Deep Learning Toolbox MLP
13 ビュー (過去 30 日間)
古いコメントを表示
I am using deep learning toolbox to make a Multi Layer Perceptron. However the accuracy that I see during training plot and then by using predict fuction are not same. Any idea how to predict for classification using Deep Learning toolbox. Because as of now I am not getting the class. I am only getting the scores
2 件のコメント
回答 (1 件)
Raunak Gupta
2020 年 9 月 5 日
Hi Shathesh,
The predict function will return the scores corresponding to each class for a particular test image. So, the highest score out of each row will correspond to class of that particular image. Alternatively, you can also use classify function which gives categorical prediction as well as scores for each test image. If you want only labels as output you can use following syntax.
% Only Labels
YPred = classify(net,XTest);
% Labels and scores
[YPred,scores] = classify(net,XTest);
As for the first query during training the accuracy is calculated for training and validation data (if given). So, the accuracy can differ with test data if test data is very different from training or validation data.
Hope this helps!
2 件のコメント
参考
カテゴリ
Help Center および File Exchange で Image Data Workflows についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!