L_4.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "bae53a09-c04c-4916-a146-00e0bb188954",
   "metadata": {},
   "source": [
    "# L4: Support Data Insight Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc6c1339-5278-4daf-8e97-04c280bde548",
   "metadata": {},
   "source": [
    "## Initial Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "41d1ce58-9d2e-4349-acbd-49227e37d31c",
   "metadata": {
    "height": 200
   },
   "outputs": [],
   "source": [
    "# Warning control\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# Load environment variables\n",
    "from helper import load_env\n",
    "load_env()\n",
    "\n",
    "import os\n",
    "import yaml\n",
    "from crewai import Agent, Task, Crew"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe1b574a-502a-49ec-9dc4-1e4e32754986",
   "metadata": {},
   "source": [
    "## Loading Tasks and Agents YAML files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0f171b23-b3cc-48c9-a8ba-07358e271e63",
   "metadata": {
    "height": 268
   },
   "outputs": [],
   "source": [
    "# Define file paths for YAML configurations\n",
    "files = {\n",
    "    'agents': 'config/agents.yaml',\n",
    "    'tasks': 'config/tasks.yaml'\n",
    "}\n",
    "\n",
    "# Load configurations from YAML files\n",
    "configs = {}\n",
    "for config_type, file_path in files.items():\n",
    "    with open(file_path, 'r') as file:\n",
    "        configs[config_type] = yaml.safe_load(file)\n",
    "\n",
    "# Assign loaded configurations to specific variables\n",
    "agents_config = configs['agents']\n",
    "tasks_config = configs['tasks']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27064ec0-4a12-4dbd-b9b2-c77d4aaaedb5",
   "metadata": {},
   "source": [
    "## Using FileReadTool"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f58e827a-8e2e-4f1d-8f8c-9486828f951d",
   "metadata": {
    "height": 47
   },
   "outputs": [],
   "source": [
    "from crewai_tools import FileReadTool\n",
    "csv_tool = FileReadTool(file_path='./support_tickets_data.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7dc509e7-5aae-4795-a5f1-822cd5bfaefa",
   "metadata": {},
   "source": [
    "## Creating Agents, Tasks and Crew"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "33d2c719-9823-4d12-bd87-219f5145f7dc",
   "metadata": {
    "height": 948
   },
   "outputs": [],
   "source": [
    "# Creating Agents\n",
    "suggestion_generation_agent = Agent(\n",
    "  config=agents_config['suggestion_generation_agent'],\n",
    "  tools=[csv_tool]\n",
    ")\n",
    "\n",
    "reporting_agent = Agent(\n",
    "  config=agents_config['reporting_agent'],\n",
    "  tools=[csv_tool]\n",
    ")\n",
    "\n",
    "chart_generation_agent = Agent(\n",
    "  config=agents_config['chart_generation_agent'],\n",
    "  allow_code_execution=True\n",
    ")\n",
    "\n",
    "# Creating Tasks\n",
    "suggestion_generation = Task(\n",
    "  config=tasks_config['suggestion_generation'],\n",
    "  agent=suggestion_generation_agent\n",
    ")\n",
    "\n",
    "table_generation = Task(\n",
    "  config=tasks_config['table_generation'],\n",
    "  agent=reporting_agent\n",
    ")\n",
    "\n",
    "chart_generation = Task(\n",
    "  config=tasks_config['chart_generation'],\n",
    "  agent=chart_generation_agent\n",
    ")\n",
    "\n",
    "final_report_assembly = Task(\n",
    "  config=tasks_config['final_report_assembly'],\n",
    "  agent=reporting_agent,\n",
    "  context=[suggestion_generation, table_generation, chart_generation]\n",
    ")\n",
    "\n",
    "\n",
    "# Creating Crew\n",
    "support_report_crew = Crew(\n",
    "  agents=[\n",
    "    suggestion_generation_agent,\n",
    "    reporting_agent,\n",
    "    chart_generation_agent\n",
    "  ],\n",
    "  tasks=[\n",
    "    suggestion_generation,\n",
    "    table_generation,\n",
    "    chart_generation,\n",
    "    final_report_assembly\n",
    "  ],\n",
    "  verbose=True\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "299bca36-395b-45f2-92e7-27b403e32a27",
   "metadata": {},
   "source": [
    "## Testing our Crew"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b6646acc-e280-44cc-8968-08db37115eba",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": [
    "support_report_crew.test(n_iterations=1, openai_model_name='gpt-4o')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d1fa2c2-6256-4420-8b7b-c8d5e09bbc01",
   "metadata": {},
   "source": [
    "## Training your crew and agents"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5a4e362d-c010-43dd-88c4-e7db87834fb4",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": [
    "support_report_crew.train(n_iterations=1, filename='training.pkl')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39e94a74-8700-4f83-b732-01b8d63dd774",
   "metadata": {},
   "source": [
    "## Comparing new test results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0fc73842-8162-49f9-b4a1-5dc7d86f3e48",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": [
    "support_report_crew.test(n_iterations=1, openai_model_name='gpt-4o')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0e8626d2-e48a-4b7e-a061-a8eb492c9036",
   "metadata": {
    "height": 115
   },
   "outputs": [],
   "source": [
    "# Display the Trello screenshot\n",
    "from IPython.display import Image, display\n",
    "\n",
    "# Load and display the image\n",
    "test_image = Image(filename='test_before_training.png', width=368)\n",
    "display(test_image)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d1437cf-ebc1-4e46-b982-7eeec5462479",
   "metadata": {},
   "source": [
    "## Kicking off Crew"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a655927f-c10c-4c06-852a-d9c96fdfbfb9",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": [
    "result = support_report_crew.kickoff()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d039f919-2d28-4871-b316-8813e9adebde",
   "metadata": {},
   "source": [
    "## Result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9e861574-0f9b-4f2c-b2d1-a230fc3a53a3",
   "metadata": {
    "height": 47
   },
   "outputs": [],
   "source": [
    "from IPython.display import display, Markdown\n",
    "display(Markdown(result.raw))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1a951d6a-8363-44f4-8783-e8397f435a32",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fa40b53e-0a49-4198-a263-c79a6a3af603",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f5fb6a8a-7ae7-4ae6-99a7-92aa09d97d7f",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "66b3b165-2a4b-46cc-93f7-696dffff1e10",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7c2a2af-3e55-4de9-86bb-d33c21ade238",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "109926d3-50e5-43e7-87f8-a1d38e45d79d",
   "metadata": {
    "height": 30
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}