{
    "nbformat": 4,
    "nbformat_minor": 0,
    "metadata": {
        "colab": {
            "provenance": [],
            "gpuType": "T4"
        },
        "kernelspec": {
            "name": "python3",
            "display_name": "Python 3"
        },
        "language_info": {
            "name": "python"
        }
    },
    "cells": [
        {
            "cell_type": "markdown",
            "source": [
                "# 🏦 FinEE - Finance Entity Extractor\n",
                "\n",
                "**Extract structured financial data from Indian banking messages in seconds.**\n",
                "\n",
                "This notebook demonstrates the `finee` Python package - a production-ready tool for parsing bank transaction messages."
            ],
            "metadata": {
                "id": "intro"
            }
        },
        {
            "cell_type": "markdown",
            "source": [
                "## 📦 Installation\n",
                "\n",
                "Install the package directly from PyPI:"
            ],
            "metadata": {
                "id": "install_header"
            }
        },
        {
            "cell_type": "code",
            "execution_count": null,
            "metadata": {
                "id": "install"
            },
            "outputs": [],
            "source": [
                "!pip install finee -q"
            ]
        },
        {
            "cell_type": "markdown",
            "source": [
                "## 🚀 Quick Demo\n",
                "\n",
                "Let's extract entities from a real HDFC Bank UPI transaction message:"
            ],
            "metadata": {
                "id": "demo_header"
            }
        },
        {
            "cell_type": "code",
            "source": [
                "from finee import extract\n",
                "\n",
                "# Sample HDFC Bank UPI transaction\n",
                "message = \"\"\"\n",
                "HDFC Bank: Rs.2,500.00 debited from A/c XX3545 on 28-12-2025.\n",
                "VPA: swiggy@ybl. UPI Ref: 534567891234.\n",
                "Not you? Call 18002586161\n",
                "\"\"\"\n",
                "\n",
                "# Extract entities (uses Regex + Rules, no GPU needed)\n",
                "result = extract(message)\n",
                "\n",
                "# Print structured output\n",
                "print(\"📊 Extracted Entities:\")\n",
                "print(f\"   Amount: ₹{result.amount}\")\n",
                "print(f\"   Type: {result.type}\")\n",
                "print(f\"   Account: ****{result.account}\")\n",
                "print(f\"   Date: {result.date}\")\n",
                "print(f\"   Reference: {result.reference}\")\n",
                "print(f\"   Merchant: {result.merchant}\")\n",
                "print(f\"   Category: {result.category}\")\n",
                "print(f\"   Confidence: {result.confidence.value}\")"
            ],
            "metadata": {
                "id": "demo_hdfc"
            },
            "execution_count": null,
            "outputs": []
        },
        {
            "cell_type": "markdown",
            "source": [
                "## 📄 JSON Output\n",
                "\n",
                "Get the result as a clean JSON object:"
            ],
            "metadata": {
                "id": "json_header"
            }
        },
        {
            "cell_type": "code",
            "source": [
                "import json\n",
                "\n",
                "# Export as JSON\n",
                "json_output = result.to_dict()\n",
                "print(json.dumps(json_output, indent=2))"
            ],
            "metadata": {
                "id": "json_output"
            },
            "execution_count": null,
            "outputs": []
        },
        {
            "cell_type": "markdown",
            "source": [
                "## 🏦 Multi-Bank Support\n",
                "\n",
                "FinEE works across all major Indian banks:"
            ],
            "metadata": {
                "id": "multibank_header"
            }
        },
        {
            "cell_type": "code",
            "source": [
                "banks = {\n",
                "    \"ICICI\": \"Dear Customer, Rs.1500 debited from Acct XX9876 on 15-01-2025 to amazon@apl. Ref: 987654321012\",\n",
                "    \"SBI\": \"SBI: Rs.350 debited from a/c XX1234 on 10-01-25. UPI txn to zomato@paytm. Ref: 456789012345\",\n",
                "    \"Axis\": \"Axis Bank: INR 800 debited from A/c 5678 on 05-01-2025. Info: UPI-UBER. Bal: Rs.12,500\",\n",
                "    \"Kotak\": \"Rs.2000 credited to Kotak A/c XX4321 on 20-01-2025 from rahul.sharma@okicici. Ref: 321654987012\"\n",
                "}\n",
                "\n",
                "print(\"🏦 Multi-Bank Extraction Results:\\n\")\n",
                "for bank, msg in banks.items():\n",
                "    r = extract(msg)\n",
                "    print(f\"{bank:6} | ₹{str(r.amount):>8} | {r.type:6} | {(r.merchant or 'N/A'):12} | {r.confidence.value}\")"
            ],
            "metadata": {
                "id": "multibank_demo"
            },
            "execution_count": null,
            "outputs": []
        },
        {
            "cell_type": "markdown",
            "source": [
                "## ⚡ Performance\n",
                "\n",
                "The Regex+Rules pipeline is blazing fast:"
            ],
            "metadata": {
                "id": "perf_header"
            }
        },
        {
            "cell_type": "code",
            "source": [
                "import time\n",
                "\n",
                "# Benchmark\n",
                "test_msg = \"Rs.500 debited from A/c 1234 to paytm@ybl on 01-01-2025\"\n",
                "\n",
                "start = time.time()\n",
                "for _ in range(1000):\n",
                "    extract(test_msg)\n",
                "elapsed = (time.time() - start) * 1000  # ms\n",
                "\n",
                "print(f\"⚡ 1000 extractions in {elapsed:.1f}ms\")\n",
                "print(f\"   Average: {elapsed/1000:.3f}ms per message\")\n",
                "print(f\"   Throughput: {1000000/elapsed:.0f} messages/second\")"
            ],
            "metadata": {
                "id": "benchmark"
            },
            "execution_count": null,
            "outputs": []
        },
        {
            "cell_type": "markdown",
            "source": [
                "## 📚 Learn More\n",
                "\n",
                "- 📦 **PyPI**: `pip install finee`\n",
                "- 🐙 **GitHub**: [Ranjitbehera0034/Finance-Entity-Extractor](https://github.com/Ranjitbehera0034/Finance-Entity-Extractor)\n",
                "- 🤗 **Model**: [Ranjit0034/finance-entity-extractor](https://huggingface.co/Ranjit0034/finance-entity-extractor)"
            ],
            "metadata": {
                "id": "links"
            }
        }
    ]
}