Mini batch accuracy in toggling between two values and settled in 75 percent. Please provide some suggestions to increase the accuracy and also not getting good testing accuracy.

Question

Chandran Venkatesan 2020-9-25

回答： Srivardhan Gadila 2020-9-30

Answer 1

Srivardhan Gadila 2020-9-30

The following are few suggestions:

Make sure that all the classes have equal number of observations.
Check how trainNetwork uses an augmented image datastore to transform training data for each epoch: Augment Images for Training with Random Geometric Transformations. Then try training the network with augmentedImageDatastore for more epochs.
Try changing the network architecture itself if there is no improvement in the accuracy when augmentedImageDatastore is used. You can refer to Choose Network Architecture.
Try Using dropout layers & increasing global L2 regularization factor in new architecture. For more information, see dropoutLayer & 'L2Regularization' option in trainingOptions.