Memory error
Adilmar opened this issue · comments
Adilmar Coelho Dantas commented
Hi Congratulations on the project, I have the following problem when I use above 2000 images I have numpy error in this line.
train_tensors = paths_to_tensor (train_files) .astype ('float32') / 255
valid_tensors = paths_to_tensor (valid_files) .astype ('float32') / 255
My computer is an i7 with 8gb have idea how to solve.
Regards
Soheil commented
Hey there!
Sure. in here code will load all images into your memory and if you have a memory problem then you can use build-in data augmentation in Keras which it can be run even with weaker memory.
https://keras.io/preprocessing/image/
Good luck!