set cudnn.benchmark to True for speed up training process (#1811)

set cudnn.benchmark to True for speed up training process

Co-authored-by: Brian Johnson <brianjo@fb.com>

Author: developer0hye

beginner_source/transfer_learning_tutorial.py

@@ -39,6 +39,7 @@
 import torch.nn as nn
 import torch.optim as optim
 from torch.optim import lr_scheduler
+import torch.backends.cudnn as cudnn
 import numpy as np
 import torchvision
 from torchvision import datasets, models, transforms
@@ -47,6 +48,7 @@
 import os
 import copy
 
+cudnn.benchmark = True
 plt.ion()   # interactive mode
 
 ######################################################################