简体   繁体   中英

Imbalanced data classification in Keras

I have a dataset consisting of classes A , B , and D . There are 100 samples A , 26 samples B , and 1 sample D out 127. If I ignore sample D, accuracy is around %95 with ANN. However, I would like to build a model that can correctly classify sample D and be successful to predict D samples in future. Any suggestion?

You can augument your D very heavily to artificially "create" more instances of D, but in general you must have enough data to teach the algorithm the variability of all classes.

If you can assume your algorithm is being shown only images from A, B and D, you can try to teach it to classify "neither A nor B" as D, with methods similar to the ones described here .

The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM