guide/14-deep-learning/address_standerdization_using_model_extension.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0b1e1740",
   "metadata": {},
   "source": [
    "# Address Standardization Using Model Extension: A Modern Approach"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dbe40b1-24a2-43d9-9e18-77affba9d661",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Table of Contents"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "477cc1a5",
   "metadata": {
    "tags": []
   },
   "source": [
    "* [Introduction](#introduction)\n",
    "* [Imports](#imports)\n",
    "* [Prompt Engineering](#prompt-engineering)\n",
    "* [Esri Model Definition(EMD) file](#esri-model-definition-emd-file)\n",
    "* [Anatomy of the Extension Function](#anatomy-of-the-extension-function)\n",
    "  * [Define the ```__init__``` function](#define-the-__init__-function)\n",
    "  * [Define the ```getParameterInfo``` function](#define-the-getconfiguration-function)\n",
    "  * [Define the ```initialize``` function](#define-the-initialize-function)\n",
    "  * [Define the ```getConfiguration``` function](#define-the-getparameterinfo-function)\n",
    "  * [Define the ```predict``` function](#define-the-predict-function)\n",
    "* [Complete NLP Function](#complete-nlp-function)\n",
    "* [Create Custom ESRI Deep Learning Package (.dlpk) file](#create-custom-esri-deep-learning-package-dlpk-file)\n",
    "* [Using the custom ESRI Deep Learning Package (.dlpk) file in arcgis.learn API](#using-the-custom-esri-deep-learning-package-dlpk-file-in-arcgislearn-api)\n",
    "* [Tool Interface](#tool-interface)\n",
    "* [Sample Input Table](#sample-input-table)\n",
    "* [Sample Output Table](#sample-output-table)\n",
    "* [Conclusion](#conclusion)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8a7b212-1558-4255-9b09-f551e7a09c65",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ed120d7-ac9b-4d75-bcbc-df010b03e80a",
   "metadata": {},
   "source": [
    "The **GeoAI toolbox** features a comprehensive **Text Analysis toolset** designed for diverse text processing tasks, including _text classification, entity extraction, and text transformation_. Leveraging advanced natural language processing (NLP) models built on powerful backbones like BERT, RoBERTa, T5, and large language models (LLMs) such as Mistral, these tools deliver high-performance text analysis.\n",
    "\n",
    "But what if you encounter a **Text model**, like a **BERT** or **T5** backbone-based language model, or a **large language model (LLM)** such as **LLaMA**, **Claude**, or **Mistral**, that isn’t included in the current learning module? Perhaps you want to utilize it from a library or an open-source code repository. Or maybe you’ve fine-tuned your own **text model** for specific tasks, such as standardizing addresses, conducting sentiment analysis, or extracting custom entities or relationships not currently supported by the Text Analysis toolset. How can you implement this within **ArcGIS Pro**?\n",
    "\n",
    "To address these needs, ArcGIS allows for the integration of external third-party language models, including both open-source LLMs and commercial LLMs accessed via web APIs. Please note that if you use a web-hosted LLM, the data processed will be sent to the LLM provider. Python developers can create custom NLP functions to interface with these external models, packaging their models as Esri deep learning packages (ESRI Deep Learning Package (.dlpk) files) for use with the following tools:\n",
    "\n",
    "1. [Classify Text Using Deep Learning (GeoAI)](https://pro.arcgis.com/en/pro-app/latest/tool-reference/geoai/classify-text-using-deep-learning.htm)\n",
    "2. [Extract Entities Using Deep Learning (GeoAI)](https://pro.arcgis.com/en/pro-app/latest/tool-reference/geoai/extract-entities-using-deep-learning.htm)\n",
    "3. [Transform Text Using Deep Learning (GeoAI)](https://pro.arcgis.com/en/pro-app/latest/tool-reference/geoai/transform-text-using-deep-learning.htm)\n",
    "\n",
    "The ESRI Deep Learning Package (.dlpk) file of the custom NLP function is also supported through the `arcgis.learn` Python API, which accommodates third-party custom models via various classes, including `TextClassifier`, `EntityRecognizer`, and `SequenceToSequence`.\n",
    "\n",
    "This document showcases a use case where a hosted commercial LLM was employed for an address standardization task. It provides a step-by-step guide detailing the components of the **NLP Function** and outlines the process for creating a ESRI Deep Learning Package (.dlpk) file using this function. Additionally, we will demonstrate how to accomplish the same task using our **Python API** interface."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47176560-f8da-4ce6-8243-14e15cd79305",
   "metadata": {},
   "source": [
    " **Address Standardization** is the process of formatting addresses to ensure consistency and accuracy. It involves correcting spelling errors, unifying abbreviations, and conforming to postal regulations.  With the rise of language models like GPT-3.5, organizations can now automate and enhance the process of address standardization. These models can quickly adapt to various address formats and nuances, making them well-suited for transforming raw addresses into standardized formats efficiently and accurately. \n",
    "\n",
    " This work aims to demonstrate the functionality of the **Model Extension** to transform text using GPT-3.5. We will explore a use case where addresses are supplied in non-standard formats, and the goal is to generate the addresses in standardized forms.\n",
    "\n",
    "#### Intended Audience\n",
    "- Analysts\n",
    "- Extension model developers\n",
    "\n",
    "#### Dataset\n",
    "- Address Dataset: A custom dataset that contains a collection of examples with **non-standardized addresses**. \n",
    "\n",
    "#### Backbone for Analysis\n",
    "- GPT-3.5\n",
    "\n",
    "#### Capability Demonstration\n",
    "- Prompt Engineering\n",
    "- Out-of-the-box use of LLM\n",
    "- Minimal or no annotation required\n",
    "- Comparable accuracy\n",
    "- Integration of custom models within the ArcGIS Pro **Transform Text Using Deep Learning (GeoAI)** and **arcgis.learn API**\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b33c40bc-1ee1-497c-8789-2d5aa6eb8b8c",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a5e30912-b73c-484e-9b56-19aa4841d56d",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import json\n",
    "import requests\n",
    "from arcgis.features import FeatureSet"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1194b485-3a2c-45a1-a095-f43e813bbe4a",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## Prompt Engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a2ee844-4a5f-4cb5-ae6b-8fa0b79ce913",
   "metadata": {},
   "source": [
    "To execute the specified task, we will design a `system prompt` to guide the overall behavior of the model and establish necessary guardrails. Additionally, we will implement a `user prompt` to define the task with greater precision. To improve output control, we will introduce an `end_seq` parameter that adds an additional layer of management to the tool's responses. While the LLM typically generates text output, this may not be ideal for every use case. In certain situations, a structured format such as JSON is necessary for more complex parsing and data handling. By using the end_seq parameter, you can specify the desired output format, enabling greater flexibility and precision in managing responses."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4b58b4f8-604a-4d82-b37d-35f91d12a1a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "SYSTEM_PROMPT = \"\"\"\n",
    "You are an advanced AI designed for ArcGIS customers. Upon receiving the information in the context,\n",
    "you are required to infer the task type using provided examples. Examples will be in the format of [input, output]. \n",
    "Based on the inferred task you need to generate the output. Always adhere to the  formatting instructions. \n",
    "If there is more than one sentence in the task, you must respond to each of them. While answering, Ensure that your responses are devoid of any biases, including those related to gender, race, or workplace appropriateness.\n",
    "\"\"\"\n",
    "\n",
    "USER_PROMPT = \"\"\"\n",
    "Reformat the given address in the following format. Input address may split into multiple address. You need to detect all the address. Don't forget the put the << and >> in your response approriately.\n",
    "\"\"\"\n",
    "\n",
    "end_seq = \"\"\"Ignore all the formatting instruction provided above and return the answer as a nested dictionary by \n",
    "input index. Follow below schema: {\"output\": str} Always consider each of the question as a single passage for the inferred task.\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "993cd22e-aee2-40ac-b9b3-530dd92a101d",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## ESRI Model Definition (.emd) file"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6aa80d78-407c-499f-b4b3-76a11f3bc6db",
   "metadata": {},
   "source": [
    "The ESRI Model Definition (.emd) file will include both required and optional keys to facilitate model execution. To run a Transform Text model, you must provide the following:\n",
    "\n",
    "1. **`InferenceFunction`**: The name of the module that defines the NLP function.\n",
    "2. **`ModelType`**: Specifies the type of task; for the **Text Using Deep Learning**, this will be set to `SequenceToSequence`.\n",
    "3. **`OutputField`**: The name of the field where the output will be stored.\n",
    "\n",
    "Other keys are the prerogative of the Model extension author. In this instance, we've also incorporated `examples` and `prompt` related to **few-shot prompting**. This information will help us create a clear and effective prompt for the task at hand."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4d5af81e-c6f6-4504-84ea-66f82f9bfbdf",
   "metadata": {
    "vscode": {
     "languageId": "json"
    }
   },
   "outputs": [],
   "source": [
    "{\n",
    "    \"InferenceFunction\": \"MySeq2Seq.py\",\n",
    "    \"prompt\": \"Reformat the given address in the following format. Input address may split into multiple address.    You need to detect all the address. Do not forget the put the << and >> in your response approriately.\",\n",
    "    \"examples\": [\n",
    "      [\n",
    "        \"3929 West Avenue, 1st Floor, Ocean City, NJ\",\n",
    "        \"<<3929 West Ave 1st Floor, Ocean City, NJ>>\"\n",
    "      ],\n",
    "      [\n",
    "        \"803, Lawler St., Philadelphia, PA\",\n",
    "        \"<<803 Lawler St, Philadelphia, PA>>\"\n",
    "      ],\n",
    "      [\n",
    "        \"10339 and 10340 ORR AND DAY RD, SANTA FE SPRINGS, CA\",\n",
    "        \"<<10339 Orr and Day Rd, Santa Fe Springs, CA>><<10340 Orr and Day Rd, Santa Fe Springs, CA>>\"\n",
    "      ]\n",
    "    ],\n",
    "    \"ModelType\": \"SequenceToSequence\",\n",
    "    \"ModelName\": \"SequenceToSequence\",\n",
    "    \"OutputField\": \"output\"\n",
    "  }"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "89cfcfb7-8c15-404f-9336-9057c72ee5e1",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## Anatomy of the Extension Function "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ccc1cbc9-d4dd-46bd-b6ad-578a0aa6d797",
   "metadata": {},
   "source": [
    "The model extension requires the process be wrapped in a class that implements the following functions:\n",
    "\n",
    "- `__init__`\n",
    "\n",
    "- `getParameterInfo`\n",
    "\n",
    "- `initialize`\n",
    "\n",
    "- `getConfiguration`\n",
    "\n",
    "- `predict`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0db98714-0c8a-4e22-a806-6ba8aa553a40",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Define the ```__init__``` function "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fed6be0-1cf4-42fc-895b-60d2231f97d4",
   "metadata": {},
   "source": [
    "The `__init__` method initializes instance variables such as `name`, `description`, and other attributes that are essential for the NLP function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "837160b7-2a7d-4e3f-88f5-56a5508ca44c",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def __init__(self, **kwargs):\n",
    "    self.description = \"Address standardization model\"\n",
    "    self.name = \"Sequence to sequence for address standardization\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34845994-8622-4831-b44d-e36276030a5d",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Define the ```getParameterInfo``` function "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9aa02e1-fc97-428e-908e-c271b33c5670",
   "metadata": {},
   "source": [
    "This function is designed to collect parameters from the user through the **Transform Text Using Deep Learning (GeoAI)** Tool. To connect to the GPT-3.5 deployment on Azure, we require the `API key`, `base URL`, and `deployment name` parameters. Since these parameters are sensitive, we will prompt the user for this information instead of hardcoding it within the file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "999eb0cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def getParameterInfo(self, **kwargs):\n",
    "    t =  [\n",
    "    {\n",
    "        \"name\": \"api_key\",\n",
    "        \"dataType\": \"string\",\n",
    "        \"required\": True,\n",
    "        \"displayName\": \"Api key of the Microsoft Azure deployment\",\n",
    "        \"description\": \"Api key of the Microsoft Azure deployment\",\n",
    "        \"value\": ''\n",
    "\n",
    "    },\n",
    "    {\n",
    "        \"name\": \"api_base\",\n",
    "        \"dataType\": \"string\",\n",
    "        \"required\": True,\n",
    "        \"displayName\": \"api base\",\n",
    "        \"description\": \"Base URL of the Azure deployment\",\n",
    "    },\n",
    "    {\n",
    "        \"name\": \"engine\",\n",
    "        \"dataType\": \"string\",\n",
    "        \"required\": False,\n",
    "        \"displayName\": \"deployment name\",\n",
    "        \"description\": \"Name of the deployment\",\n",
    "    },\n",
    "    {\n",
    "        \"name\": \"prompt\",\n",
    "        \"dataType\": \"string\",\n",
    "        \"required\": False,\n",
    "        \"displayName\": \"custom user prompt\",\n",
    "        \"description\": \"custom user prompt\",\n",
    "    }\n",
    "    ]\n",
    "\n",
