The inference-time model's output are not the same as the training-time model when the batch size is large

[Harry-zzh]

Hi, I try python repmlpnet.py, and set the size of x at [100, 3, 224, 224]. But I find the deviation between the new_y and origin_y is 0.0019, which is not negligible. I wonder why it happens, thank you in advance.