Doc fix.

README.md

@@ -4,9 +4,8 @@ Copyright (C) 2016, Ayan Chakrabarti <[email protected]>
 This is a reference implementation of the algorithm described in the
 paper:
 
-Ayan Chakrabarti, "**A Neural Approach to Blind Motion Deblurring**", 
-Proc. of the European Conference on Computer Vision (ECCV) 2016.
-[arXiv](http://arxiv.org/abs/1603.04771) 
+Ayan Chakrabarti, ["**A Neural Approach to Blind Motion Deblurring**"](http://arxiv.org/abs/1603.04771), 
+Proc. of the European Conference on Computer Vision (ECCV) 2016. 
 
 This implementation is being made available for non-commercial research
 use only. If you find this code useful in your research, please consider 
@@ -18,7 +17,7 @@ and contact <[email protected]> with any questions.
 ### Requirements
 
 1. You will need to download the trained neural model, available as
-   MAT file at the [project page](http://www.ttic.edu/chakrabarti/ndeblur/).
+   a MAT file at the [project page](http://www.ttic.edu/chakrabarti/ndeblur/).
 
 2. This implementation requires a modern CUDA-capable GPU (it has been
    tested on an NVIDIA Titan X), and a recent version of MATLAB's

training/README.md

@@ -70,6 +70,8 @@ deblurring code. Please read through the following steps, individual
 7. Once you're done training, create a model .mat file by saving the
    following variables:
 
-   `load([wname '.mat'],'net','best_wts');` 
-   `wts = best_wts;`
-   `save('-mat','model.mat','tfm','net','wts');`
+```MATLAB
+>>> load([wname '.mat'],'net','best_wts');
+>>> wts = best_wts;
+>>> save('-mat','model.mat','tfm','net','wts');
+```