why is error happen even I write the exact code for alexnet

Error using SeriesNetwork/predict (line 149)
For an image input layer, the input data for predict must a single image, a 4D array of images, or an
imageDatastore with the correct size.

Error in SeriesNetwork/classify (line 225)
            scores = this.predict( X, varargin{:} );

Error in amjh (line 10)

[Walter Roberson]

We need to see more of your code.

Also, which MATLAB release are you using?

[muhmed omer]

clear all,close,clc
camera=webcam;
net= alexnet;

while true

picture=camera.snapshot;
    picture= imresize(picture,[277,277]);

      lable=classify(net,picture);

      image(picture);
    title(char,(lable));
     drawnow;
  end

and I am using matlab 2016b

Answer 1

[Walter Roberson]

Change

      picture= imresize(picture,[277,277]);

to

      picture= imresize(picture,[227,227]);

That gets me past that point. I run out of GPU memory eventually.

Also you need to change to

      title(char(lable));

Answer 2

[Pradeep Punniyakotti]

To process color images in AlexNet, the input image size has to be 227 x 227 x 3.

227 x 227 indicates the height and width of the image and 3 represents the RGB intensities.

Resize your existing image using the imresize command.

To view the input layer size of AlexNet, use the command variable.Layers(1)

Answer 3

[abdelilah mbarek]

Hi; in alexNet you should resize or images. you can call functions that you create it in file resizeFunction.m for example:

            function I = resizeFunction(filename)
            I = imread(filename);
            I = imresize(I, [227 227]);

if you are using imageDatastore:

            images= imageDatastore();

use this call after:

            images.ReadFcn = @resizeFunction;