fix tqdm usage (#246)

maskrcnn_benchmark/engine/inference.py

@@ -17,7 +17,7 @@ def compute_on_dataset(model, data_loader, device):
     model.eval()
     results_dict = {}
     cpu_device = torch.device("cpu")
-    for i, batch in tqdm(enumerate(data_loader)):
+    for i, batch in enumerate(tqdm(data_loader)):
         images, targets, image_ids = batch
         images = images.to(device)
         with torch.no_grad():