So according to the pytorch documentation, it offers a lot of different network pruning techniques. 2 of them being prune.L1Unstructured which Prune (currently unpruned) units in a tensor by zeroing out the ones with the lowest L1-norm and there’s prune.l1_unstructured which Prunes tensor by removing units with the lowest L1-norm. I want to know how these two are different? And if they cause a change on the neural net’s architecture?
Can someone help?