Second Prediction for CNN network

1 次查看（过去 30 天）
Tsvetislava Ivanova
编辑： Tsvetislava Ivanova 2022-7-9
Hello,
I am developing an artificial intelligence system for recognizing dog breeds, which is currently still being tested and working with flowers. It is made through a Convolutional Neural Network.
I have the first predicted result, which is considered the closest in percentage. Can anyone help me to make a second and possibly a third prediction?
This is the code brief for the result field:
% Button pushed function: pushbutton2
function pushbutton2_Callback(app, event)
net = app.net;
I = app.I;
I= imresize(I,[500,500],'nearest');
[Pred,scores] = classify(net,I);
scores = max(double(scores*100));
app.edit2.Value = scores + "%" + " " + string(Pred);
*edit2 is the the field for the first prediction*
Thanks in advance :)

请先登录，再回答此问题。

回答（1 个）

Image Analyst
Image Analyst 2022-7-9
Try maxk
topScores = maxk(scores , 3);
  1 个评论
Tsvetislava Ivanova
Tsvetislava Ivanova 2022-7-9
编辑：Tsvetislava Ivanova 2022-7-9
Hello, and thank you for your responce.
For this field i wrote:
[YPred2,topScores] = classify(net,I);
YPred2 = maxk(YPred , 2);
topScores = maxk(scores , 2);
app.edit2_2.Value = topScores + "%" + " " + string(YPred2);
Am i on the right way? I am getting an error...
(Error using categorical/maxk
Relational comparisons are not allowed for categorical arrays that are not ordinal.)
I will be glad to help :)
Sorry, but I'm a beginner..:)

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Delaunay Triangulation 的更多信息

标签

产品


版本

R2019a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by