Amend

National Institute for Materials Science

Actif depuis 2015

Followers: 0 Following: 0

Message

Statistiques

Afficher les badges

Feeds

Question

How to return estimated intervals or standard deviation for support machine regression model
For GPR model, we can use [ypred,ysd,yint] = predict(gprMdl,Xnew) to returns the estimated standard deviations and the 95% predi...

environ 7 ans il y a | 1 réponse | 0

0

réponse

Question

prediction without prior fitting in GPR
We can use _fitrgp_ to fit a Gaussian process regression model, then use this model to predict the output on new input. The proc...

presque 8 ans il y a | 1 réponse | 0

1

réponse

Question

how to use cross-validation in fitrgp
I find that there are two places in fitrgp() that we can do cross-validation: * cvgprMdl = fitrgp(x,y,'KernelFunction','squ...

environ 8 ans il y a | 1 réponse | 0

1

réponse

Question

how to understand generalized linear regression result
After using mdl=fitglm() to create generalized linear regression model, the matlab displays the regression results. How to under...

environ 8 ans il y a | 1 réponse | 0

0

réponse

Question

How to change numeric display in tree view.
I use command "view(model, 'mode','graph')" to display a tree. The default numeric format is short, which displays 5-digit scale...

plus de 8 ans il y a | 1 réponse | 0

0

réponse

Question

How does fsrnca() not work?
I have installed matlab R2016a and Statistics and Machine Learning Toolbox. I just simply try mdl = fsrnca(X,y) command (X is 10...

plus de 8 ans il y a | 1 réponse | 0

1

réponse

Question

how to deal with compiler error
I am a new one to matlab compiler. I just follow steps of "the matlab compiler user guide - Create Standalone Application from M...

plus de 8 ans il y a | 1 réponse | 0

1

réponse

Question

Are Matlab Compiler and Matlab Runtime the same?
I want to turn my GUI program into a standalone .exe file. So I can provide it to my end users who have no any Matlab products i...

plus de 9 ans il y a | 1 réponse | 0

1

réponse

Question

The effect of rng() on neural network training result
I use rng('default') to control random number generation, so each neural network has the same divide data (train, validation, te...

plus de 9 ans il y a | 1 réponse | 0

1

réponse

Question

Does ANN really needs pre-process?
Many papers suggest to do certain preprocessing steps to improve ANN performance. The common way is the normalization of input a...

plus de 9 ans il y a | 1 réponse | 0

0

réponse

Question

How to keep ANN training result stable?
When we train a neural network, normally we divide the data randomly into three (trainlm) or two (trainbr) groups (training data...

plus de 9 ans il y a | 1 réponse | 0

1

réponse

Question

how to produce different ensemble object by bag algorithm
Using fitensemble(X,y,'bag',NLearn, Learners) can produce an ensemble object. Changing Learners can obtain different objects. So...

plus de 9 ans il y a | 1 réponse | 0

0

réponse

Question

how does regression tree calculate "prunelist"
After creating regression tree, the prune sequence is established in "rtree.PruneList". Who can tell me how this PruneList is ca...

presque 10 ans il y a | 1 réponse | 0

0

réponse

Question

what does it mean: "MaxNumSplits is not a valid parameter name"
*I use follow code in command line* templ = templateTree('MaxNumSplits',4); ens = fitensemble(X,Y,'LSBoost',50,templ); ...

presque 10 ans il y a | 1 réponse | 0

1

réponse

Question

how to understand "Decision tree for classification (text description)"
The "Decision tree for classification (text description)" is difficult to understand. Who can give detail explanation. Thank ...

presque 10 ans il y a | 2 réponses | 0

2

réponses

Question

Why are pValue and FStat 'NaN'?
I am new to statistic tool. 1). I use "stepwiselm(ds1,'poly1432')" to perform regression and got some results like: Removi...

presque 10 ans il y a | 1 réponse | 0

0

réponse

Question

How do I perform nonlinear regression
I have 60 data pairs. Each data pair include 3 inputs and 1 output. The trends (2 dimension) between output and individual input...

presque 10 ans il y a | 1 réponse | 0

0

réponse

Question

how to improve data fit accuracy with neural network?
I have a one hidden layer network with 6 inputs and 4 outputs. I use neural network tool to perform 'Function Approximation and ...

environ 10 ans il y a | 1 réponse | 0

0

réponse

Réponse apportée
how many datasets are used to training neural network in one iteration?
*"Once randomly chosen, the trn/val/tst sets remain fixed.",* do you mean: 1. trn/val/tst data sets keep fixed until the er...

environ 10 ans il y a | 0

Réponse apportée
how many datasets are used to training neural network in one iteration?
*Thank you for your answer.* For above question,if I use default setting (net.trainParam.epochs=1000), each time network rando...

environ 10 ans il y a | 0

Question

how many datasets are used to training neural network in one iteration?
If neural network has two inputs and one output, and there are 100 datasets for two inputs and one output, how many datasets are...

environ 10 ans il y a | 3 réponses | 0

3

réponses

Question

how to display individual output regression in multiple output neural network
I create a neural network with 3 inputs, 1 hidden layer and 3 outputs. I use the "nftool" to train this neural network. When fin...

environ 10 ans il y a | 1 réponse | 0

1

réponse