How to improve my graph?
3 views (last 30 days)
Show older comments
Hello guys, I made an object detector YOLOv4 and after runnin this code:
[ap,recall,precision] = evaluateDetectionPrecision(detectionResults,testData);
recallv = cell2mat(recall);
precisionv = cell2mat(precision);
[r,index] = sort(recallv);
p = precisionv(index);
figure
plot(r,p)
xlabel("Recall")
ylabel("Precision")
grid on
title(sprintf("Average Precision = %.2f",mean(ap)))
My graph looks like this :
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/1390514/image.png)
But in Matlab example their graph looks like this :
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/1390519/image.jpeg)
I have 5 classes in my object detector...
What can be wrong ? Thanks for help
1 Comment
Ranjeet
on 1 Jun 2023
Hi Adrian,
Testing the following script in R2022b, I do not face any issues.
However, I wanted to ask "Why are you using cell2mat with recall and precision? They are already be a returned as a vector and not cell. Is the operation not throwing any error?"
Also, providing the script that is being used for training might help.
Answers (0)
See Also
Categories
Find more on Recognition, Object Detection, and Semantic Segmentation in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!