diff --git a/INSTALL.md b/INSTALL.md
index 37bc0c9a0..2bdfb7fe3 100644
--- a/INSTALL.md
+++ b/INSTALL.md
@@ -7,7 +7,7 @@
 - yacs
 - matplotlib
 - GCC >= 4.9
-- (optional) OpenCV for the webcam demo
+- OpenCV
 
 
 ### Option 1: Step-by-step installation
diff --git a/README.md b/README.md
index 5c68de5f0..b7cda0a62 100644
--- a/README.md
+++ b/README.md
@@ -138,7 +138,7 @@ and we have divided the learning rate by 8x.
 We also changed the batch size during testing, but that is generally not necessary because testing
 requires much less memory than training.
 
-Furthermore, we set ```MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN 2000``` as the proposals are selected for per the batch rather than per image. The value is calculated by **1000 x images-per-gpu**. Here we have 2 images per GPU, therefore we set the number as 1000 x 2 = 2000. If we have 8 images per GPU, the value should be set as 8000. See [#672](https://github.com/facebookresearch/maskrcnn-benchmark/issues/672) for more details.
+Furthermore, we set `MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN 2000` as the proposals are selected for per the batch rather than per image in the default training. The value is calculated by **1000 x images-per-gpu**. Here we have 2 images per GPU, therefore we set the number as 1000 x 2 = 2000. If we have 8 images per GPU, the value should be set as 8000. Note that this does not apply if `MODEL.RPN.FPN_POST_NMS_PER_BATCH` is set to `False` during training. See [#672](https://github.com/facebookresearch/maskrcnn-benchmark/issues/672) for more details.
 
 ### Multi-GPU training
 We use internally `torch.distributed.launch` in order to launch
@@ -150,7 +150,7 @@ process will only use a single GPU.
 export NGPUS=8
 python -m torch.distributed.launch --nproc_per_node=$NGPUS /path_to_maskrcnn_benchmark/tools/train_net.py --config-file "path/to/config/file.yaml" MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN images_per_gpu x 1000
 ```
-Note we should set ```MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN``` follow the rule in Single-GPU training.
+Note we should set `MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN` follow the rule in Single-GPU training.
 
 ## Abstractions
 For more information on some of the main abstractions in our implementation, see [ABSTRACTIONS.md](ABSTRACTIONS.md).
diff --git a/maskrcnn_benchmark/utils/model_zoo.py b/maskrcnn_benchmark/utils/model_zoo.py
index bd644d7cc..2128ad7bb 100644
--- a/maskrcnn_benchmark/utils/model_zoo.py
+++ b/maskrcnn_benchmark/utils/model_zoo.py
@@ -3,13 +3,13 @@
 import sys
 
 try:
-    from torch.utils.model_zoo import _download_url_to_file
-    from torch.utils.model_zoo import urlparse
-    from torch.utils.model_zoo import HASH_REGEX
-except:
     from torch.hub import _download_url_to_file
     from torch.hub import urlparse
     from torch.hub import HASH_REGEX
+except ImportError:
+    from torch.utils.model_zoo import _download_url_to_file
+    from torch.utils.model_zoo import urlparse
+    from torch.utils.model_zoo import HASH_REGEX
 
 from maskrcnn_benchmark.utils.comm import is_main_process
 from maskrcnn_benchmark.utils.comm import synchronize
@@ -35,8 +35,8 @@ def cache_url(url, model_dir=None, progress=True):
         >>> cached_file = maskrcnn_benchmark.utils.model_zoo.cache_url('https://s3.amazonaws.com/pytorch/models/resnet18-5c106cde.pth')
     """
     if model_dir is None:
-        torch_home = os.path.expanduser(os.getenv('TORCH_HOME', '~/.torch'))
-        model_dir = os.getenv('TORCH_MODEL_ZOO', os.path.join(torch_home, 'models'))
+        torch_home = os.path.expanduser(os.getenv("TORCH_HOME", "~/.torch"))
+        model_dir = os.getenv("TORCH_MODEL_ZOO", os.path.join(torch_home, "models"))
     if not os.path.exists(model_dir):
         os.makedirs(model_dir)
     parts = urlparse(url)