Improve edge and node update images (#116)

Use edge update and node update images which demonstrate more of the process of the learner Swaps out two images for the corresponding diagrams given in the powerpoint presentation.
typedb · Dec 16, 2019 · 80ce7a5 · 80ce7a5
1 parent fd8acae
commit 80ce7a5
Show file tree

Hide file tree

Showing 3 changed files with 2 additions and 2 deletions.
diff --git a/kglib/kgcn/.images/edge_update.png b/kglib/kgcn/.images/edge_update.png
diff --git a/kglib/kgcn/.images/node_update.png b/kglib/kgcn/.images/node_update.png
diff --git a/kglib/kgcn/README.md b/kglib/kgcn/README.md
@@ -68,13 +68,13 @@ for step in supersteps:
 
 Use as input the current edge features and the features of the nodes it connects. Update the edge's features as the output of some neural network layers. Do this for all edges.
 
-<img src = ".images/edge_update.png" style="width:50%">
+![Edge Update](.images/edge_update.png)
 
 ### Node Block
 
 Use as input the node's features and the most up-to-date features of the edges that connect to it. Update the node's features as the output of some neural network layers. Do this for all nodes.
 
-<img src = ".images/node_update.png" style="width:50%">
+![Node Update](.images/node_update.png)
 
 ### How is this Convolutional?