Deep network designer - pretrained network

10 次查看（过去 30 天）
Arivan Ramachandran
Arivan Ramachandran 2020-11-26
回答： Raynier Suresh 2020-11-30
I have been working on a image classification problem, I trained resnet50 and GoogleNet using the input data. Can anyone help me on how to derive the results (F1, recall, accuracy and ROC). I would be really grateful.

请先登录，再回答此问题。

回答（1 个）

Raynier Suresh
Raynier Suresh 2020-11-30
For Accuracy:
accuracy = mean(YPred == YValidation)
Refer the below link for more information
https://www.mathworks.com/help/deeplearning/ref/alexnet.html
For Roc:
refer the below links
https://www.mathworks.com/help/deeplearning/ref/roc.html
https://www.mathworks.com/help/deeplearning/ref/plotroc.html
For Precision, Recall and F1 Score:
From the confusion matrix you can calculate precision and recall. Using Precision and Recall you can calculate the F1 score.
Refer the below links for confusion matrix
https://www.mathworks.com/help/deeplearning/ref/plotconfusion.html

类别

在 Help Center 和 File Exchange 中查找有关 Image Data Workflows 的更多信息

标签

产品


版本

R2020b

Community Treasure Hunt

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

Start Hunting!

Translated by