social network analysis with pyg and dgl

.github/workflows/ci.yaml

@@ -31,11 +31,11 @@ jobs:
             --rm --detach -v "$(pwd)/data:/data" \
             --name graph-machine-learning-box \
             graph-machine-learning:latest
-
-          # Run tests
-          cd docker
-
+
           # Run the tests only for chapters managed by poetry
           CHAPTERS=$(find Ch* -name poetry.lock -print0 | sed -e 's/\/poetry.lock//g' | xargs -0)
-
+
+          # Run tests
+          cd docker
+
           ./tests.sh $CHAPTERS

Chapter03/04_Graph_Neural_Network.ipynb

@@ -217,7 +217,7 @@
     "id": "VHU1UGiHfw1e"
    },
    "source": [
-    "In this demo, we will be using the PROTEINS dataset, already integrated in StellarGraph"
+    "In this demo, we will be using the PROTEINS dataset, already integrated in StellarGraph (although we need to override the url, given that the original data was removed)"
    ]
   },
   {
@@ -246,6 +246,7 @@
     }
    ],
    "source": [
+    "sg.datasets.PROTEINS.url = 'https://www.chrsmrrs.com/graphkerneldatasets/PROTEINS.zip'\n",
     "dataset = sg.datasets.PROTEINS()\n",
     "display(HTML(dataset.description))\n",
     "graphs, graph_labels = dataset.load()"

Chapter06/01_Social_network_analysis.ipynb

@@ -11,7 +11,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 1,
    "metadata": {
     "executionInfo": {
      "elapsed": 2015,
@@ -90,8 +90,7 @@
    "source": [
     "!wget http://snap.stanford.edu/data/facebook_combined.txt.gz\n",
     "!wget http://snap.stanford.edu/data/facebook.tar.gz\n",
-    "!gzip -d facebook_combined.txt.gz\n",
-    "!tar -xf facebook.tar.gz"
+    "!gzip -d facebook_combined.txt.gz"
    ]
   },
   {
@@ -180,7 +179,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "metadata": {
     "executionInfo": {
      "elapsed": 923,
@@ -228,7 +227,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 3,
    "metadata": {
     "executionInfo": {
      "elapsed": 787,
@@ -296,7 +295,10 @@
    "outputs": [],
    "source": [
     "plt.axis(\"off\")\n",
-    "nx.draw_networkx(G, pos=spring_pos, node_color=default_node_color, edge_color=default_edge_color, with_labels=False, node_size=35)"
+    "plt.figure(1,figsize=(12,12)) \n",
+    "nx.draw_networkx(G, pos=spring_pos, node_color=default_node_color, edge_color=default_edge_color, with_labels=False, node_size=35)\n",
+    "ax = plt.gca() # to get the current axis\n",
+    "ax.collections[0].set_edgecolor(\"#fff\")"
    ]
   },
   {
@@ -341,7 +343,10 @@
     "                         pos=spring_pos, \n",
     "                         nodelist=max_keys, \n",
     "                         node_color=enhanced_edge_color,\n",
-    "                         node_size=max_vals)"
+    "                         node_size=max_vals)\n",
+    "  \n",
+    "  ax = plt.gca() # to get the current axis\n",
+    "  ax.collections[0].set_edgecolor(\"#fff\")"
    ]
   },
   {
@@ -675,14 +680,25 @@
     "  print(node, \"is in community number\", parts.get(node))\n",
     "  \n",
     "n_sizes = [5]*len(G.nodes())\n",
+    "\n",
+    "plt.axis(\"off\")\n",
+    "nx.draw_networkx(G, pos=spring_pos, cmap=plt.get_cmap(\"Blues\"), edge_color=default_edge_color, node_color=values, node_size=n_sizes, with_labels=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# enhance color and size of the ego-nodes\n",
     "for node in ego_nodes:\n",
     "  n_sizes[node] = 250\n",
     "\n",
     "plt.axis(\"off\")\n",
     "nx.draw_networkx(G, pos=spring_pos, cmap=plt.get_cmap(\"Blues\"), edge_color=default_edge_color, node_color=values, node_size=n_sizes, with_labels=False)\n",
     "\n",
-    "# enhance color and size of the ego-nodes\n",
-    "nodes = nx.draw_networkx_nodes(G,spring_pos,ego_nodes,node_color=[parts.get(node) for node in ego_nodes])\n",
+    "nodes = nx.draw_networkx_nodes(G, spring_pos, ego_nodes,node_color=[parts.get(node) for node in ego_nodes])\n",
     "nodes.set_edgecolor(enhanced_node_color)"
    ]
   },
@@ -933,7 +949,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "metadata": {
     "id": "baistC-ZdaRf"
    },
@@ -1046,7 +1062,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 5,
    "metadata": {
     "id": "2C3TwPVTdgeM"
    },
@@ -1059,7 +1075,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 6,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -1078,12 +1094,36 @@
     "id": "oZIj_NqvVzmU",
     "outputId": "990227b3-7f01-4ef6-d036-9346c7ae5eca"
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'features': array([1., 1., 1., ..., 0., 0., 0.])}"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
     "# check features has been correctly assigned\n",
     "G.nodes[0]"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# dump the graph on file\n",
+    "import pickle\n",
+    "\n",
+    "with open('test.gpickle', 'wb') as f:\n",
+    "    pickle.dump(G, f, pickle.HIGHEST_PROTOCOL)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {
@@ -1096,35 +1136,6 @@
     "More in detail, TODO"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 1000
-    },
-    "executionInfo": {
-     "elapsed": 14891,
-     "status": "ok",
-     "timestamp": 1616871410804,
-     "user": {
-      "displayName": "Aldo Marzullo",
-      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjBD_mZewcZ8LCqkD20Nku4DR5OCGFqYkxawoUjgg=s64",
-      "userId": "17245895923239449231"
-     },
-     "user_tz": -60
-    },
-    "id": "Li-IypwmIW9V",
-    "outputId": "580084bf-ff08-406d-b335-7f791b93e9fe"
-   },
-   "outputs": [],
-   "source": [
-    "!pip install stellargraph\n",
-    "!pip install node2vec==0.3.3\n",
-    "!pip install git+https://github.com/palash1992/GEM.git"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -1710,9 +1721,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "metadata": {
-    "id": "O1YkgCx5rh25"
-   },
+   "metadata": {},
    "outputs": [],
    "source": []
   }
@@ -1726,7 +1735,7 @@
    "provenance": []
   },
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -1740,7 +1749,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.5.0"
+   "version": "3.9.12"
   }
  },
  "nbformat": 4,