feat: Update GraphRAG notebook with Couchbase setup instructions

examples_notebooks/community_contrib/couchbase/GraphRAG_with_Couchbase.ipynb

@@ -7,7 +7,21 @@
    },
    "source": [
     "# Tutorial on GraphRAG Local Search with Couchbase Vector Store\n",
-    "This notebook walks through the process of setting up a search engine that combines Couchbase for storing embeddings, OpenAI's models for generating embeddings, and a local search engine for querying structured data. This is useful when you need to search through structured data using natural language queries, leveraging both machine learning and a database."
+    "\n",
+    "This notebook walks through the process of setting up a search engine that combines Couchbase for storing embeddings, OpenAI's models for generating embeddings, and a local search engine for querying structured data. \n",
+    "This is useful when you need to search through structured data using natural language queries, leveraging both machine learning and a database.This approach is useful for searching structured data using natural language queries, leveraging both machine learning and a database.\n",
+    "\n",
+    "## Setting up Couchbase\n",
+    "\n",
+    "Before running this notebook, set up the following in Couchbase:\n",
+    "\n",
+    "1. Create a bucket named \"graphrag-demo\" (or as specified in COUCHBASE_BUCKET_NAME)\n",
+    "2. Within the bucket, create a scope named \"shared\" (or as specified in COUCHBASE_SCOPE_NAME)\n",
+    "3. Within the scope, create a collection named \"entity_description_embeddings\" (or as specified in COUCHBASE_COLLECTION_NAME)\n",
+    "\n",
+    "These settings should match the environment variables defined in your .env file or the default values in the code.\n",
+    "\n",
+    "4. In the Couchbase Full Text Search (FTS) index section, create a new index by importing the `graphrag_demo_index.json` file. This file contains the necessary configuration for the vector search index.\n"
    ]
   },
   {