Adds notebooks for Telco-Network application with GNN and LLM.

ChangeLog.md

@@ -4,6 +4,9 @@ Starting with v1.31.6, this file will contain a record of major features and upd
 
 ## Upcoming
 
+- New notebooks showing Telco examples leveraging GNN and LLM ([Link to PR](https://github.com/aws/graph-notebook/pull/587))
+  - Path: 02-Neptune-ML > 03-Sample-Applications > 04-Telco-Networks
+
 ## Release 4.2.0 (April 4, 2024)
 
 - New Neptune Analytics notebooks - SBOM ([Link to PR](https://github.com/aws/graph-notebook/pull/566))

src/graph_notebook/notebooks/03-Neptune-ML/03-Sample-Applications/04-Telco-Networks/1b-Graph_init.ipynb

@@ -0,0 +1,64 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "21451344-eff7-4c5b-b75d-046fbde1c9a2",
+   "metadata": {},
+   "source": [
+    "# Loading the Graph data\n",
+    "\n",
+    "The data that we use is adapted from the [Stanford Network Analysis Project (SNAP) public Telecom dataset](https://snap.stanford.edu/data/#tcn)\n",
+    "\n",
+    "The relationships between users and cells in dataset were extracted and was augmented to add gnodeb nodes and edges between cells and gnodebs to make it similar to a 5G cell network.\n",
+    "\n",
+    "The data transformation implementation with AWS Glue can be found in the script `Transform2Neptune.py`\n",
+    "\n",
+    "The final dataset used in this application contains Users, Cells, gnodeb, etc. and the known links in between\n",
+    "\n",
+    "- s3://neptuneml-data/telcograph/data/node_users.csv\n",
+    "\n",
+    "- s3://neptuneml-data/telcograph/node_cell.csv\n",
+    "\n",
+    "- s3://neptuneml-data/telcograph/gnodeb_node.csv\n",
+    "\n",
+    "- s3://neptuneml-data/telcograph/edge_all.csv\n",
+    "\n",
+    "- s3://neptuneml-data/telcograph/edge_gnodeb.csv\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "960dfe5d-67ce-43fd-85a4-6b1593790581",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "# #load command for loading the graph from S3 to Neptunedb\n",
+    "%load -s s3://neptuneml-data/telco-graph/data -f csv -p OVERSUBSCRIBE"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

src/graph_notebook/notebooks/03-Neptune-ML/03-Sample-Applications/04-Telco-Networks/2a-GraphQueryGremlin.ipynb

@@ -0,0 +1,216 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "source": [
+    "# Querying the Graph data\n",
+    "\n",
+    "Once the data has been loaded into the database, we can run some gremlin queries to query the data in the graph database.\n",
+    "\n",
+    "Below we have examples of a few Gremlin queries to get the count of vertices and edges in the graph as well as to get the counts by vertex label and edge label."
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "cd27cae0-9aad-47cc-a1fa-46ea561d4e1e",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.V().count().next()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7f434bc3-ddca-4979-a580-f1f416a1443a",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.E().count().next()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8f22769f-1f97-4e03-b3a3-aa62e44f4b42",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.V().groupCount().by(label).unfold().order().by(keys)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ae383337-74bb-4105-8c4a-6acf93edde42",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.E().groupCount().by(label).unfold().order().by(keys)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "4240519a-2dba-401c-81f2-4860848672f0",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.V().values('user_attr_1').mean().next()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "13b6adf4-5d33-4957-8e32-db897a0d4a6e",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.V().values('user_attr_1').min().next()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ededeb32-e1f5-4b29-bf03-e7dc557ad960",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "# Find the distinct values of a property in vertices\n",
+    "distinct_values = g.V().values('user_attr_1').dedup().toList()\n",
+    "#print(\"Distinct values:\", distinct_values)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "76a5b6da-e855-49d2-b120-374dd7a8ea2e",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.V().hasLabel('user').properties().count()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "f99bb270-0d9c-4b44-bef2-332306bab4fb",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.V().hasLabel('user').properties().key().dedup().count()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c1fd63ad-f0ad-4f38-9e5c-4832d2d7fc51",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.V().hasLabel('cell').properties().key().dedup().count() "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "4b96732c-f6fa-4c60-a0a9-a15610885b6b",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.V().hasLabel('cell').properties().count()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "f23d18bb-1ee8-4374-8197-1d0b8dc1e62b",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.E().hasLabel('user_live_cell').properties().key().dedup().count() "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "13594594-0ff0-42d5-a5b2-205c8a39860c",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "%%gremlin\n",
+    "g.E().hasLabel('user_live_cell').properties().count()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8a705292-849d-4a55-98ca-f75dd7b3892b",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}