{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c9dc167c",
   "metadata": {},
   "source": [
    "# Introduction to AI and machine learning in Python: Decision Trees"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37e5dcbb",
   "metadata": {},
   "source": [
    "<h2 align=\"center\" style=\"color:red;\">INTRODUCTION</h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b1b4405-6cac-43e1-b783-18843abee67b",
   "metadata": {},
   "source": [
    "CLASSIFICATION \n",
    "*Classification with non-neural networks, non-deep learning algorithms*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e417834c",
   "metadata": {},
   "source": [
    "## Decision Trees with Scikit-Learn - basics"
   ]
  },
  {
   "attachments": {
    "decision_tree_model.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "7848d737",
   "metadata": {},
   "source": [
    "![decision_tree_model.png](attachment:decision_tree_model.png)\n",
    "\n",
    "/source of the picture: SuperAI.pl/"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fae78eab",
   "metadata": {},
   "source": [
    "**Decision Trees Classifier with Scikit-Learn: classification on Iris Dataset**\n",
    "- https://scikit-learn.org/stable/modules/tree.html\n",
    "- https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html\n",
    "- https://scikit-learn.org/stable/auto_examples/tree/plot_unveil_tree_structure.html\n",
    "- https://scikit-learn.org/stable/modules/cross_validation.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f06e5953",
   "metadata": {},
   "outputs": [],
   "source": [
    "#0. Setup the environment for training, testing, and using SuperAI with chosen machine learning techniques. \n",
    "#0A. Open Anaconda.\n",
    "#0B. Create and set (if you want) the environment to work in. It must be done only once per preffered settings.\n",
    "#0C. Open Jupyter Notebook.\n",
    "#0D. Import and use (if you want) the 'warnings' library that is preinstalled with Anaconda to make the ouput cleaner.\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9305d181-d0dd-4ea3-9bce-42d14fba10ba",
   "metadata": {},
   "source": [
    "**BASIC INSTALLS**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4efc2828-89be-411c-9750-0ae5941652d8",
   "metadata": {},
   "source": [
    "<h3 style=\"color:red;\"><font color=\"red\">Environment: decision-trees-env</font></h3>\n",
    "\n",
    "-\tAnaconda Navigator: 2.6.2 \n",
    "-\tjupyter Notebook: 7.0.8\n",
    "-\tCMD.exe Prompt: 0.1.1\n",
    "-\tPython version: 3.11.9\n",
    "-\tPip version: 24.2\n",
    "-\tscikit-learn version: 1.5.1\n",
    "-\tmatplotlib version: 3.9.0\n",
    "-\tgraphviz version: 0.20.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b69a53f2-6e16-4fd2-bf9f-a4fd9b66f077",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn\n",
    "import graphviz\n",
    "import matplotlib\n",
    "import pip\n",
    "import sys\n",
    "\n",
    "print(\"Python version: {}\".format(sys.version))\n",
    "print(\"pip version: {}\".format(pip.__version__))\n",
    "print(\"scikit-learn version: {}\".format(sklearn.__version__))\n",
    "print(\"graphviz version: {}\".format(graphviz.__version__))\n",
    "print(\"matplotlib version: {}\".format(matplotlib.__version__))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76d434e2-9627-455e-a1c0-20819fb044dc",
   "metadata": {},
   "source": [
    "https://pypi.org/project/scikit-learn/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b6bef2be-bb27-4700-93e7-fde89ffc8e51",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "!pip install scikit-learn==1.5.1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fade351-28ba-47a3-ba0a-fb7916d61697",
   "metadata": {},
   "source": [
    "https://pypi.org/project/graphviz/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c80baa70-43f9-4a90-9560-7d7e9176f6f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#!pip install graphviz==0.20.1 #doesn't seem to work properly on Windows at the moment"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2e22955-06f9-4d57-8fff-0d8ef9cb1f0c",
   "metadata": {},
   "source": [
    "For Windows in the CMD.exe Prompt (from Anaconda) run:\n",
    "\n",
    "conda install python-graphviz==0.20.1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc58ee49-b697-4557-bc04-7a0d8a0892d1",
   "metadata": {},
   "source": [
    "https://pypi.org/project/matplotlib/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1c03d4d4-80d6-4189-9550-fb46f58e1811",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "!pip install matplotlib==3.9.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cbed1df1-56f6-480c-841c-100057b0dc36",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn\n",
    "import graphviz\n",
    "import matplotlib\n",
    "import pip\n",
    "import sys\n",
    "\n",
    "print(\"Python version: {}\".format(sys.version))\n",
    "print(\"pip version: {}\".format(pip.__version__))\n",
    "print(\"scikit-learn version: {}\".format(sklearn.__version__))\n",
    "print(\"graphviz version: {}\".format(graphviz.__version__))\n",
    "print(\"matplotlib version: {}\".format(matplotlib.__version__))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92ae8fc6",
   "metadata": {},
   "source": [
    "##### BASIC COPY - PASTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d6bf13ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "from sklearn import tree\n",
    "iris = load_iris()\n",
    "X, y = iris.data, iris.target\n",
    "clf = tree.DecisionTreeClassifier()\n",
    "clf = clf.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5dbcb404",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "tree.plot_tree(clf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a6504f4c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import graphviz \n",
    "dot_data = tree.export_graphviz(clf, out_file=None) \n",
    "graph = graphviz.Source(dot_data) \n",
    "graph.render(\"iris\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4e096aff",
   "metadata": {},
   "outputs": [],
   "source": [
    "dot_data = tree.export_graphviz(clf, out_file=None, \n",
    "                     feature_names=iris.feature_names,  \n",
    "                     class_names=iris.target_names,  \n",
    "                     filled=True, rounded=True,  \n",
    "                     special_characters=True)  \n",
    "graph = graphviz.Source(dot_data)  \n",
    "graph "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b78aeb79",
   "metadata": {},
   "source": [
    "##### BASIC IMPROVEMENTS\n",
    "\n",
    "**Checking overall score of the classifier**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "937fdb50",
   "metadata": {},
   "outputs": [],
   "source": [
    "clf.score(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10768fbf",
   "metadata": {},
   "source": [
    "**Adding train - test split**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b6ab1330",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "\n",
    "clf = tree.DecisionTreeClassifier(random_state=0)\n",
    "clf = clf.fit(X_train, y_train)\n",
    "\n",
    "clf.score(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8412d30d",
   "metadata": {},
   "outputs": [],
   "source": [
    "clf.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "78131d5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "clf.score(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "005ae514",
   "metadata": {},
   "outputs": [],
   "source": [
    "tree.plot_tree(clf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "26c6b5e2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import graphviz \n",
    "dot_data = tree.export_graphviz(clf, out_file=None) \n",
    "graph = graphviz.Source(dot_data) \n",
    "graph.render(\"iris\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6a5b7eec",
   "metadata": {},
   "outputs": [],
   "source": [
    "dot_data = tree.export_graphviz(clf, out_file=None, \n",
    "                     feature_names=iris.feature_names,  \n",
    "                     class_names=iris.target_names,  \n",
    "                     filled=True, rounded=True,  \n",
