fix: address lint and notebook structure (ruff, intro, imports, pinning)

Co-authored-by: Cursor <cursoragent@cursor.com>

Author: jhamon

learn/generation/openai/fine-tuning/gpt-3.5-agent-training/00-fine-tuning.ipynb

@@ -1,5 +1,16 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "intro"
+   },
+   "source": [
+    "# Fine-tuning GPT-3.5 for Agent Training\n",
+    "\n",
+    "This notebook demonstrates fine-tuning a GPT-3.5 model for use in an agent that uses a retrieval tool."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 1,
@@ -8,11 +19,22 @@
    },
    "outputs": [],
    "source": [
-    "!pip install -qU \\\n",
-    "    datasets==2.14.4 \\\n",
-    "    langchain==0.0.274 \\\n",
-    "    pinecone>=5.0.0 \\\n",
-    "    openai==0.27.9"
+    "!pip install -qU datasets==2.14.4 langchain==0.0.274 pinecone==5.0.0 openai==0.27.9\n",
+    "\n",
+    "import os\n",
+    "from getpass import getpass\n",
+    "from time import sleep\n",
+    "\n",
+    "import openai\n",
+    "import requests\n",
+    "from chains import VectorDBChain\n",
+    "from datasets import load_dataset\n",
+    "from langchain.agents import AgentType, Tool, initialize_agent\n",
+    "from langchain.chat_models import ChatOpenAI\n",
+    "from langchain.memory import ConversationBufferWindowMemory\n",
+    "\n",
+    "os.environ[\"OPENAI_API_KEY\"] = os.getenv(\"OPENAI_API_KEY\") or \"YOUR_API_KEY\"\n",
+    "openai.api_key = os.environ[\"OPENAI_API_KEY\"]"
    ]
   },
   {
@@ -41,8 +63,6 @@
     }
    ],
    "source": [
-    "from datasets import load_dataset\n",
-    "\n",
     "data = load_dataset(\"jamescalam/agent-conversations-retrieval-tool\", split=\"train\")\n",
     "data"
    ]
@@ -186,12 +206,6 @@
     }
    ],
    "source": [
-    "import os\n",
-    "import openai\n",
-    "\n",
-    "os.environ[\"OPENAI_API_KEY\"] = os.getenv(\"OPENAI_API_KEY\") or \"YOUR_API_KEY\"\n",
-    "openai.api_key = os.environ[\"OPENAI_API_KEY\"]\n",
-    "\n",
     "res = openai.File.create(file=open(\"conversations.jsonl\", \"r\"), purpose=\"fine-tune\")\n",
     "res"
    ]
@@ -460,11 +474,9 @@
     }
    ],
    "source": [
-    "from time import sleep\n",
-    "\n",
     "while True:\n",
     "    res = openai.FineTuningJob.retrieve(job_id)\n",
-    "    if res[\"finished_at\"] != None:\n",
+    "    if res[\"finished_at\"] is not None:\n",
     "        break\n",
     "    else:\n",
     "        print(\".\", end=\"\")\n",
@@ -540,8 +552,6 @@
    },
    "outputs": [],
    "source": [
-    "import requests\n",
-    "\n",
     "res = requests.get(\n",
     "    \"https://raw.githubusercontent.com/pinecone-io/examples/master/learn/generation/openai/fine-tuning/gpt-3.5-agent-training/chains.py\"\n",
     ")\n",
@@ -557,12 +567,6 @@
    },
    "outputs": [],
    "source": [
-    "from getpass import getpass\n",
-    "from langchain.agents import Tool\n",
-    "from langchain.chat_models import ChatOpenAI\n",
-    "from langchain.memory import ConversationBufferWindowMemory\n",
-    "from chains import VectorDBChain\n",
-    "\n",
     "llm = ChatOpenAI(temperature=0.5, model_name=ft_model)\n",
     "\n",
     "memory = ConversationBufferWindowMemory(\n",
@@ -591,8 +595,6 @@
    },
    "outputs": [],
    "source": [
-    "from langchain.agents import AgentType, initialize_agent\n",
-    "\n",
     "agent = initialize_agent(\n",
     "    agent=AgentType.CHAT_CONVERSATIONAL_REACT_DESCRIPTION,\n",
     "    tools=[vdb_tool],\n",

learn/generation/openai/fine-tuning/gpt-3.5-agent-training/chains.py

@@ -1,6 +1,7 @@
-from pinecone import Pinecone, ServerlessSpec
-import openai
 from uuid import uuid4
+
+import openai
+from pinecone import Pinecone, ServerlessSpec
 from tqdm.auto import tqdm