CPU Memory increasing when training

[YRlin-12]

Hi! I'm training the pretraining process on RefCOCO* datasets. But, the CPU memory increasing from 18G to 22G after 1 epoch. I'm worry about after a few epochs later, the CPU MEMORY will face OOM problem. (My device is 32G CPU) Do you have any idea about this?

[wjn922]

Hi, I suppose you mean the GPU memory. We also use the GPUs with 32G memory, so we think there won't be a problem.

[YRlin-12]

Sorry, I don't mean the GPU memory.