916645 5.8647533237626330305 -2.0366709109032425928 -0.03230609582282549519 -0.45153114887074347061 -0.41073787161036301452 -0.29637452863745211706 4.6307285727632629957 -6.1897665100087602141 -4.9375418589340762665 3.0572208702528951285 0.86598908243470218427 0.088502782264388554778 0.018384056907826663851;-0.25941227099638924924 0.063406182949320835651 0.2898381096706423854 0.24973696881609430687 0.26029070601865406109 -0.40185491662611277919 0.015474073295345712811 -0.02508452775774228094 0.077690403050594561418 -0.30565638309244791415 0.41188126482277648277 -0.44787653693434970048 -0.24252597465925465614 -0.30311943160463483693 0.099002080551864629632 -0.024950089213650807063;0.030759135019659097215 0.94181177269536908714 3.6670327629645007583 -11.415418738248394703 -0.69896391614851804963 1.5000929045007520557 -0.70203127176361501682 0.21463474573401522538 -0.15313332384712705081 4.3391630473931526168 -2.213139985476724636 7.3148054494482455112 1.1760794888162648952 -1.4745394717106763416 0.63411206065683889399 -0.018336766470073331825];

IW1_2 = [0.45027388189551742492 -0.15852801255842990935 -0.18609364727177077348 -0.17561782204204934299 0.16837450190128627758 0.22851425836508831857 -0.041943113144883588383 -0.090302365049412816611 0.34396646901746591451 0.19343404793062582425 0.25468606340414329248 0.0025276393312470071817 0.17105694848600205837 -0.30597890471533917278 0.12389756943607706241 0.061501483808284518329;-1.3728824510181860674 4.4673370187048799806 7.0276169331556346265 12.736811827044830281 -0.51013745636253216009 -4.1709574704372220921 -0.31347120706149989022 -0.077928843119455859512 -0.90472211140410685459 -2.5286849788290415297 -6.5067099679789706101 -13.256573300584872044 -0.21796187219574009175 1.2231444212962074847 0.36149963970403287128 -0.22325279702952233185;-0.25437147246944058487 -0.31158402374328453277 0.60707901941238839871 -1.6844381193085358195 0.80141072198180240971 4.2465095709269462532 0.23891394212372749073 -0.011162087986435319936 0.39040841292426203823 2.1437717624575505226 0.19431567685076875374 -0.10456104405412595948 -1.0387512136289052123 1.5407254608248621697 0.10767662429159777226 0.18964517606363953672;0.048659095247014103047 13.056140025540624805 8.2536014610758527965 -2.0917423014896132649 -0.39775826458192387047 5.2745734208552201849 -0.76890085880214875669 0.018359457104715429937 -0.047676695695904977579 4.8832075052038170782 -3.162542868143285002 -13.699130844523898176 -3.8505399329144922582 -5.053663714513916716 0.67240715919014859558 0.41991001379602049814;0.33747423938282433697 -1.1474387235490939307 -0.052096559191877396533 1.3561688889839746963 -0.04063719312689976354 -0.11644402305866034808 0.31847147705416845032 1.7845723937867581377 -0.23143773176002999103 0.30118927256016492233 0.071676173687516289701 -0.72535285061859866573 -0.40460258676065752326 -0.060908838105890712733 0.46444808351938748903 -0.37978938236604126155;0.26037007147038665478 0.30778752637179662699 0.34242846805777726882 -0.024895239461724501945 -0.15027067863755394495 0.30027830211899053658 0.3472317157030836432 -0.19724341604467018896 0.1772084208113364201 -0.18898483262770263136 0.0012437928481804187697 -0.30637756150111344899 0.097550915541807003972 0.051970399968714373062 -0.23148179441726871031 -0.14040346606786519024;-0.025767834508597441817 1.5722517997246980848 0.072872997155605745423 -5.0074741507044766564 1.4748599781434839073 -0.65595457011106750311 -0.18049381385230520158 0.13254481491048408781 0.28762856264705105591 2.5609881681715696011 1.3034877989748223648 1.7096390080014400858 -1.8300730230173241075 1.6974661616201545122 0.27445591941233288269 0.003466366557726536634;-0.29274352193631486552 -5.9356402957429894585 4.5742567911115408208 6.0276476558534666239 -1.7651389009219586868 -0.65420478947856552931 -0.37488696993546277714 0.32744653885925362191 0.22702081129889017941 4.9970449875415221541 -5.5470200276101744308 -4.8595673125976057705 3.2448051006448808131 1.1841114439054023499 0.65095129304911791834 0.17883127367264753382;-0.091640448272191254997 0.61733629854507277024 -0.46256249878367911021 -0.096647615192860941713 -0.16132382041499823688 0.19957594866539010714 0.015228027532853471299 0.02691597503798133853 -0.39348815995164088122 -0.14835683714733230443 -0.22201185698056391526 0.15852918281460037786 0.079164197869389274498 0.22235575174995250713 -0.12227932631048023249 0.016725043257196484797;0.29596373764304850784 1.436223060250261474 3.7791089637525665523 -11.546784149862991953 -1.1159433577299526785 2.0358271903818541126 0.060484066477757607305 -0.33515736836658810427 -0.090689260878449251546 4.4079454384035043901 -2.5206891428631919005 7.370223536368952999 0.913092837798264223 -1.3085283601084902738 -0.0034593534307655413734 0.23352480390266031063];

