update docs (#378)

examples/ResNet/README.md

@@ -26,10 +26,11 @@ See the [tutorial](http://tensorpack.readthedocs.io/en/latest/tutorial/efficient
 ## load-resnet.py
 
 This script only converts and runs ImageNet-ResNet{50,101,152} Caffe models [released by Kaiming](https://github.com/KaimingHe/deep-residual-networks).
+Note that the architecture is different from the `imagenet-resnet.py` script and the models are not compatible.
 
-Example usage:
+Usage:
 ```bash
-# convert caffe model to npy format
+# download and convert caffe model to npy format
 python -m tensorpack.utils.loadcaffe PATH/TO/{ResNet-101-deploy.prototxt,ResNet-101-model.caffemodel} ResNet101.npy
 # run on an image
 ./load-resnet.py --load ResNet-101.npy --input cat.jpg --depth 101