Initial Release.

PiperOrigin-RevId: 264337446

graph_matching_networks/README.md

@@ -0,0 +1,23 @@
+# Graph Matching Networks for Learning the Similarity of Graph Structured Objects
+
+This is the example code for the following ICML 2019 paper.  If you use the code
+here please cite this paper.
+
+> Yujia Li, Chenjie Gu, Thomas Dullien, Oriol Vinyals, Pushmeet Kohli.  *Graph Matching Networks for Learning the Similarity of Graph Structured Objects*.  ICML 2019.  [\[arXiv\]](https://arxiv.org/abs/1904.12787).
+
+
+## Running the code
+
+The code is in the format of a colab notebook, which includes:
+
+*  an example implementation of the model,
+*  an example graph similarity learning task,
+*  an example training loop, and
+*  some attention visualization tools.
+
+To launch the notebook in Google colab, [click here](https://colab.research.google.com/github/deepmind/deepmind_research/blob/master/graph_matching_networks/graph_matching_networks.ipynb).
+
+You can also download the notebook and run it locally with jupyter.  The
+notebook assumes you are on python 3 and have the latest (as of July 24,
+2019) tensorflow, sonnet, numpy etc. installed.  You can install the dependencies by running `pip3 install --user -r requirements.txt`.
+

graph_matching_networks/requirements.txt

@@ -0,0 +1,6 @@
+matplotlib==3.1.1
+networkx==2.3
+dm-sonnet==1.34
+numpy==1.16.4
+tensorflow==1.14
+six==1.12