% Layer 2

b2 = [0.12126044776618830068;-0.43953616222107300127;-0.34803718137709632741;-0.31522252311469378672;-0.27683913416911037908;-1.8877011899976434695;-1.1291005859790823607;0.10442226002437747512];

LW2_1 = [-0.0020632840476992931646 0.00019995611827562680867 -0.0013069253671432381342 0.00044729898732825598955 -2.8557554176220376556e-05 -5.6668587754342425455 7.9939433168376881293e-05 0.0037027391640624541906 0.12847669923354496091 0.00098578777974368293757;-1.6123411029436434116 0.089746353146229196351 -0.14094762420972525852 0.11498522003309508133 -0.018688127012124206711 -4.0557047343705630027 0.30725814383470562152 -0.070134260577484980326 -1.4245069133806955719 -0.093265225444317215175;1.4727717296487634613 -0.061167929195984842294 0.46481706092491997806 -0.12999438442841601549 0.0067825451836336023148 5.7122573205690692077 -0.28719853564913344623 -0.0018390056090217491575 0.52534088144817658872 0.2245230396730550082;-1.7321455125257800844 0.12491886114300818078 -0.141275329420620821 0.093158767721493646041 -0.018879355161943948277 -3.7443888676306631424 0.16853264443904616976 -0.01362519212355450815 -1.7814828606704862235 -0.073782578373859680765;1.8020573723565211921 -0.070874686781871185093 0.49748731466903478582 -0.12133307116092452438 0.010117027249749952281 6.6735897757492930182 -0.19959340472577327685 -0.030748811194918866663 0.74037792724440809611 0.21740026062640838145;-1.1659153725188242667 -0.048218710980271098365 1.2179836210374754835 0.030625751393465690253 -0.0082046602919589647213 7.5091801127596449206 -0.21447538198618165683 -0.11397282170787839672 -4.0918674751413259827 -0.12515091130036049005;-0.7238552540580567296 -0.074261618103721657724 0.47439168330151088027 -0.026366843288217468172 0.0043893543755136252282 -4.0696509499894064632 0.086673529377084834735 -0.0041894922628651553795 0.38657743630255875233 -0.0066502116318456067079;0.18673592139670339152 -0.0029064305247215645872 0.048222854119912852977 -0.0019582448785448814191 0.57134363577213242991 0.78368824439956275008 0.0090698090009147386281 -0.024875640327847232947 0.058396843832262901319 0.017312172549151860074];

% Output 1

y1_step1.ymin = -1;

y1_step1.gain = [0.000228336568101381;0.0417105921738372;0.0242836806711481;0.0429184549356223;0.023121387283237;0.000556669762247836;0.149253731343284;0.00557103064066852];

y1_step1.xoffset = [1;6.193857405;6.594200472;1.4;10;0;0.1;0];

% ===== SIMULATION ========

% Format Input Arguments

isCellX = iscell(X);

if ~isCellX

X = {X};

end

if (nargin < 2), error('Initial input states Xi argument needed.'); end

% Dimensions

TS = size(X,2); % timesteps

if ~isempty(X)