    "\n",
    "    return t"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad320582-1ed2-414b-b567-128f3ac332dc",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Define the ```initialize``` function"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7b7f80e-6eb3-48e3-bf37-7079460cb73c",
   "metadata": {},
   "source": [
    "The `initialize` function acts as the central hub for the extension. Within this function, we will set up the necessary variables. It accepts two parameters via `kwargs`:\n",
    "\n",
    "#### Parameters in `kwargs`\n",
    "- **`model`**: The path to the EMD file.\n",
    "- **`device`**: The name of the device (either GPU or CPU), which is particularly important for on-premises models.\n",
    "\n",
    "Unlike the `__init__` method, which creates an instance of the NLP function, `initialize` reads the EMD file and configures the essential variables needed for inference."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bd943109",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def initialize(self, **kwargs):\n",
    "    with open(kwargs[\"model\"]) as f:\n",
    "        temp_payload = json.load(f)\n",
    "    kwargs.update(temp_payload)\n",
    "    # This block of the code will parse params from emd or kwargs passed from arcgis.learn python API\n",
    "    self.API_BASE = kwargs.get(\"api_base\", \"\")\n",
    "    self.API_KEY = kwargs.get(\"api_key\", \"\")\n",
    "    self.engine = kwargs.get(\"engine\", \"\")\n",
    "    self.API_VERSION = \"2024-02-01\"\n",
    "    self.examples = kwargs.get(\"examples\", [])\n",
    "    self.header = None\n",
    "    self.model = \"gpt-35-turbo-16k\"\n",
    "    self.prompt = kwargs.get(\"prompt\", \"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d121c9d-e10e-4bbb-96db-402c34128e10",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Define the ```getConfiguration``` function "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85296efe-2889-466f-a985-82c9353d0065",
   "metadata": {},
   "source": [
    "This function receives `kwargs` that contain parameter names and values collected from the user.. Additionally, it includes parameters like `batch_size`, which controls the tool's behavior. The Text Transform Tool supports customization of `batch_size`, determining how many records are processed in a single batch. This parameter plays a crucial role in optimizing the tool's performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6687993f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def getConfiguration(self, **kwargs):\n",
    "    # Edit the batch size\n",
    "    self.API_BASE = kwargs.get(\"api_base\", \"\")\n",
    "    self.API_KEY = kwargs.get(\"api_key\", \"\")\n",
    "    self.engine = kwargs.get(\"engine\", \"\") \n",
    "    self.prompt = kwargs.get(\"prompt\", \"\") \n",
    "    if kwargs[\"batch_size\"] > 4:\n",
    "        kwargs[\"batch_size\"] = 4\n",
    "    return kwargs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93e54e4f-4460-4215-8170-fabf551bc9a2",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Define the ```predict``` function "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a5196c4-e58f-42df-aa0e-1a00b8370d45",
   "metadata": {},
   "source": [
    "This function serves as the entry point for performing inference on the input. It takes a `FeatureSet` as input along with the field name that contains the relevant data. First, we will retrieve the field name from the provided `kwargs` and use it to extract the list of sentences for inference.\n",
    "\n",
    "In the next step, we will create the payload for GPT-3.5, preparing the input in the specified format of the `system message` and `user message` sequence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "07728e8b-bc5c-4892-8c88-db2de7b01358",
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict(self, feature_set: FeatureSet, **kwargs):\n",
    "        field = kwargs[\"input_field\"] #Fetch the field name which will have input\n",
    "        text_list = feature_set.df[field].to_list() # convert input to text\n",
    "        #set up the security keys and url \n",
    "        self.header = {\"Content-Type\": \"application/json\", \"api-key\": self.API_KEY}\n",
    "        API_BASE = f\"https://{self.API_BASE}/openai/deployments/{self.engine}/chat/completions?api-version={self.API_VERSION}\"\n",
    "\n",
    "        # Format the examples for few-shot prompt\n",
    "        if self.examples is not None:\n",
    "            example_formatted = [\n",
    "                f\"### Example {idx} \\n\\n {i}\\n\\n\" for idx, i in enumerate(self.examples)\n",
    "            ]\n",
    "        else:\n",
    "            example_formatted = []\n",
    "\n",
    "        if len(example_formatted):\n",
    "            t = \"\\n\".join(\n",
    "                example_formatted\n",
