cleanup

Author: retroryan

Lab_7_Agent_Memory/01_memory_context_provider.ipynb

@@ -4,18 +4,7 @@
    "cell_type": "markdown",
    "id": "0",
    "metadata": {},
-   "source": [
-    "# Memory Context Provider\n",
-    "\n",
-    "This notebook uses the [`neo4j-agent-memory`](https://github.com/neo4j-labs/agent-memory) package — a graph-native memory system for AI agents, and the second Neo4j context provider package covered in this workshop. Unlike the `agent-framework-neo4j` provider in Lab 6 that retrieves from a static knowledge base, agent memory is dynamic: it grows with every conversation.\n",
-    "\n",
-    "Here you'll use its `Neo4jContextProvider` (distinct from the one in Lab 6) as a MAF context provider for automatic memory injection. The provider supports three memory types:\n",
-    "- **Short-term**: Recent conversation history with semantic search\n",
-    "- **Long-term**: Entities, preferences, and facts extracted from conversations\n",
-    "- **Reasoning**: Similar past task traces for learning from experience\n",
-    "\n",
-    "***"
-   ]
+   "source": "# Memory Context Provider\n\nThis notebook uses the [`neo4j-agent-memory`](https://github.com/neo4j-labs/agent-memory) package — a graph-native memory system for AI agents, and the second Neo4j context provider package covered in this workshop. Unlike the `agent-framework-neo4j` provider in Lab 6 that retrieves from a static knowledge base, agent memory is dynamic: it grows with every conversation.\n\nHere you'll use its `Neo4jContextProvider` (distinct from the one in Lab 6) as a MAF context provider for automatic memory injection. The provider supports three memory types:\n- **Short-term**: Recent conversation history with semantic search\n- **Long-term**: Entities, preferences, and facts extracted from conversations\n- **Reasoning**: Similar past task traces for learning from experience\n\n***"
   },
   {
    "cell_type": "markdown",
@@ -31,26 +20,7 @@
    "id": "2",
    "metadata": {},
    "outputs": [],
-   "source": [
-    "import sys\n",
-    "sys.path.insert(0, '../shared')\n",
-    "\n",
-    "import asyncio\n",
-    "\n",
-    "from pydantic import SecretStr\n",
-    "\n",
-    "from agent_framework.azure import AzureAIClient\n",
-    "from azure.identity.aio import AzureCliCredential\n",
-    "\n",
-    "from neo4j_agent_memory import MemoryClient, MemorySettings\n",
-    "from neo4j_agent_memory.integrations.microsoft_agent import (\n",
-    "    Neo4jContextProvider,\n",
-    "    Neo4jMicrosoftMemory,\n",
-    ")\n",
-    "\n",
-    "from azure_embedder import get_memory_embedder\n",
-    "from config import get_agent_config, Neo4jConfig"
-   ]
+   "source": "import sys\nsys.path.insert(0, '../shared')\n\nimport asyncio\n\nfrom pydantic import SecretStr\n\nfrom agent_framework.azure import AzureAIClient\nfrom azure.identity.aio import AzureCliCredential\n\nfrom neo4j_agent_memory import MemoryClient, MemorySettings\nfrom neo4j_agent_memory.integrations.microsoft_agent import (\n    Neo4jContextProvider,\n    Neo4jMicrosoftMemory,\n)\n\nfrom azure_embedder import get_memory_embedder\nfrom config import get_agent_config, Neo4jConfig"
   },
   {
    "cell_type": "markdown",
@@ -68,23 +38,7 @@
    "id": "4",
    "metadata": {},
    "outputs": [],
-   "source": [
-    "# Get workshop configuration\n",
-    "config = get_agent_config()\n",
-    "neo4j_config = Neo4jConfig()\n",
-    "\n",
-    "# Configure memory settings\n",
-    "settings = MemorySettings(\n",
-    "    neo4j={\n",
-    "        \"uri\": neo4j_config.uri,\n",
-    "        \"username\": neo4j_config.username,\n",
-    "        \"password\": SecretStr(neo4j_config.password),\n",
-    "    },\n",
-    ")\n",
-    "\n",
-    "print(f\"Neo4j URI: {neo4j_config.uri}\")\n",
-    "print(f\"Memory settings configured\")"
-   ]
+   "source": "# Get workshop configuration\nconfig = get_agent_config()\nneo4j_config = Neo4jConfig()\n\n# Configure memory settings\nsettings = MemorySettings(\n    neo4j={\n        \"uri\": neo4j_config.uri,\n        \"username\": neo4j_config.username,\n        \"password\": SecretStr(neo4j_config.password),\n    },\n    extraction={\"extractor_type\": \"spacy\"},\n)\n\nprint(f\"Neo4j URI: {neo4j_config.uri}\")\nprint(f\"Memory settings configured\")"
   },
   {
    "cell_type": "markdown",
@@ -105,26 +59,7 @@
    "id": "6",
    "metadata": {},
    "outputs": [],
-   "source": [
-    "# Create memory client\n",
-    "memory_client = MemoryClient(settings, embedder=get_memory_embedder())\n",
-    "await memory_client.__aenter__()\n",
-    "\n",
-    "# Create unified memory interface\n",
-    "session_id = \"workshop-demo\"\n",
-    "\n",
-    "memory = Neo4jMicrosoftMemory.from_memory_client(\n",
-    "    memory_client=memory_client,\n",
-    "    session_id=session_id,\n",
-    "    include_short_term=True,\n",
-    "    include_long_term=True,\n",
