Torch.manual_seed(3407) is all you need: On the influence of random seeds in deep learning architectures for computer vision
In this paper i investigate the effect of random seed selection on the accuracy when using popular deep learning architectures for computer vision.
The conclusions are that even if the variance is not very large, it issurprisingly easy to find an outlier that performs much better or much worsethan the average.
The conclusions are that even if the variance is not very large, it issurprisingly easy to find an outlier that performs much better or much worsethan the average.