    "            )  # have to do it outside due to limitation of f-string\n",
    "            prompt = f\"{self.prompt}\\n\\nBelow are the representative examples\\n\\n{t}\"\n",
    "        else:\n",
    "            prompt = f\"{self.prompt}\\n\\n {','.join(example_formatted)}\"\n",
    "\n",
    "        prompt = f\"{SYSTEM_PROMPT}\\n\\n{prompt}\\n\\n{end_seq}\"\n",
    "\n",
    "        payload = []\n",
    "        features_list = []\n",
    "        # Create the payload in the system and user prompt for GPT\n",
    "        for message in text_list:\n",
    "            payload = (\n",
    "                {\n",
    "                    \"messages\": [\n",
    "                        {\"role\": \"system\", \"content\": prompt},\n",
    "                        {\"role\": \"user\", \"content\": f\"{message}\"},\n",
    "                    ],\n",
    "                    \"temperature\": 0.2,\n",
    "                }\n",
    "            )\n",
    "\n",
    "            payload = json.dumps(payload)\n",
    "            # Make the call to GPT\n",
    "            s = requests.request(method=\"POST\", url=API_BASE, headers=self.header, data=payload, verify=True) \n",
    "            if s.status_code == 200:\n",
    "                output = s.json()[\"choices\"][0][\"message\"][\"content\"]\n",
    "            else:\n",
    "                output = \"\"\n",
    "\n",
    "            if len(output):\n",
    "                try:\n",
    "                    _, val = list(json.loads(output).items())[0] # Load the output from JSON as dictionary\n",
    "                    sub_val = [i.replace(\"\\n\", \"\") for i in [i.replace(\"<<\", \"\") for i in val.split(\">>\")] if len(i)][0] # parse the output which are enclosed in << and >>\n",
    "                    features_list.append({'attributes': {'input_raw': message, \"output\": sub_val }})\n",
    "                except: \n",
    "                    features_list.append({'attributes': {'input_raw': message, \"output\": \"NA\" }})\n",
    "            else:\n",
    "                features_list.append({'attributes': {'input_raw': message, \"output\": \"NA\" }})\n",
    "        # Create the feature set. here we will define the fields and their data type. All the feature list collected \n",
    "        # during LLM call will added into the feature set. \n",
    "        feature_dict = {\n",
    "            \"fields\": [\n",
    "                { \"name\": \"input_raw\", \"type\": \"esriFieldTypeString\" },  \n",
    "                { \"name\": \"output\", \"type\": \"esriFieldTypeString\" }, \n",
    "            ],\n",
    "            'geometryType': \"\",\n",
    "            'features':features_list\n",
    "        }\n",
    "\n",
    "        return FeatureSet.from_dict(feature_dict)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1493d621-a9ef-4767-a730-76b0229a130d",
   "metadata": {},
   "source": [
    "Finally, the tool will receive a `FeatureSet` as output from the **extensible function**."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a6c209b-413a-454a-bcce-1167dc11600a",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## Complete NLP Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7f25510-4ec8-473d-ad36-5a4b6d9d33e0",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import requests\n",
    "from arcgis.features import FeatureSet\n",
    "\n",
    "SYSTEM_PROMPT = \"\"\"You are an advanced AI designed for ArcGIS customers. Upon receiving the information in the context,\n",
    " you are required to infer the task type using provided examples. Examples will be in the format of [input, output]. \n",
    " Based on the inferred task you need to generate the output. Always adhere to the  formatting instructions. \n",
    " If there is more than one sentence in the task, you must respond to each of them. While answering, ensure you are \n",
    " devoid of any biases, such as gender, racial, and not suitable for the workplace. \n",
    " \"\"\"\n",
    "\n",
    "end_seq =  \"\"\"Ignore all the formatting instruction provided above and return the answer as a nested dictionary by \n",
    "input index. Follow below schema: {\"output\": str} Always consider each of the question as a single passage for the inferred task.\"\"\"\n",
    "\n",
    "class MySeq2Seq:\n",
    "    def __init__(self, **kwargs):\n",
    "        self.description = \"Address standardization model\"\n",
    "        self.name = \"Sequence to sequence for address standardization\"\n",
    "\n",
    "    def initialize(self, **kwargs):\n",
    "        with open(kwargs[\"model\"]) as f:\n",
    "            temp_payload = json.load(f)\n",
    "        kwargs.update(temp_payload)\n",
    "        # This block of the code will parse params from emd or kwargs passed from arcgis.learn python API\n",
    "        self.API_BASE = kwargs.get(\"api_base\", \"\")\n",
