Fix a typo

This commit is contained in:
Ilya Kostrikov
2017-09-19 09:31:49 -04:00
committed by GitHub
parent 9e07584872
commit ccc261bcb9
+1 -1
View File
@@ -5,7 +5,7 @@ import torch.optim as optim
# TODO: In order to make this code faster:
# 1) Implement _extract_patches as a single cuda kernel_size
# 1) Implement _extract_patches as a single cuda kernel
# 2) Compute QR decomposition in a separate process
# 3) Actually make a general KFAC optimizer so it fits PyTorch