Mnist And Transfer Learning With Vgg16 In Keras- Low Validation Accuracy

I recently started taking advantage of Keras's flow_from_dataframe() feature for a project, and decided to test it with the MNIST dataset. I have a directory full of the MNIST sam

Solution 1:

Have you tried explicitly defining the classes of the images? as such:

train_generator=image.ImageDataGenerator().flow_from_dataframe(classes=[0,1,2,3,4,5,6,7,8,9])

in both the train and validation generators.

I have found that sometimes the train and validation generators create different correspondence dictionaries.