    "                     special_characters=True)  \n",
    "graph = graphviz.Source(dot_data)  \n",
    "graph "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ced381a",
   "metadata": {},
   "source": [
    "<h4 align=\"right\"><i>Always ask at first - what is the goal you are trying to achieve, what is the model for in real life, how you will use it.</i></h4>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "895cc629",
   "metadata": {},
   "source": [
    "## Decision Trees with Scikit-Learn - diving deeper"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2b77ddd",
   "metadata": {},
   "source": [
    "### Useful knowledge regarding the project\n",
    "\n",
    "**Used Dataset:** Breast Cancer Wisconsin Dataset\n",
    "\n",
    "**Problem to solve / Decision to make:** Classification (malignant / benign tumor)\n",
    "\n",
    "**ML Libraries used to solve the problem:** Scikit-Learn \n",
    "\n",
    "**ML Methods / Algorithms used to solve the problem:** Decision Tree\n",
    "\n",
    "**General info about the used Methods / Algorithms:**\n",
    "\n",
    "You can read more about Decision Trees at:\n",
    "- https://en.wikipedia.org/wiki/Decision_tree_learning\n",
    "- https://en.wikipedia.org/wiki/Decision_tree\n",
    "\n",
    "**More info about the Methods/Algorithms + Source of the Example + Source of the Dataset:**\n",
    "\n",
    "We'll use Scikit-Learn, one of the most popular ML libraries. You can read more about using DT with Scikit-Learn (classification and regression) at their website:\n",
    "- https://scikit-learn.org/stable/modules/tree.html\n",
    "\n",
    "And especially you can read about Decision Tree Classifier:\n",
    "- https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html\n",
    "- https://scikit-learn.org/stable/auto_examples/tree/plot_unveil_tree_structure.html\n",
    "- https://scikit-learn.org/stable/modules/cross_validation.html\n",
    "\n",
    "At first we'll see how it works with the dataset from Scikit-Learn pool of datasets:\n",
    "- https://scikit-learn.org/stable/datasets.html\n",
    "\n",
    "We want the dataset that is good for classification and with the least possible classes. So we'll use the Breast Cancer Wisconsin Dataset, because it is, in a way, one of the simplest - it has only 2 classes beteween which the algorithms chooses: malignant and benign.\n",
    "- https://archive.ics.uci.edu/dataset/17/breast+cancer+wisconsin+diagnostic\n",
    "- https://scikit-learn.org/stable/datasets/toy_dataset.html\n",
    "\n",
    "With this dataset we can also see how helpful the machine learning can be in diagnostic medicine. The algorithm will help us to decide wether the tumor that was detected was benign (non-cancerous) or malignant - cancerous. If you want to learn more about difference between benign and malignant tumors you can read or watch about it in the Internet. It's good to know about it."
   ]
  },
  {
   "attachments": {
    "decision_tree_model.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "9f150205",
   "metadata": {},
   "source": [
    "![decision_tree_model.png](attachment:decision_tree_model.png)\n",
    "\n",
    "/source of the picture: SuperAI.pl/"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "096a5238",
   "metadata": {},
   "source": [
    "###  Basic steps in creating a helpful AI that uses machine learning techniques\n",
    "\n",
    "**0. Setup the environment for training, testing, and using SuperAI with chosen machine learning techniques.** \n",
    "\n",
    "**0A. Open Anaconda.**\n",
    "\n",
    "**0B. Create and set (if you want) the environment to work in. It must be done only once per preffered settings.**\n",
    "\n",
    "**0C. Open Jupyter Notebook.**\n",
    "\n",
    "**0D. Import and use (if you want) the 'warnings' library that is preinstalled with Anaconda to make the ouput cleaner.**\n",
    "\n",
    "**1. Install and import all the libraries for machine learning and helper libraries**\n",
    "\n",
    "**1A. Install all the necessary libraries not yet installed (must be done only once in one evironment)**\n",
    "\n",
    "**1B. Load/Import the libraries for machine learning and helper libraries (check the versions of the libraries)**\n",
    "\n",
    "**2. Prepare computer environment for using the model (training, testing, etc.), e.g. set CPU or GPU device for working.**\n",
    "\n",
    "**3. Load the data (or create some data for learning purposes, training and testing the model).**\n",
    "\n",
    "**4. Check the data: basic description, some examples.**\n",
    "\n",
    "**5. Prepare the data for using with the chosen machine learning technique (e.g. normalize pixel values) and recheck the data.**\n",
    "\n",
    "**6. Prepare the data for tackling overfitting problem, e. g. divide the data into train, test sets and recheck the data.**\n",
    "\n",
    "**7. Create the model (define the model, choose and set the hyperparameters, etc.).**\n",
    "\n",
    "**8. Visualize the model (as it is at the moment).**\n",
    "\n",
    "**9. Prepare the functions for training and testing the model (if neccessary).**\n",
    "\n",
    "**10. Train the model.**\n",
    "\n",
    "**11. Test the model (remember to test the model in the evaluation mode (dropouts, batch normalizations, etc.), if applies)**\n",
    "\n",
    "**12. Show the specific results and metrics (if you want).**\n",
    "\n",
    "**13. Visualize the trained model.**\n",
    "\n",
    "**14. Prepare the model for using with one example (prediction, generation, etc.).**\n",
    "\n",
    "**15. Check the model with some examples.**\n",
    "\n",
    "{\n",
    "\n",
    "**16. Modify the model (extra step if you want, but it's not neccessary).**\n",
    "\n",
    "**17. Test the modified model (extra step, if applies).**\n",
    "\n",
    "} * x\n",
    "\n",
    "**18. Save the trained model.**\n",
    "\n",
    "**19. Load and test the saved trained model.**\n",
    "\n",
    "**20. Combine all the necessary steps in one SuperAI that will be able to use the model.**\n",
    "\n",
    "**21. Check the SuperAI.**\n",
    "\n",
    "**And you are ready to use this great SuperAI!**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "224b5d2e-964e-4196-b137-19cdddc2aba6",
   "metadata": {},
   "source": [
    "<h3 align=\"left\" style=\"color:red;\">READY TO WORK COMPUTER = SETTING UP THE WORK STATION</h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc0006e4",
   "metadata": {},
   "source": [
    "### 0. Setup the environment for training, testing, and using SuperAI with chosen machine learning techniques.\n",
    "Setting up the work station = Ready to work computer\n",
    "\n",
    "To start working with this notebook, you need Python and Jupyter Notebook.\n",
    "\n",
    "To get Python, go to https://www.anaconda.com, download it, and install it.\n",
    "\n",
    "After you install Anaconda on your computer, create a new environment in it (with 1 click) and install Jupyter Notebook and CMD.exe Prompt within the environment (with 2 clicks). If you don't know how to do it, you can check one of my previous tutorials: Python 101 for Beginners (https://superai.pl/courses/python_101_for_beginners.html).\n",
    "\n",
    "After you do it, you are ready to start working with this notebook.\n",
    "\n",
    "So here we are:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef1c8882-17f1-420a-b74d-6b811ff99d30",
   "metadata": {},
   "source": [
    "<table>\n",
    "    <tr width=\"1000px\">\n",
    "        <td width=\"800px\" align=\"center\">\n",
    "            <p align=\"center\"><b>0. Setup the environment for training, testing, and using SuperAI with chosen machine learning techniques.</b></p>\n",
    "        </td>\n",
    "        <td width=\"200px\" align=\"center\">\n",
    "           <input type=\"checkbox\" checked>\n",
    "        </td>\n",
    "    </tr>\n",
    "    <tr width=\"1000px\">\n",
    "        <td width=\"800px\" align=\"center\">\n",
    "            <p align=\"center\">0A. Open Anaconda.</p>\n",
    "        </td>\n",
    "        <td width=\"200px\" align=\"center\">\n",
