Timeline for Training Image Classifier with 7 classes but my model is overfitting resulting the accuracy of the model to behave weirdly during training
Current License: CC BY-SA 4.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 8 at 11:32 | comment | added | nbro | So, basically, your (training?) dataset consists of 308 * 7 images in total. That might not be enough to train the model for your purposes. If you can't collect more images, you could try to augment the dataset by creating more images from the existing ones by e.g. rotating even more the existing ones, blurring, or whatever. Also, how many images are you using for validation or testing? | |
| S Aug 7 at 21:38 | review | First questions | |||
| Aug 21 at 21:49 | |||||
| S Aug 7 at 21:38 | history | asked | rekoilS | CC BY-SA 4.0 |