I would like to use the confusion matrix as a measure of classification accuracy. So, I wonder whether is it reasonable to find the accumulative confusion matrix for k-folds and m-times of repeats or not.
The output matrix will consist of big numbers for the totals and at the same time percentages for all accuracy's, which is the most useful for me.
I would be grateful if you directed me to any reference that implemented such this work.
thanks in advance
Sadir Fadhil
yes, I end up with k x m number of confusion matrices. There is no calculation for the average CM, only total. My aim as I said before is to get different calculations represented on the graph as a percentage of the final accuracy, precision, recall, specificity, and others.
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Philosophy
- Philosophy Of Science
- Reasoning
More Sadir Fadhil's questions See All
Similar questions and discussions