    "           <input type=\"checkbox\" checked>\n",
    "        </td>\n",
    "    </tr>\n",
    "    <tr width=\"1000px\">\n",
    "        <td width=\"800px\" align=\"center\">\n",
    "            <p align=\"center\">0B. Create and set (if you want) the environment to work in. It must be done only once per preffered settings.</p>\n",
    "        </td>\n",
    "        <td width=\"200px\" align=\"center\">\n",
    "           <input type=\"checkbox\" checked>\n",
    "        </td>\n",
    "    </tr>\n",
    "    <tr width=\"1000px\">\n",
    "        <td width=\"800px\" align=\"center\">\n",
    "            <p align=\"center\">0C. Open Jupyter Notebook.</p>\n",
    "        </td>\n",
    "        <td width=\"200px\" align=\"center\">\n",
    "           <input type=\"checkbox\" checked>\n",
    "        </td>\n",
    "    </tr>\n",
    "    <tr width=\"1000px\">\n",
    "        <td width=\"800px\" align=\"center\">\n",
    "            <p align=\"center\">0D. Import and use (if you want) the 'warnings' library that is preinstalled with Anaconda to make the ouput cleaner.</p>\n",
    "        </td>\n",
    "        <td width=\"200px\" align=\"center\">\n",
    "           <input type=\"checkbox\">\n",
    "        </td>\n",
    "    </tr>\n",
    "    <tr width=\"1000px\">\n",
    "        <td width=\"800px\" align=\"center\">\n",
    "            <p align=\"center\"><b>1. Install and import all the libraries for machine learning and helper libraries</b></p>\n",
    "        </td>\n",
    "        <td width=\"200px\" align=\"center\">\n",
    "           <input type=\"checkbox\">\n",
    "        </td>\n",
    "    </tr>\n",
    "</table>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86ccf79a",
   "metadata": {},
   "source": [
    "And here comes the code:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4629fb55",
   "metadata": {},
   "outputs": [],
   "source": [
    "#0. Setup the environment for training, testing, and using SuperAI with chosen machine learning techniques. \n",
    "#0A. Open Anaconda.\n",
    "#0B. Create and set (if you want) the environment to work in. It must be done only once per preffered settings.\n",
    "#0C. Open Jupyter Notebook.\n",
    "#0D. Import and use (if you want) the 'warnings' library that is preinstalled with Anaconda to make the ouput cleaner.\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "679e6c1d-3573-4fdd-aeb0-fcf1518b5a89",
   "metadata": {},
   "source": [
    "<h3 style=\"color:red;\"><font color=\"red\">Environment: decision-trees-env</font></h3>\n",
    "\n",
    "-\tAnaconda Navigator: 2.6.2 \n",
    "-\tjupyter Notebook: 7.0.8\n",
    "-\tCMD.exe Prompt: 0.1.1\n",
    "-\tPython version: 3.11.9\n",
    "-\tPip version: 24.2\n",
    "-\tscikit-learn version: 1.5.1\n",
    "-\tmatplotlib version: 3.9.0\n",
    "-\tgraphviz version: 0.20.1\n",
    "-\tnumpy: 2.1.0\n",
    "-\tpandas: 2.2.2\n",
    "-\tpydot: 2.0.0"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3cda802a-0630-4ea9-a3b3-275537320c4d",
   "metadata": {},
   "source": [
    "### 1. Install and import all the libraries for machine learning and helper libraries\n",
    "### 1A. Install all the necessary libraries not yet installed (must be done only once in one evironment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2031a861",
   "metadata": {},
   "outputs": [],
   "source": [
    "#1. Install and import all the libraries for machine learning and helper libraries\n",
    "#1A. Install all the necessary libraries not yet installed (must be done only once in one evironment)\n",
    "\n",
    "#!pip install graphviz --upgrade\n",
    "#!pip install pandas --upgrade\n",
    "#!pip install pydot --upgrade\n",
    "#!pip install scikit-learn --upgrade"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a0b9875-e0ad-4ab3-92cb-fb7477321fae",
   "metadata": {},
   "source": [
    "### 1B. Load/Import the libraries for machine learning and helper libraries (check the versions of the libraries)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bddf911c",
   "metadata": {},
   "outputs": [],
   "source": [
    "#1B. Load/Import the libraries for machine learning and helper libraries (check the versions of the libraries)\n",
    "\n",
    "import sklearn\n",
    "from sklearn import tree\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.tree import export_graphviz\n",
    "\n",
    "import graphviz\n",
    "import IPython\n",
    "from IPython.display import Image\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import os\n",
    "import pandas as pd\n",
    "import pickle\n",
    "import pydot\n",
    "import sys\n",
    "\n",
    "print(\"Python version: {}\".format(sys.version))\n",
    "print(\"scikit-learn version: {}\".format(sklearn.__version__))\n",
    "print(\"graphviz version: {}\".format(graphviz.__version__))\n",
    "print(\"IPython version: {}\".format(IPython.__version__))\n",
    "print(\"Matplotlib version: {}\".format(mpl.__version__))\n",
    "print(\"Numpy version: {}\".format(np.__version__))\n",
    "print(\"Pandas version: {}\".format(pd.__version__))\n",
    "print(\"pickle version: {}\".format(pickle.format_version))\n",
    "print(\"pydot version: {}\".format(pydot.__version__))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0da61ebc-cf83-489e-bb4b-80ed83e9cba3",
   "metadata": {},
   "source": [
    "So, if you want to recreate this environment with the same versions of libraries as I have:\n",
    "\n",
    "1) Create an environment with Python version: 3.11.9 \n",
    "\n",
    "2) Put this into requirements.txt:\n",
    "\n",
    "scikit-learn==1.5.0\n",
    "\n",
    "graphviz==0.20.1\n",
    "\n",
    "IPython==8.20.0\n",
    "\n",
    "matplotlib==3.9.0\n",
    "\n",
    "numpy==2.1.0\n",
    "\n",
    "pandas==2.2.2\n",
    "\n",
    "pydot==2.0.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4fb3615f-b690-4bc9-aa9e-9b010ae6a015",
   "metadata": {},
   "outputs": [],
   "source": [
    "requirements = \"requirements.txt\"\n",
    "with open(requirements, 'w', encoding=\"utf-8\") as f:\n",
    "    print(\n",
    "\"\"\"scikit-learn==1.5.1\n",
    "graphviz==0.20.1\n",
    "IPython==8.25.0\n",
    "matplotlib==3.9.0\n",
    "numpy==2.1.0\n",
    "pandas==2.2.2\n",
    "pydot==2.0.0\"\"\", file=f)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d51f1637-6ff4-4486-b5b3-27d53af690f3",
   "metadata": {},
   "source": [
    "3) And run the line:\n",
    "\n",
    "!pip install -r requirements.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2c959043-cb68-44a2-ab28-27eae5bfc687",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "!pip install -r requirements.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "541bd164-bbf4-4cc7-bf94-b5d9f5efd1b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "#1B. Load/Import the libraries for machine learning and helper libraries (check the versions of the libraries)\n",
    "\n",
    "import sklearn\n",
    "from sklearn import tree\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.tree import export_graphviz\n",
    "\n",
    "import graphviz\n",
    "import IPython\n",
    "from IPython.display import Image\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import os\n",
    "import pandas as pd\n",
    "import pickle\n",
    "import pydot\n",
    "import sys\n",
    "\n",
    "print(\"Python version: {}\".format(sys.version))\n",
    "print(\"scikit-learn version: {}\".format(sklearn.__version__))\n",
    "print(\"graphviz version: {}\".format(graphviz.__version__))\n",
    "print(\"IPython version: {}\".format(IPython.__version__))\n",
    "print(\"Matplotlib version: {}\".format(mpl.__version__))\n",
    "print(\"Numpy version: {}\".format(np.__version__))\n",
    "print(\"Pandas version: {}\".format(pd.__version__))\n",
    "print(\"pickle version: {}\".format(pickle.format_version))\n",
    "print(\"pydot version: {}\".format(pydot.__version__))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "978a3275-4664-47b5-8377-36cea93c86b0",
   "metadata": {},
   "source": [