    "        self.API_KEY = kwargs.get(\"api_key\", \"\")\n",
    "        self.engine = kwargs.get(\"engine\", \"\")\n",
    "        self.API_VERSION = \"2024-02-01\"\n",
    "        self.examples = kwargs.get(\"examples\", [])\n",
    "        self.header = None\n",
    "        self.model = \"gpt-35-turbo-16k\"\n",
    "        self.prompt = kwargs.get(\"prompt\", \"\")\n",
    "\n",
    "    def getParameterInfo(self, **kwargs):\n",
    "        t =  [\n",
    "        {\n",
    "            \"name\": \"api_key\",\n",
    "            \"dataType\": \"string\",\n",
    "            \"required\": True,\n",
    "            \"displayName\": \"Api key of the Microsoft Azure deployment\",\n",
    "            \"description\": \"Api key of the Microsoft Azure deployment\",\n",
    "            \"value\": ''\n",
    "\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"api_base\",\n",
    "            \"dataType\": \"string\",\n",
    "            \"required\": True,\n",
    "            \"displayName\": \"api base\",\n",
    "            \"description\": \"Base URL of the Azure deployment\",\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"engine\",\n",
    "            \"dataType\": \"string\",\n",
    "            \"required\": False,\n",
    "            \"displayName\": \"deployment name\",\n",
    "            \"description\": \"Name of the deployment\",\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"prompt\",\n",
    "            \"dataType\": \"string\",\n",
    "            \"required\": False,\n",
    "            \"displayName\": \"custom user prompt\",\n",
    "            \"description\": \"custom user prompt\",\n",
    "        }\n",
    "        ]\n",
    "\n",
    "\n",
    "        return t\n",
    "\n",
    "    def getConfiguration(self, **kwargs):\n",
    "        # Edit the batch size\n",
    "        self.API_BASE = kwargs.get(\"api_base\", \"\")\n",
    "        self.API_KEY = kwargs.get(\"api_key\", \"\")\n",
    "        self.engine = kwargs.get(\"engine\", \"\") \n",
    "        self.prompt = kwargs.get(\"prompt\", \"\") \n",
    "        if kwargs[\"batch_size\"] > 4:\n",
    "            kwargs[\"batch_size\"] = 4\n",
    "        return kwargs\n",
    "\n",
    "    def predict(self, feature_set: FeatureSet, **kwargs):\n",
    "        field = kwargs[\"input_field\"] #Fetch the field name which will have input\n",
    "        text_list = feature_set.df[field].to_list() # convert input to text\n",
    "        #set up the security keys and url \n",
    "        self.header = {\"Content-Type\": \"application/json\", \"api-key\": self.API_KEY}\n",
    "        API_BASE = f\"https://{self.API_BASE}/openai/deployments/{self.engine}/chat/completions?api-version={self.API_VERSION}\"\n",
    "\n",
    "        # Format the examples for few-shot prompt\n",
    "        if self.examples is not None:\n",
    "            example_formatted = [\n",
    "                f\"### Example {idx} \\n\\n {i}\\n\\n\" for idx, i in enumerate(self.examples)\n",
    "            ]\n",
    "        else:\n",
    "            example_formatted = []\n",
    "\n",
    "        if len(example_formatted):\n",
    "            t = \"\\n\".join(\n",
    "                example_formatted\n",
    "            )  # have to do it outside due to limitation of f-string\n",
    "            prompt = f\"{self.prompt}\\n\\nBelow are the representative examples\\n\\n{t}\"\n",
    "        else:\n",
    "            prompt = f\"{self.prompt}\\n\\n {','.join(example_formatted)}\"\n",
    "\n",
    "        prompt = f\"{SYSTEM_PROMPT}\\n\\n{prompt}\\n\\n{end_seq}\"\n",
    "\n",
    "        payload = []\n",
    "        features_list = []\n",
    "        # Create the payload in the system and user prompt for GPT\n",
    "        for message in text_list:\n",
    "            payload = (\n",
    "                {\n",
    "                    \"messages\": [\n",
    "                        {\"role\": \"system\", \"content\": prompt},\n",
    "                        {\"role\": \"user\", \"content\": f\"{message}\"},\n",
    "                    ],\n",
    "                    \"temperature\": 0.2,\n",
    "                }\n",
    "            )\n",
    "\n",
    "            payload = json.dumps(payload)\n",
    "            # Make the call to GPT\n",
    "            s = requests.request(method=\"POST\", url=API_BASE, headers=self.header, data=payload, verify=True) \n",
    "            if s.status_code == 200:\n",
    "                output = s.json()[\"choices\"][0][\"message\"][\"content\"]\n",
    "            else:\n",
    "                output = \"\"\n",
    "\n",
    "            if len(output):\n",
    "                try:\n",
    "                    _, val = list(json.loads(output).items())[0] # Load the output from JSON as dictionary\n",