-    "    include_reasoning=True,\n",
-    "    extract_entities=True,\n",
-    ")\n",
-    "\n",
-    "print(f\"Session: {session_id}\")\n",
-    "print(f\"Memory created with context provider and chat store\")"
-   ]
+   "source": "# Create memory client\nmemory_client = MemoryClient(settings, embedder=get_memory_embedder())\nawait memory_client.__aenter__()\n\n# Create unified memory interface\nsession_id = \"workshop-demo\"\n\nmemory = Neo4jMicrosoftMemory.from_memory_client(\n    memory_client=memory_client,\n    session_id=session_id,\n    include_short_term=True,\n    include_long_term=True,\n    include_reasoning=True,\n    extract_entities=True,\n)\n\nprint(f\"Session: {session_id}\")\nprint(f\"Memory created with context provider and chat store\")"
   },
   {
    "cell_type": "markdown",
@@ -154,44 +89,7 @@
    "id": "9",
    "metadata": {},
    "outputs": [],
-   "source": [
-    "async def run_conversation():\n",
-    "    async with AzureCliCredential() as credential:\n",
-    "        async with AzureAIClient(\n",
-    "            project_endpoint=config.project_endpoint,\n",
-    "            model_deployment_name=config.model_name,\n",
-    "            credential=credential,\n",
-    "        ) as client:\n",
-    "            agent = client.as_agent(\n",
-    "                name=\"workshop-memory-agent\",\n",
-    "                instructions=(\n",
-    "                    \"You are a helpful assistant with persistent memory. \"\n",
-    "                    \"You can remember previous conversations and user preferences. \"\n",
-    "                    \"When you notice the user expressing a preference, acknowledge it.\"\n",
-    "                ),\n",
-    "                context_providers=[memory.context_provider],\n",
-    "            )\n",
-    "            session = agent.create_session()\n",
-    "\n",
-    "            # First interaction\n",
-    "            queries = [\n",
-    "                \"Hi! I'm interested in learning about Apple's products.\",\n",
-    "                \"What about their risk factors? I'm particularly concerned about supply chain issues.\",\n",
-    "                \"Can you remind me what we discussed about Apple?\",\n",
-    "            ]\n",
-    "\n",
-    "            for query in queries:\n",
-    "                print(f\"User: {query}\\n\")\n",
-    "                print(\"Assistant: \", end=\"\", flush=True)\n",
-    "\n",
-    "                response = await agent.run(query, session=session)\n",
-    "                print(response.text)\n",
-    "                print(\"\\n\" + \"-\" * 50 + \"\\n\")\n",
-    "\n",
-    "    await asyncio.sleep(0.1)\n",
-    "\n",
-    "await run_conversation()"
-   ]
+   "source": "async def run_conversation():\n    async with AzureCliCredential() as credential:\n        async with AzureAIClient(\n            project_endpoint=config.project_endpoint,\n            model_deployment_name=config.model_name,\n            credential=credential,\n        ) as client:\n            agent = client.as_agent(\n                name=\"workshop-memory-agent\",\n                instructions=(\n                    \"You are a helpful assistant with persistent memory. \"\n                    \"You can remember previous conversations and user preferences. \"\n                    \"When you notice the user expressing a preference, acknowledge it.\"\n                ),\n                context_providers=[memory.context_provider],\n            )\n            session = agent.create_session()\n\n            # First interaction\n            queries = [\n                \"Hi! I'm interested in learning about Apple's products.\",\n                \"What about their risk factors? I'm particularly concerned about supply chain issues.\",\n                \"Can you remind me what we discussed about Apple?\",\n            ]\n\n            for query in queries:\n                print(f\"User: {query}\\n\")\n                print(\"Assistant: \", end=\"\", flush=True)\n\n                response = await agent.run(query, session=session)\n                print(response.text)\n                print(\"\\n\" + \"-\" * 50 + \"\\n\")\n\n    await asyncio.sleep(0.1)\n\nawait run_conversation()"
   },
   {
    "cell_type": "markdown",
@@ -253,13 +151,7 @@
    "cell_type": "markdown",
    "id": "13",
    "metadata": {},
-   "source": [
-    "***\n",
-    "\n",
-    "[View the complete code](../financial_data_load/solution_srcs/07_01_memory_context_provider.py)\n",
-    "\n",
-    "[Move on to the Memory Tools Agent Notebook](02_memory_tools_agent.ipynb)"
-   ]
+   "source": "***\n\n[View the complete code](../financial_data_load/solution_srcs/07_01_memory_context_provider.py)\n\n[Move on to the Memory Tools Agent Notebook](02_memory_tools_agent.ipynb)"
   },
   {
    "cell_type": "code",
@@ -294,4 +186,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 5
-}
+}