How many images in the training set for each class?

Hi,thanks for sharing the code and model, it helps me a lot
Can you tell me how many images in the training set for the 5 classes, I'm not familiar about keras, does the code below means that only 500*batch_size images are trained every epoch, and not every images in the training set of [nsfw_data_scraper](https://github.com/alexkimxyz/nsfw_data_scraper) is trained in an epoch?

btw,I test the model on the test set(2000 images for each class), the correct rate of class neutral is 0.1 lower than the result confusion matrix shows while other classes performs well

Start with the training code shown in the issue and inspect how it loads the nsfw_data_scraper dataset and sets the epoch or batch limits. Determine the per-class image counts and whether the reported neutral-class discrepancy is explained by the evaluation setup. Done means documenting those findings and clarifying the confusion-matrix comparison.

Written by the indexing model from the issue text.