massive datasets loading and training

[davidvct]

if I have a massive datasets of images, and unable to fit all of them into memory, how do I load and train them by batches?
my folder structure is:

datasets
|
|----Train
| |---- images
| |-----masks
|
|---- Val
|---- images
|---- masks

[Vipin2705]

If you are using GPU instead of loading all image at ones, try using datagenerator it allows you to supply images to GPU in batches, earlier I used to get OOM issue, now I am able to train my model.