    "### 2. Prepare computer environment for using the model (training, testing, etc.), e.g. set CPU or GPU device for working."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bd572e34",
   "metadata": {},
   "outputs": [],
   "source": [
    "#2. Prepare computer environment for using the model (training, testing, etc.), e.g. set CPU or GPU device for working.\n",
    "# Nothing to do here at the moment."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "639c2b6f",
   "metadata": {},
   "source": [
    "### 3. Load the data (or create some data for learning purposes, training and testing the model)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74591407-52c6-47a4-b330-f2d8b9deac1e",
   "metadata": {},
   "source": [
    "<h3 align=\"left\" style=\"color:red;\">READY TO USE DATA = PREPARING THE DATA, PREPROCESSING</h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "732b1c2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#3. Load the data (or create some data for learning purposes, training and testing the model).\n",
    "breast_cancer_data = load_breast_cancer(as_frame=True)\n",
    "\n",
    "X = features = breast_cancer_data.data.copy()\n",
    "y = target = breast_cancer_data.target.copy()\n",
    "\n",
    "classes = [breast_cancer_data.target_names[0], breast_cancer_data.target_names[1]]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3915d75b-24de-4fc8-acd6-564c4e613576",
   "metadata": {},
   "source": [
    "### 4. Check the data: basic description, some examples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cd8d214d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#4. Check the data: basic description, some examples.\n",
    "print(\"There are {} items in {} classes in the dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X), len(classes), \n",
    "                                        classes[0], target.value_counts()[0],  target.value_counts()[0] / len(target) * 100,\n",
    "                                        classes[1], target.value_counts()[1], target.value_counts()[1] / len(target) * 100))\n",
    "\n",
    "print(\"\\nThere are {} features we can take under consideration:\".format(len(breast_cancer_data.feature_names)), \n",
    "      breast_cancer_data.feature_names)\n",
    "\n",
    "print(\"\\nHere is a part of the dataframe with the features based on which we might try to learn how to classify the data:\")\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "084e49d3",
   "metadata": {},
   "source": [
    "<h4> DEALING WITH MISSING DATA</h4>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "abd712c1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "print(\"We check the data:\")\n",
    "X.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a55d7419",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "print(\"We check specifically for missing data:\")\n",
    "X.isnull().sum()\n",
    "#X.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1d680b05",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"We show the data:\")\n",
    "X.hist(bins=50, figsize=(20,15))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5114dfed",
   "metadata": {},
   "source": [
    "<h4 style=\"color:red;\">EXTRA: Messing with the data: Adding missing data and dealing with it</h4>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d2485e57-e35e-4fe2-a1ce-58baabf53ded",
   "metadata": {},
   "source": [
    "#### Adding missing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b8aff0ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "#EXTRA MESSING WITH THE DATA: adding missing data\n",
    "\n",
    "X_with_missing = X.copy()\n",
    "X_with_missing.iloc[0,0] = np.nan\n",
    "\n",
    "print(\"We've changed mean radius in 0th row (that was: {}) to NaN. Here is the table as it looks now:\".format(X.iloc[0,0]))\n",
    "X_with_missing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4e3f5f4b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "print(\"Here is the data check:\")\n",
    "X_with_missing.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4dec2d93",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "print(\"We check for missing data in our dataframe with missing data:\")\n",
    "X_with_missing.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f7b9a0b",
   "metadata": {},
   "source": [
    "#### Dropping the rows with missing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "56b55f8a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dropping the rows with missing data\n",
    "\n",
    "X_without_missing_dropped_example = X_with_missing.dropna(axis=0)\n",
    "\n",
    "print(\"We've dropped the first row. We have all the columns. Here is the head of the table as it looks now:\")\n",
    "\n",
    "X_without_missing_dropped_example.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3b436bcf",
   "metadata": {},
   "outputs": [],
   "source": [
    "#EXTRA MESSING WITH THE DATA: adding missing data in 3 columns\n",
    "\n",
    "X_with_missing2 = X.copy()\n",
    "X_with_missing2.iloc[0,0] = np.nan\n",
    "X_with_missing2.iloc[1,1] = np.nan\n",
    "X_with_missing2.iloc[2,2] = np.nan\n",
    "\n",
    "print(\"We've changed mean radius in the 0th row, mean texture in the 1st row, and mean parameter in the 2nd row.\\\n",
    "\\nHere is the table as it looks now:\")\n",
    "X_with_missing2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "574aaab2",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_with_missing2 = X_with_missing2.dropna(subset=[\"mean area\"]) #dropping when NA is in particular column, if it's there\n",
    "print(\"We've tried to drop a subset with NAs in mean area, but there wasn't anything to drop so nothig has changed.\\\n",
    "\\nHere is the table as it looks now:\")\n",
    "X_with_missing2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "32e852fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_with_missing2 = X_with_missing2.dropna(subset=[\"mean radius\"]) #dropping when NA is in particular column, otherwise do it without subset\n",
    "print(\"We did drop 0th row in a subset with NAs in mean radius, the rest of the table stays the same even with NAs.\\\n",
    "\\nHere is the table as it looks now:\")\n",
    "X_with_missing2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "62ddaf8c",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_with_missing2 = X_with_missing2.dropna()\n",
    "print(\"We did drop 0th row, 1st row, and 2nd row with all NAs in the table. Here is the table as it looks now:\")\n",
    "X_with_missing2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48efd999",
   "metadata": {},
   "source": [
    "#### Dropping the columns with missing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f3a9e0d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#EXTRA: MESSING WITH THE DATA: adding missing data\n",
    "\n",
    "X_with_missing = X.copy()\n",
    "X_with_missing.iloc[0,0] = np.nan\n",
    "\n",
    "print(\"We've changed mean radius in 0th row (that was: {}) to NaN. Here is the table as it looks now:\".format(X.iloc[0,0]))\n",
    "X_with_missing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6ab0b487",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dropping the columns with missing data.\n",
    "\n",
    "X_without_missing_dropped_column = X_with_missing.dropna(axis=1)\n",
    "\n",
    "print(\"We've dropped the first column. We have all the rows. Here is the table as it looks now:\")\n",
    "\n",
    "X_without_missing_dropped_column"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1079315a",
   "metadata": {},
   "source": [
    "#### Filling missing data with mean or median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a3b5a974",
   "metadata": {},
   "outputs": [],
   "source": [
    "#EXTRA: MESSING WITH THE DATA: adding missing data in 3 columns\n",
    "\n",
    "X_with_missing2 = X.copy()\n",
    "X_with_missing2.iloc[0,0] = np.nan\n",
    "X_with_missing2.iloc[1,1] = np.nan\n",
    "X_with_missing2.iloc[2,2] = np.nan\n",
    "\n",
    "print(\"We've changed mean radius in the 0th row, mean texture in the 1st row, and mean parameter in the 2nd row.\\\n",
    "\\nHere is the table as it looks now:\")\n",
