FIX small bugs in `run_classifier_pytorch.py`

run_classifier_pytorch.py

@@ -512,7 +512,7 @@ def main():
         train_dataloader = DataLoader(train_data, sampler=train_sampler, batch_size=args.train_batch_size)
 
         model.train()
-        for epoch in range(args.num_train_epochs):
+        for epoch in range(int(args.num_train_epochs)):
             for input_ids, input_mask, segment_ids, label_ids in train_dataloader:
                 input_ids = input_ids.to(device)
                 input_mask = input_mask.float().to(device)