    "                    sub_val = [i.replace(\"\\n\", \"\") for i in [i.replace(\"<<\", \"\") for i in val.split(\">>\")] if len(i)][0] # parse the output which are enclosed in << and >>\n",
    "                    features_list.append({'attributes': {'input_raw': message, \"output\": sub_val }})\n",
    "                except: \n",
    "                    features_list.append({'attributes': {'input_raw': message, \"output\": \"NA\" }})\n",
    "            else:\n",
    "                features_list.append({'attributes': {'input_raw': message, \"output\": \"NA\" }})\n",
    "        # Create the feature set. here we will define the fields and their data type. All the feature list collected \n",
    "        # during LLM call will added into the feature set. \n",
    "        feature_dict = {\n",
    "            \"fields\": [\n",
    "                { \"name\": \"input_raw\", \"type\": \"esriFieldTypeString\" },  \n",
    "                { \"name\": \"output\", \"type\": \"esriFieldTypeString\" }, \n",
    "            ],\n",
    "            'geometryType': \"\",\n",
    "            'features':features_list\n",
    "        }\n",
    "\n",
    "        return FeatureSet.from_dict(feature_dict)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d4cac82-b182-4bab-8159-cba5cea70eeb",
   "metadata": {},
   "source": [
    "#### Key Points:\n",
    "- All of the aforementioned functions must be encapsulated within a class.\n",
    "- The class name should match the Inference Function file name, excluding the file extension.\n",
    "- The author can define helper functions either within the class or in the module.\n",
    "- The author is responsible for handling exceptions when calling an endpoint.\n",
    "- The input field name must correspond to the name of the input field selected from the input table or feature layer."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56346a42-7d53-46f0-af9c-0e3835e101a3",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## Create Custom ESRI Deep Learning Package (.dlpk) file"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "295a59e6-0d2c-4276-941e-45b8b1f554c8",
   "metadata": {},
   "source": [
    "To complete a custom NLP function setup, create an ESRI Deep Learning Package (.dlpk) file.\n",
    "\n",
    "Organize the files as follows:\n",
    "\n",
    "**Create a Folder**: Create a folder and include the custom NLP function file. In our case, we will create a folder `text_transform_gpt`. Inside this, we will place `text_transform_gpt.emd` and the NLP function file \n",
    "`MySeq2Seq.py`.\n",
    "\n",
    "  ```      \n",
    "text_transform_gpt\n",
    "├── MySeq2Seq.py\n",
    "└── text_transform_gpt.emd\n",
    "```\n",
    "\n",
    "**Zip the Folder**: Create a ZIP archive of the folder. Rename the .zip file to match the name of the ESRI Model Definition (.emd) file but with the .dlpk extension. The dlpk's name will be `text_transform_gpt.dlpk`\n",
    "\n",
    "This ESRI Deep Learning Package (.dlpk) file is now ready for use with the arcgis.learn Python API and **Transform Text Using Deep Learning (GeoAI)** Tool."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2caad17d-3b32-4b8c-9ec1-5f4e52591277",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Using the custom ESRI Deep Learning Package (.dlpk) file in arcgis.learn API"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c353e363-96d1-4ced-a103-96805e8f3f6b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from arcgis.learn.text import SequenceToSequence\n",
    "\n",
    "# Initialize the model with the dlpk\n",
    "model = SequenceToSequence.from_model(r\"Path_to_the_dlpk\\text_transform_gpt.dlpk\",\n",
    "                                     api_key=\"SECRET_KEY\",\n",
    "                                     api_base=\"SECRET_KEY\", \n",
    "                                     engine=\"SECRET_KEY\")\n",
    "\n",
    "# perform the inference\n",
    "model.predict([\"2965 AND 2967 W GLENMEADOW DRIVE, FAYETTEVILLE, AR\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4b8d60b-84dd-4978-bfa8-7ddef51edffc",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Tool Interface"
   ]
  },
  {
   "attachments": {
    "1269cd6c-e866-4952-addb-6683c4595e81.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "56abb090-1933-4041-8e7d-0086e5ca635a",
   "metadata": {},
   "source": [
    "![image.png](attachment:1269cd6c-e866-4952-addb-6683c4595e81.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ea9c10a-dfa4-4b4d-9116-78ed84ef3be3",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Sample Input Table"
   ]
  },
  {
   "attachments": {
    "c366ddb2-8825-4e0f-adde-5efd338c209f.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "35b1bea7-a8bf-4b3a-a356-24322dbb0e59",
   "metadata": {},
   "source": [