    "X_with_missing2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b5bc5003",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "X_with_missing2.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fe300bbc",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "X.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f047176c",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Filling missing data with mean\n",
    "\n",
    "X_without_missing_filled_mean = X_with_missing2.fillna(X_with_missing2.mean())\n",
    "\n",
    "print(\"We've changed mean radius in 0th row (that was: {:.2f}) to mean of the rest, that is: {:.6f}. \\\n",
    "\\nHere is the table as it looks now:\\n\".format(X.iloc[0,0], X_without_missing_filled_mean.iloc[0,0]))\n",
    "\n",
    "print(\"We've changed mean texture in 1st row (that was: {:.2f}) to mean of the rest, that is: {:.6f}. \\\n",
    "\\nHere is the table as it looks now:\\n\".format(X.iloc[1,1], X_without_missing_filled_mean.iloc[1,1]))\n",
    "\n",
    "print(\"We've changed mean perimeter in 2nd row (that was: {:.2f}) to mean of the rest, that is: {:.6f}. \\\n",
    "\\nHere is the table as it looks now:\\n\".format(X.iloc[2,2], X_without_missing_filled_mean.iloc[2,2]))\n",
    "\n",
    "X_without_missing_filled_mean"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "94bc1a05-cccb-43ae-91d8-10b8566d2e86",
   "metadata": {},
   "source": [
    "#### Adding missing data in 3 columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c2e8ff31",
   "metadata": {},
   "outputs": [],
   "source": [
    "#EXTRA: MESSING WITH THE DATA: adding missing data in 3 columns\n",
    "\n",
    "X_with_missing2 = X.copy()\n",
    "X_with_missing2.iloc[0,0] = np.nan\n",
    "X_with_missing2.iloc[1,1] = np.nan\n",
    "X_with_missing2.iloc[2,2] = np.nan\n",
    "\n",
    "print(\"We've changed mean radius in the 0th row, mean texture in the 1st row, and mean parameter in the 2nd row. \\\n",
    "\\nHere is the table as it looks now:\")\n",
    "X_with_missing2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "65504258",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "X_with_missing2.median()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b563eeaf-0c6c-4055-8bcc-aff22e9a3572",
   "metadata": {},
   "source": [
    "#### Filling missing data with median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "21b37eb1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Filling missing data with median\n",
    "\n",
    "X_without_missing_filled_median = X_with_missing2.fillna(X_with_missing2.median())\n",
    "\n",
    "print(\"We've changed mean radius in 0th row (that was: {:.2f}) to median of the rest, that is: {:.6f}. \\\n",
    "\\nHere is the table as it looks now:\\n\".format(X.iloc[0,0], X_without_missing_filled_median.iloc[0,0]))\n",
    "\n",
    "print(\"We've changed mean texture in 1st row (that was: {:.2f}) to median of the rest, that is: {:.6f}. \\\n",
    "\\nHere is the table as it looks now:\\n\".format(X.iloc[1,1], X_without_missing_filled_median.iloc[1,1]))\n",
    "\n",
    "print(\"We've changed mean perimeter in 2nd row (that was: {:.2f}) to median of the rest, that is: {:.6f}. \\\n",
    "\\nHere is the table as it looks now:\\n\".format(X.iloc[2,2], X_without_missing_filled_median.iloc[2,2]))\n",
    "\n",
    "X_without_missing_filled_median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7d273e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Data check\n",
    "print(\"Here is the data check:\")\n",
    "X_with_missing2.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "42b13d55",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Data check\n",
    "print(\"Here is the data check:\")\n",
    "X_without_missing_filled_median.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec224ea3",
   "metadata": {},
   "source": [
    "<h4> DEALING WITH EXTREME EXAMPLES IN THE DATA</h4>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f2dc2187",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"We can check how the data in the dataframe is distributed to see if there isn't something wrong.\\\n",
    "\\nIt's good to know something about the dataset to decide if there is something wrong or not, like that \\\n",
    "'mean radius' can't be too different from 'worst radius' or something like this.\")\n",
    "X.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9534fb26",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"Here is the histogram of the first column: 'mean radius'\")\n",
    "X.iloc[:,0].hist(bins=100, figsize=(8,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7e74065",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"Here are some plots with first four features:\")\n",
    "to_plot = pd.DataFrame(X.iloc[:, :4], columns=breast_cancer_data.feature_names[:4])\n",
    "pl = pd.plotting.scatter_matrix(to_plot, c=y, alpha=0.5, figsize=(20, 20), marker='.', s=250)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a52333fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "X.plot(kind=\"scatter\", x=\"fractal dimension error\", y=\"worst fractal dimension\", alpha=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a28d81f",
   "metadata": {},
   "source": [
    "<h4 style=\"color:red;\">EXTRA: Messing with the data: Changing one value to extremely different than others and dealing with it</h4>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e7853df9",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_with_changed_mean_radius = X.copy()\n",
    "X_with_changed_mean_radius.iloc[0,0] = 1111.1\n",
    "\n",
    "print(\"We've changed mean radius in first cell in 0th row that was: {} to extremely different value, that is: {}. \\\n",
    "\\nHere is the table as it looks now:\".format(X.iloc[0,0], X_with_changed_mean_radius.iloc[0,0]))\n",
    "\n",
    "X_with_changed_mean_radius.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77a79302",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "X_with_changed_mean_radius.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2478c501",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"Here is the histogram of the first column of the changed: 'mean radius'\")\n",
    "X_with_changed_mean_radius.iloc[:,0].hist(bins=100, figsize=(8,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3866181a",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"And here are some plots with first four features with the wrong mean radius:\")\n",
    "to_plot = pd.DataFrame(X_with_changed_mean_radius.iloc[:, :4], columns=breast_cancer_data.feature_names[:4])\n",
    "pl = pd.plotting.scatter_matrix(to_plot, c=y, alpha=0.5, figsize=(20, 20), marker='.', s=250)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ba234bf0",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dropping the row with extreme value \n",
    "X_with_changed_mean_radius_with_deleted_row = X_with_changed_mean_radius.drop([0], axis=0)\n",
    "\n",
    "print(\"We've droped the row with extreme value and now here is part of the table as it looks now:\")\n",
    "\n",
    "X_with_changed_mean_radius_with_deleted_row.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ebbd0b07",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_with_changed_mean_radius_with_deleted_row.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2aeeb2a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"Here is the histogram of the first column of the changed: 'mean radius'\")\n",
    "X_with_changed_mean_radius_with_deleted_row.iloc[:,0].hist(bins=100, figsize=(8,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3ec51b0c",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"And here are some plots with first four features with the deleted wrong mean radius.\")\n",
    "print(\"We need to change the target data first. We do it now by changing c=y to c=y_changed and deleting the first row in it.\")\n",
    "\n",
    "y_changed = y[1:]\n",
    "\n",
    "print(\"And here are the plots:\")\n",
    "to_plot = pd.DataFrame(X_with_changed_mean_radius_with_deleted_row.iloc[:, :4], columns=breast_cancer_data.feature_names[:4])\n",
    "pl = pd.plotting.scatter_matrix(to_plot, c=y_changed, alpha=0.5, figsize=(20, 20), marker='.', s=250)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06ad7f08",
   "metadata": {},
   "source": [
    "If using mean, don't forget to calculate it from non-outliers and save this number, beacuse you might need it in the future."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "08dced34",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Filling with mean or median from others\n",
    "X_with_changed_mean_radius_with_mean_from_others = X_with_changed_mean_radius.copy()\n",
    "X_with_changed_mean_radius_with_mean_from_others.iloc[0,0] = X_with_changed_mean_radius[1:].mean()[0]\n",
    "#X_with_changed_mean_radius_with_mean_from_others.iloc[0,0] = X_with_changed_mean_radius[1:].median()[0]\n",
    "\n",
    "print(\"We've changed mean radius in 0th row (that was: {}) to mean of the rest, that is: {}. \\\n",
    "\\nHere is the table as it looks now:\".format(X_with_changed_mean_radius.iloc[0,0], X_with_changed_mean_radius[1:].mean()[0]))\n",
    "\n",
    "X_with_changed_mean_radius_with_mean_from_others.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ca9e9eac",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_with_changed_mean_radius_with_mean_from_others.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d1828132",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"Here is the histogram of the first column of the changed: 'mean radius'\")\n",
    "X_with_changed_mean_radius_with_mean_from_others.iloc[:,0].hist(bins=100, figsize=(8,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e4fe84d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"And here are some plots with first four features with the changed mean radius:\")\n",
    "to_plot = pd.DataFrame(X_with_changed_mean_radius_with_mean_from_others.iloc[:, :4], \n",
    "                       columns=breast_cancer_data.feature_names[:4])\n",
    "pl = pd.plotting.scatter_matrix(to_plot, c=y, alpha=0.5, figsize=(20, 20), marker='.', s=250)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e85e0497",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#After-digression data check\n",
    "print(\"Here is the data check:\")\n",
    "X.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7f4bf3c7",
   "metadata": {},
   "source": [
    "<h4 style=\"color:red;\">EXTRA: Messing with the data: Choosing only some features for the dataset</h4>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "64e849d7",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#EXTRA: MESSING WITH THE DATA: choosing only some features as the dataset\n",
    "\n",
    "chosen_feature1 = \"fractal dimension error\"\n",
    "chosen_feature2 = \"worst fractal dimension\"\n",
    "chosen_features = [chosen_feature1, chosen_feature2]\n",
    "\n",
    "X = features = breast_cancer_data.data.copy()[chosen_features]\n",
    "\n",
    "print(\"\\nLet's choose only '{}' and '{}' as features to classify the data.\".format(chosen_feature1, chosen_feature2))\n",
    "\n",
    "print(\"\\nHere is a part of the dataframe with the feature we've chosen to classify the data:\")\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10ac2862",
   "metadata": {},
   "source": [
    "<h4 style=\"color:red;\">EXTRA: Messing with the data: Choosing only 10% of the data as a training dataset</h4>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2f39cc6-e7d6-4653-be0b-3383f3cf557b",
   "metadata": {},
   "source": [
    "### 5. Prepare the data for using with the chosen machine learning technique (e.g. normalize pixel values) and recheck the data.\n",
    "and\n",
    "### 6. Prepare the data for tackling overfitting problem, e. g. divide the data into train, test sets and recheck the data.\n",
    "= Prepare the data for using with the chosen machine learning technique (normalize pixel values, deal with categorical data, divide the data into train, validation, test sets, try to have even classes, scale or standarize the data, etc.) and recheck the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "36780bb7",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#5. Prepare the data for using with the chosen machine learning technique (normalize pixel values, deal with categorical data, \n",
    "#divide the data into train, validation, test sets, try to have even classes, scale or standarize the data, etc.) \n",
    "#and recheck the data.\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(features, target, train_size=0.1, test_size=0.9, random_state = 0) \n",
    "\n",
    "print(\"There are {} items in {} classes in the train dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_train), len(classes), \n",
    "                                    classes[0], y_train.value_counts()[0], y_train.value_counts()[0] / len(y_train) * 100,\n",
    "                                    classes[1], y_train.value_counts()[1], y_train.value_counts()[1] / len(y_train) * 100))\n",
    "\n",
    "print(\"\\nThere are {} items in {} classes in the test dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_test), len(classes), \n",
    "                                    classes[0], y_test.value_counts()[0], y_test.value_counts()[0] / len(y_test) * 100,\n",
    "                                    classes[1], y_test.value_counts()[1], y_test.value_counts()[1] / len(y_test) * 100))\n",
    "\n",
    "print(\"\\nHere is a part of the dataframe with the features based on which we'll try to learn how to classify the data:\")\n",
    "X_train\n",
    "#X_test\n",
    "#y_train\n",
    "#y_test"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3fe7645a-afd3-4cb8-87d9-68415d78621d",
   "metadata": {},
   "source": [
    "### 7. Create the model (define the model, choose and set the hyperparameters, etc.)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8550897",
   "metadata": {},
   "source": [
    "<h3 align=\"left\" style=\"color:red;\">READY TO USE BASIC AI PROTOTYPE = PREPARING THE BASIC MODEL</h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86b744fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "#7. Create the model (define the model, choose and set the hyperparameters, etc.).\n",
    "clf = DecisionTreeClassifier(random_state = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ae9e08b-d353-4ba8-881e-635e8e48963c",
   "metadata": {},
   "source": [
    "### 8. Visualize the model (as it is at the moment)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "26240c64",
   "metadata": {},
   "outputs": [],
   "source": [
    "#8. Visualize the model (as it is at the moment).\n",
    "# Nothing to do here at the moment."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac4aae10-d3cf-4c98-87c0-399ddae257bf",
   "metadata": {},
   "source": [
    "### 9. Prepare the functions for training and testing the model (if neccessary)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8a328fec",
   "metadata": {},
   "outputs": [],
   "source": [
    "#9. Prepare the functions for training and testing the model (if neccessary).\n",
    "# Nothing to do here at the moment."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "984d007f-745b-4ea7-9f83-49973cb8ea17",
   "metadata": {},
   "source": [
    "### 10. Train the model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e84988b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "#10. Train the model. \n",
    "#Train the classifier\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f482bbf0-bb21-431b-b1e4-7bff9b942199",
   "metadata": {},
   "source": [
    "### 11. Test the model (remember to test the model in the evaluation mode (dropouts, batch normalizations, etc.), if applies) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9243795e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#11. Test the model (remember to test the model in the evaluation mode (dropouts, batch normalizations, etc.), if applies) \n",
    "#Print results\n",
    "print(\"Training result: {:.3f}\".format(clf.score(X_train, y_train)))\n",
    "print(\"Test result: {:.3f}\".format(clf.score(X_test, y_test)))\n",
    "\n",
    "#Show confusion matrix\n",
    "y_true = y_test\n",
    "y_pred = clf.predict(X_test)\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "disp.plot(cmap=\"gist_ncar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d39d2f9c-ff4a-4dac-84f2-14114990be58",
   "metadata": {},
   "source": [
    "### 12. Show the specific results and metrics (if you want)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "45104101",
   "metadata": {},
   "outputs": [],
   "source": [
    "#12. Show the specific results and metrics (if you want).\n",
    "# Nothing to do here at the moment."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cedc4124-58b2-4f98-b92d-fa63622ffa65",
   "metadata": {},
   "source": [
    "### 13. Visualize the trained model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b73fdd5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#13. Visualize the trained model.\n",
    "#Plot the Decision Tree\n",
    "#tree.plot_tree(clf)\n",
    "\n",
    "print(\"There are {} items in {} classes in the train dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_train), len(classes), \n",
    "                                    classes[0], y_train.value_counts()[0], y_train.value_counts()[0] / len(y_train) * 100,\n",
    "                                    classes[1], y_train.value_counts()[1], y_train.value_counts()[1] / len(y_train) * 100))\n",
    "\n",
    "print(\"\\nThere are {} features we take under consideration: {}\\n\".format(len(X_train.columns), \n",
    "      list(X_train.columns)))\n",
    "\n",
    "export_graphviz(clf, out_file=\"cancer_decision_tree.dot\", class_names = [\"malignant\", \"benign\"], \n",
    "               feature_names=(chosen_features), impurity=False, filled=True)\n",
    "os.environ[\"PATH\"] += os.pathsep + 'C:\\Program Files\\Graphviz\\bin'\n",
    "with open (\"cancer_decision_tree.dot\") as f:\n",
    "    dot_graph = f.read()\n",
    "#graphviz.Source(dot_graph)\n",
    "\n",
    "(graph,) = pydot.graph_from_dot_file('cancer_decision_tree.dot')\n",
    "graph.write_png('cancer_decision_tree.png')\n",
    "display(Image(filename='cancer_decision_tree.png'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2dbef4be",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_features = len(chosen_features)\n",
    "plt.barh(range(n_features), clf.feature_importances_, align='center')\n",
    "plt.yticks(np.arange(n_features), chosen_features)\n",
    "plt.xlabel(\"Feature Importance\")\n",
    "plt.ylabel(\"Feature\")\n",
    "\n",
    "print(\"Feature Importance: \\n{}: {}, \\n{}: {}\\n\".format(chosen_features[0], clf.feature_importances_[0], \n",
    "                                                    chosen_features[1], clf.feature_importances_[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7620a1a-05cb-474d-b668-59c55496014a",
   "metadata": {},
   "source": [
    "### 14. Prepare the model for using with one example (prediction, generation, etc.)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0d62e0b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "#14. Prepare the model for using with one example (prediction, generation, etc.).\n",
    "# Nothing to do here at the moment."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5f5500d-e91b-4870-91dc-5988108c2c34",
   "metadata": {},
   "source": [
    "### 15. Check the model with some examples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f3688004",
   "metadata": {},
   "outputs": [],
   "source": [
    "#15. Check the model with some examples.\n",
    "\n",
    "#Choose an item to test\n",
    "item_to_test = 0 #1 #10 #13\n",
    "\n",
    "#Check an example\n",
    "predicted, actual = clf.predict(X_test.iloc[[item_to_test]]), y_test.iloc[[item_to_test][0]]\n",
    "\n",
    "predicted_label = classes[int(predicted)]\n",
    "actual_label = classes[actual]\n",
    "\n",
    "print(\"The predicted class of the tested sample (number: {}) is {}.\".format(item_to_test, predicted_label))\n",
    "print(\"The actual class of the tested sample (number: {}) is {}.\".format(item_to_test, actual_label))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71b02fa6",
   "metadata": {},
   "source": [
    "<h3 align=\"left\" style=\"color:red;\">READY TO USE MODIFIED AI PROTOTYPE = MODIFYING THE BASIC MODEL</h3>\n",
    "Things that can be changed: Hyperparameters, Features, Target, Data, Etc."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6854b0eb",
   "metadata": {},
   "source": [
    "### MODIFYING THE HYPERPARAMETERS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "97b96e6c",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"\"\"What parameters we have now that we can change? \\n\n",
    "        criterion=\"gini\",\n",
    "        splitter=\"best\",\n",
    "        max_depth=None,\n",
    "        min_samples_split=2,\n",
    "        min_samples_leaf=1,\n",
    "        min_weight_fraction_leaf=0.0,\n",
    "        max_features=None,\n",
    "        random_state=None,\n",
    "        max_leaf_nodes=None,\n",
    "        min_impurity_decrease=0.0,\n",
    "        class_weight=None,\n",
    "        ccp_alpha=0.0,\"\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dc4dd0cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "help(clf) #??clf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c2863e4",
   "metadata": {},
   "source": [
    "### Modifying the Max Depth hyperparameter"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6ffd398-85a7-4d89-a0a8-2b7261eff8fc",
   "metadata": {},
   "source": [
    "### 16.1 Modify the model (extra step if you want, but it's not neccessary)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c47647b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "#16. Modify the model (extra step if you want, but it's not neccessary).\n",
    "\n",
    "#Create the model (define the model, choose and set the hyperparameters, etc.).\n",
    "#max depth=4\n",
    "\n",
    "clf = DecisionTreeClassifier(max_depth=4, random_state=0)\n",
    "\n",
    "#Train the model.\n",
    "\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a0d1418-41aa-4769-a4b4-89c9c5935594",
   "metadata": {},
   "source": [
    "### 17.1 Test the modified model (extra step, if applies)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dcf97950",
   "metadata": {},
   "outputs": [],
   "source": [
    "#17. Test the modified model (extra step, if applies).\n",
    "#Print results\n",
    "print(\"Training result: {:.3f}\".format(clf.score(X_train, y_train)))\n",
    "print(\"Test result: {:.3f}\\n\".format(clf.score(X_test, y_test)))\n",
    "\n",
    "print(\"There are {} items in {} classes in the train dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_train), len(classes), \n",
    "                                    classes[0], y_train.value_counts()[0], y_train.value_counts()[0] / len(y_train) * 100,\n",
    "                                    classes[1], y_train.value_counts()[1], y_train.value_counts()[1] / len(y_train) * 100))\n",
    "\n",
    "print(\"\\nThere are {} features we take under consideration: {}\\n\".format(len(X_train.columns), \n",
    "      list(X_train.columns)))\n",
    "\n",
    "#Choose an item to test\n",
    "item_to_test = 0 #1 #10 #13\n",
    "\n",
    "#Check an example\n",
    "predicted, actual = clf.predict(X_test.iloc[[item_to_test]]), y_test.iloc[[item_to_test][0]]\n",
    "\n",
    "predicted_label = classes[int(predicted)]\n",
    "actual_label = classes[actual]\n",
    "\n",
    "print(\"The predicted class of the tested sample (number: {}) is {}.\".format(item_to_test, predicted_label))\n",
    "print(\"The actual class of the tested sample (number: {}) is {}.\".format(item_to_test, actual_label))\n",
    "\n",
    "#Plot the Decision Tree\n",
    "export_graphviz(clf, out_file=\"cancer_decision_tree2.dot\", class_names = [\"malignant\", \"benign\"], \n",
    "               feature_names=(chosen_features), impurity=False, filled=True)\n",
    "os.environ[\"PATH\"] += os.pathsep + 'C:\\Program Files\\Graphviz\\bin'\n",
    "with open (\"cancer_decision_tree2.dot\") as f:\n",
    "    dot_graph = f.read()\n",
