Cleaned up structure

benjaminpatrickevans · Nov 8, 2017 · f5c787f · f5c787f
1 parent c5e4cf5
commit f5c787f
Show file tree

Hide file tree

Showing 4 changed files with 1 addition and 1 deletion.
diff --git a/ConvGP/.DS_Store → .DS_Store b/ConvGP/.DS_Store → .DS_Store
diff --git a/README.md b/README.md
@@ -6,7 +6,7 @@ The key idea is to combine aspects from genetic programming and convolutional ne
 to overcome various limitations of ConvNets, i.e.
 
 - Need for manually crafted architectures
-- Poor Interpretability (Although google brain appears to be doing some promising research in this area [here]("https://distill.pub/2017/feature-visualization/"), however a large limitation is still interpretability of feature interaction). 
+- Poor Interpretability. Although google brain appears to be doing some promising research in this area [here](https://distill.pub/2017/feature-visualization/), however a large limitation is still interpretability of feature interaction 
 - Require large amounts of training data
 
 Future work will look at propagting gradients back through the tree, as seen with Convoultional neural networks

diff --git a/ConvGP/Report.pdf → Report.pdf b/ConvGP/Report.pdf → Report.pdf
diff --git a/ConvGP/Slides.pdf → Slides.pdf b/ConvGP/Slides.pdf → Slides.pdf