    "![image.png](attachment:c366ddb2-8825-4e0f-adde-5efd338c209f.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b678e7d-9119-4f21-a88a-2803fc3360d7",
   "metadata": {},
   "source": [
    "## After Extension file loading"
   ]
  },
  {
   "attachments": {
    "aa4089b3-9138-45cb-b121-198701748e9d.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "66f9f3ef-e53b-4c71-8532-852c1120687b",
   "metadata": {},
   "source": [
    "![Untitled (1).png](attachment:aa4089b3-9138-45cb-b121-198701748e9d.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43492168-00b0-44ac-b1a8-a1d5aaee40fa",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Sample Output Table"
   ]
  },
  {
   "attachments": {
    "2d5efb37-89cd-4bb0-8309-014684744518.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "6aaa9fed-0bd4-4aa2-8cfb-aefcdd54fcd2",
   "metadata": {},
   "source": [
    "![image.png](attachment:2d5efb37-89cd-4bb0-8309-014684744518.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0242f6d5",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ce393c87",
   "metadata": {},
   "source": [
    "In this notebook, we developed an address standardization model using GPT-3.5 with extensibility function. The dataset consisted of non-standard, incorrect house addresses from the United States. We leveraged the capabilities of GPT-3.5 to efficiently standardize input house addresses. Below are the results from the sample inputs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c7022abf",
   "metadata": {
    "vscode": {
     "languageId": "tex"
    }
   },
   "outputs": [],
   "source": [
    "| Input                                                                   | Output                                                         |\n",
    "| ----------------------------------------------------------------------- | -------------------------------------------------------------- |\n",
    "| Lot 59 Butternut Ridge Weston Meadows Subdivision, Oconomowoc, WI       | Lot 59 Butternut Ridge, Oconomowoc, WI                         |\n",
    "| LOT 8 MARBELLA (B) ST MONTEMAR, GUAYAMA, PR                             | Lot 8 Marbella St, Montemar, Guayama, PR                       |\n",
    "| LOT C SR 165 KM2.5CUCHILLAS ROAD LOMAS VERDES, NARANJITO, PR            | LOT C SR 165 KM2.5 CUCHILLAS ROAD, LOMAS VERDES, NARANJITO, PR |\n",
    "| LOT9 RD 4494 C/BRISAS, ISABELA, PR                                      | Lot 9 Rd 4494 C/Brisas, Isabela, PR                            |\n",
    "| LOT309 SABANERA DE DORADO, DORADO, PR                                   | Lot 309 Sabanera de Dorado, Dorado, PR                         |\n",
    "| LOT1 PUERTO RICO URBTANAMA, ARECIBO, PR                                 | Lot 1 Puerto Rico Urbanama, Arecibo, PR                        |\n",
    "| LOT300, 17 STLA ARBOLEDA DEV, SALINAS, PR                               | Lot 300, 17 Stla Arboleda Dev, Salinas, PR                     |\n",
    "| LOT 51 2STREPARTO DEL CARMEN DEV, PONCE, PR                             | Lot 51 2st Reparto Del Carmen Dev, Ponce, PR                   |\n",
    "| LOT 14 B ST REAPRTO ROSARIO SECT BUENA VISTABAYAMO, BAYAMON, PR         | Lot 14 B St Reaprto Rosario Sect Buena Vista, Bayamon, PR      |\n",
    "| LOT84  1 STLA CAMPINA DEVELOPMENT, SAN JUAN, PR                         | Lot 84 1 Stla Campina Development, San Juan, PR                |\n",
    "| LOT 3 KM 5 7 SR 816NUEVO WD, BAYAMON, PR                                | Lot 3 KM 5 7 SR 816 Nuevo Wd, Bayamon, PR                      |\n",
    "| LOT 5 KM 1 8 ST 6690COSTA AZUL SABANA WD, VEGA ALTA, PR                 | Lot 5 Km 1.8 St 6690 Costa Azul Sabana Wd, Vega Alta, PR       |\n",
    "| LOT 37- A 5 STREET LA JURADO  COMMUNITY, Caguas, PR                     | Lot 37-A 5 Street La Jurado Community, Caguas, PR              |\n",
    "| LOT 193 17 STPIEDRA GORDA WD, CAMUY, PR                                 | Lot 193 17 St Piedra Gorda Wd, Camuy, PR                       |\n",
    "| LOT 175  P. FLORES STREETVILLA CLEMENTE COMMUNITY, TOA BAJA, PR         | Lot 175 P. Flores St, Villa Clemente Community, Toa Baja, PR   |\n",
    "| LOT 22 KM 27.0 SR 1CAMPO REAL DEV, CAGUAS, PR                           | Lot 22 Km 27.0 Sr 1, Campo Real Dev, Caguas, PR                |\n",
    "| PARCELA SR 825 KM 0.9QUEBRADA CRUZ WD, TOA ALTA, PR                     | Parcela SR 825 KM 0.9 Quebrada Cruz WD, Toa Alta, PR           |\n",
    "| Lot #11 Bl 4 SEC 3 Sunrise Lak, Milford, PA                             | Lot 11 Bl 4 Sec 3 Sunrise Lak, Milford, PA                     |\n",
    "| Smith W B D, TRACT 28B-3, ACRES 1.075, Montgomery, TX                   | Smith W B D, Tract 28B-3, Montgomery, TX                       |\n",
    "| 1741, 1741 1/2, 1743, 1743 1/2 Hause, Los Angeles, CA                   | 1741 Hause, Los Angeles, CA                                    |"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}