Implementing different length arrays into Classification Learner
1 vue (au cours des 30 derniers jours)
Afficher commentaires plus anciens
Hi all, I am new to the whole deep learning stuff and I am just looking for some clarification. I have been collecting data about machine failures, the data tends to vary in length due to the nature of how the machine fails. In order to use this data with the classification learner I have been stretching the data to equal lengths using imresize(A, [B,C], 'bilinear') and then compiling the data into 1 array. I have added an extra column in order to classify each row based on a scale of 0-100 to try predict a wear percent essentially. As I am currently still learning how to use the software I have only been using a small data pool size of 9 samples, the samples are recorded from the machine at 0% wear up to failure. Am I on the right track for implementing the classification learner?
Along with this I have used the 'All Quick-To-Train' and have got an accuracy of 85.0% with accuracy percentages on the confusion matrix around 90% at wear of 70% and above. Am I correct in thinking that I can export this model and then input data points in order to output a predicted wear level of the machine?
Thank you
10 commentaires
Stephan
le 8 Jan 2019
There is also a regression learner app... It is the same look and feel - should be no problem.
Réponses (0)
Voir également
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!