You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Slows down training substantially (at least for batch sizes of ~4). For example, in my own experiments I get ~2.5 batches/s per GPU without redrawing, and ~1.4 batches/s with redrawing.
I found one solution from pytorch GP, which dispatches to CPU for small QR factorizations:
Perhaps a similar strategy could be used? I think num_cols should never really be more than about ~100 though, so perhaps you should always use cpu here?
The text was updated successfully, but these errors were encountered:
Doing the QR-decomposition:
performer-pytorch/performer_pytorch/performer_pytorch.py
Lines 67 to 70 in f9765c4
Slows down training substantially (at least for batch sizes of ~4). For example, in my own experiments I get ~2.5 batches/s per GPU without redrawing, and ~1.4 batches/s with redrawing.
I found one solution from pytorch GP, which dispatches to CPU for small QR factorizations:
cornellius-gp/gpytorch#1224
Perhaps a similar strategy could be used? I think num_cols should never really be more than about ~100 though, so perhaps you should always use cpu here?
The text was updated successfully, but these errors were encountered: