fix download link for features

README.md

@@ -90,16 +90,26 @@ computation cost.
 
 Make sure you are on a machine with a NVIDIA GPU and Python 2 with about 70 GB disk space.
 
-1. Install [PyTorch](http://pytorch.org/) with CUDA and Python 2.7.
+1. Install [PyTorch v0.3](http://pytorch.org/) with CUDA and Python 2.7.
 2. Install [h5py](http://docs.h5py.org/en/latest/build.html).
 
 #### Data Setup
 
-All data should be downloaded to a data/ directory in the root directory of this repository.
+All data should be downloaded to a 'data/' directory in the root
+directory of this repository.
 
-The easiest way to download the data is to run the provided script `tools/download.sh` from the repository root. If the script does not work, it should be easy to examine the script and modify the steps outlined in it according to your needs. Then run `tools/process.sh` from the repository root to process the data to the correct format.
+The easiest way to download the data is to run the provided script
+`tools/download.sh` from the repository root. The features are
+provided by and downloaded from the original authors'
+[repo](https://github.com/peteanderson80/bottom-up-attention). If the
+script does not work, it should be easy to examine the script and
+modify the steps outlined in it according to your needs. Then run
+`tools/process.sh` from the repository root to process the data to the
+correct format.
 
 #### Training
 
-Simply run `python main.py` to start training. The training and validation scores will be printed
-every epoch, and the best model will be saved under the directory "saved_models". The default flags should give you the result provided in the table above.
+Simply run `python main.py` to start training. The training and
+validation scores will be printed every epoch, and the best model will
+be saved under the directory "saved_models". The default flags should
+give you the result provided in the table above.

tools/download.sh

@@ -28,6 +28,6 @@ unzip data/v2_Annotations_Val_mscoco.zip -d data
 rm data/v2_Annotations_Val_mscoco.zip
 
 # Image Features
-wget -P data https://storage.googleapis.com/bottom-up-attention/trainval_36.zip
+wget -P data https://imagecaption.blob.core.windows.net/imagecaption/trainval_36.zip
 unzip data/trainval_36.zip -d data
 rm data/trainval_36.zip