Merge pull request #57 from kli-nlpr/patch-1

Update transfer_learning_tutorial.py

Author: chsasank

beginner_source/transfer_learning_tutorial.py

@@ -297,7 +297,8 @@ def optim_scheduler_ft(model, epoch, init_lr=0.001, lr_decay_epoch=7):
     param.requires_grad = False
 
 # Parameters of newly constructed modules have requires_grad=True by default
-model.fc = nn.Linear(512, 100)
+num_ftrs = model.fc.in_features
+model.fc = nn.Linear(num_ftrs, 2)
 
 if use_gpu:
     model = model.cuda()