Q = size(X{1},2); % samples/series

elseif ~isempty(Xi)

Q = size(Xi{1},2);

else

Q = 0;

end

% Input 1 Delay States

Xd1 = cell(1,3);

for ts=1:2

Xd1{ts} = mapminmax_apply(Xi{1,ts},x1_step1);

end

% Input 2 Delay States

Xd2 = cell(1,3);

for ts=1:2

Xd2{ts} = mapminmax_apply(Xi{2,ts},x2_step1);

end

% Allocate Outputs

Y = cell(1,TS);

% Time loop

for ts=1:TS

% Rotating delay state position

xdts = mod(ts+1,3)+1;

% Input 1

Xd1{xdts} = mapminmax_apply(X{1,ts},x1_step1);

% Input 2

Xd2{xdts} = mapminmax_apply(X{2,ts},x2_step1);

% Layer 1

tapdelay1 = cat(1,Xd1{mod(xdts-[1 2]-1,3)+1});

tapdelay2 = cat(1,Xd2{mod(xdts-[1 2]-1,3)+1});

a1 = tansig_apply(repmat(b1,1,Q) + IW1_1*tapdelay1 + IW1_2*tapdelay2);

% Layer 2

a2 = repmat(b2,1,Q) + LW2_1*a1;

% Output 1

Y{1,ts} = mapminmax_reverse(a2,y1_step1);

end

% Final Delay States

finalxts = TS+(1: 2);

xits = finalxts(finalxts<=2);

xts = finalxts(finalxts>2)-2;

Xf = [Xi(:,xits) X(:,xts)];

Af = cell(2,0);

% Format Output Arguments

if ~isCellX

Y = cell2mat(Y);

end

% ===== MODULE FUNCTIONS ========

% Map Minimum and Maximum Input Processing Function

function y = mapminmax_apply(x,settings)

y = bsxfun(@minus,x,settings.xoffset);

y = bsxfun(@times,y,settings.gain);

y = bsxfun(@plus,y,settings.ymin);

end

% Sigmoid Symmetric Transfer Function

function a = tansig_apply(n,~)

a = 2 ./ (1 + exp(-2*n)) - 1;

end

% Map Minimum and Maximum Output Reverse-Processing Function

function x = mapminmax_reverse(y,settings)

x = bsxfun(@minus,y,settings.ymin);

x = bsxfun(@rdivide,x,settings.gain);

x = bsxfun(@plus,x,settings.xoffset);

end

Ahmad Sedaghat 2024-4-1

This is not Regression Learner. I am using nnstart. The menu do not give any Export. They only provide a function or a script. What version do you use?

Tejas 2024-4-2

I am currently working with MATLAB R2022b.

If the "Export to Workspace" option is missing, it may prove beneficial to reinstall or update MATLAB to newer versions for access to enhanced features and compatibility. This method simplifies the process, as selecting the "Export to Workspace" option immediately provides a structure into which the sample dataset can be input with minimal need for data conversion.

To obtain predictions from the function,save the function in a MATLAB file with the same name as the function itself. The following syntax can then be utilized: [Y, Xf, Af] = trainedModel(X, Xi, []);

This syntax is consistent with the code previously provided. To ensure the sample dataset is compatible with the model intended for predictions, appropriate conversion of the dataset is necessary. For guidance on this conversion, refer to the specific documentation for the desired neural network model. After inputting the dataset into the function, the variable "Y" will contain the predictions.

请先登录，再进行评论。

How to create an output from trained model using "nnstart" with new set of input data with unknown output?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

回答（1 个）

4 个评论
显示 2更早的评论隐藏 2更早的评论

另请参阅

类别

标签

产品

版本

Community Treasure Hunt

How to create an output from trained model using "nnstart" with new set of input data with unknown output?

0 个评论 显示 -2更早的评论隐藏 -2更早的评论

回答（1 个）

4 个评论 显示 2更早的评论隐藏 2更早的评论

另请参阅

类别

标签

产品

版本

Community Treasure Hunt

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

4 个评论
显示 2更早的评论隐藏 2更早的评论