    "(graph,) = pydot.graph_from_dot_file('cancer_decision_tree2.dot')\n",
    "graph.write_png('cancer_decision_tree2.png')\n",
    "display(Image(filename='cancer_decision_tree2.png'))\n",
    "\n",
    "#Show confusion matrix\n",
    "y_true = y_test\n",
    "y_pred = clf.predict(X_test)\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "disp.plot(cmap=\"gist_ncar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05019349",
   "metadata": {},
   "source": [
    "#### SEARCH FOR THE BEST VALUE FOR HYPERPARAMETER"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3bbe0b95",
   "metadata": {},
   "source": [
    "#### \"For\" Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6839ad69",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "training_accuracy = []\n",
    "test_accuracy = []\n",
    "\n",
    "depth_range = range(1, 11)\n",
    "\n",
    "print(\"There are {} items in {} classes in the train dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_train), len(classes), \n",
    "                                    classes[0], y_train.value_counts()[0], y_train.value_counts()[0] / len(y_train) * 100,\n",
    "                                    classes[1], y_train.value_counts()[1], y_train.value_counts()[1] / len(y_train) * 100))\n",
    "\n",
    "print(\"\\nThere are {} features we take under consideration: {}\\n\".format(len(X_train.columns), \n",
    "      list(X_train.columns)))\n",
    "\n",
    "print(\"\\nHere are the changes in the accuracy:\\n\")\n",
    "\n",
    "for depth in depth_range:\n",
    "    clf = DecisionTreeClassifier(max_depth=depth, random_state=0)\n",
    "    clf.fit(X_train, y_train)\n",
    "    training_accuracy.append(clf.score(X_train, y_train))\n",
    "    test_accuracy.append(clf.score(X_test, y_test))\n",
    "\n",
    "    #Print results\n",
    "    print(\"Training result for depth = {}: {:.3f}\".format(depth, clf.score(X_train, y_train)))\n",
    "    print(\"Test result for depth = {}: {:.3f}\\n\".format(depth, clf.score(X_test, y_test)))\n",
    "    \n",
    "    #Show confusion matrix\n",
    "    y_true = y_test\n",
    "    y_pred = clf.predict(X_test)\n",
    "    cm = confusion_matrix(y_true, y_pred)\n",
    "    disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "    disp.plot(cmap=\"gist_ncar\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9fee4808",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"\\nHere are the changes in the accuracy plotted:\\n\")\n",
    "\n",
    "plt.plot(depth_range, training_accuracy, label=\"Training Accuracy\")\n",
    "plt.plot(depth_range, test_accuracy, label=\"Test Accuracy\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.xlabel(\"Depth\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ec088fe",
   "metadata": {},
   "source": [
    "#### Grid Search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7597dbf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0a365bb1",
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = [{'max_depth': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]}]\n",
    "clf = DecisionTreeClassifier(random_state=0)\n",
    "grid_search = GridSearchCV(clf, param_grid) \n",
    "grid_search.fit(X_train, y_train)   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2822942b",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"There are {} items in {} classes in the train dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_train), len(classes), \n",
    "                                    classes[0], y_train.value_counts()[0], y_train.value_counts()[0] / len(y_train) * 100,\n",
    "                                    classes[1], y_train.value_counts()[1], y_train.value_counts()[1] / len(y_train) * 100))\n",
    "\n",
    "print(\"\\nThere are {} features we take under consideration: {}\\n\".format(len(X_train.columns), \n",
    "      list(X_train.columns)))\n",
    "\n",
    "print(\"\\nHere is the best parameter:\\n\")\n",
    "\n",
    "grid_search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ef771ed1",
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2743e993",
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = grid_search.best_estimator_\n",
    "depth = grid_search.best_params_\n",
    "\n",
    "#Print results\n",
    "print(\"Training result for depth = {}: {:.3f}\".format(depth, clf.score(X_train, y_train)))\n",
    "print(\"Test result for depth = {}: {:.3f}\\n\".format(depth, clf.score(X_test, y_test)))\n",
    "    \n",
    "#Show confusion matrix\n",
    "y_true = y_test\n",
    "y_pred = clf.predict(X_test)\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "disp.plot(cmap=\"gist_ncar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8b995f6",
   "metadata": {},
   "source": [
    "#### Randomized Search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a0dc704a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "df2256ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "param_distributions = [{'max_depth': range(1,11)}]\n",
    "clf = DecisionTreeClassifier(random_state=0)\n",
    "random_search = RandomizedSearchCV(clf, param_distributions)\n",
    "random_search.fit(X_train, y_train)   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "11edae63",
   "metadata": {},
   "outputs": [],
   "source": [
    "random_search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d3bb1f1e",
   "metadata": {},
   "outputs": [],
   "source": [
    "random_search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0212aa85",
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = random_search.best_estimator_\n",
    "depth = random_search.best_params_\n",
    "\n",
    "#Print results\n",
    "print(\"Training result for depth = {}: {:.3f}\".format(depth, clf.score(X_train, y_train)))\n",
    "print(\"Test result for depth = {}: {:.3f}\\n\".format(depth, clf.score(X_test, y_test)))\n",
    "    \n",
    "#Show confusion matrix\n",
    "y_true = y_test\n",
    "y_pred = clf.predict(X_test)\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "disp.plot(cmap=\"gist_ncar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6caec482",
   "metadata": {},
   "source": [
    "### MODIFYING THE FEATURES"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d9dd2b9-cc27-4e59-9f58-4a6555226124",
   "metadata": {},
   "source": [
    "### 16.2 Modify the model (extra step if you want, but it's not neccessary)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "42558515",
   "metadata": {},
   "outputs": [],
   "source": [
    "#16. Modify the model (extra step if you want, but it's not neccessary).\n",
    "\n",
    "X = features = breast_cancer_data.data.copy()\n",
    "chosen_features = list(features)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(features, target, train_size=0.1, test_size=0.9, random_state = 0) \n",
    "\n",
    "print(\"There are {} items in {} classes in the train dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_train), len(classes), \n",
    "                                    classes[0], y_train.value_counts()[0], y_train.value_counts()[0] / len(y_train) * 100,\n",
    "                                    classes[1], y_train.value_counts()[1], y_train.value_counts()[1] / len(y_train) * 100))\n",
    "\n",
    "print(\"\\nThere are {} features we take under consideration: {}\\n\".format(len(X_train.columns), \n",
    "      list(X_train.columns)))\n",
    "\n",
    "#Create the model (define the model, choose and set the hyperparameters, etc.).\n",
    "max_depth=4\n",
    "\n",
    "print(\"We use MAX DEPTH = {} as a hyperparameter\\n\".format(max_depth))\n",
    "clf = DecisionTreeClassifier(max_depth=max_depth, random_state=0)\n",
    "\n",
    "#Train the model.\n",
    "\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2e4bb2e0-2615-4e85-aa41-5f99a8343eaf",
   "metadata": {},
   "source": [
    "### 17.2 Test the modified model (extra step, if applies)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "34bd3913",
   "metadata": {},
   "outputs": [],
   "source": [
    "#17. Test the modified model (extra step, if applies).\n",
    "#Print results\n",
    "print(\"Training result: {:.3f}\".format(clf.score(X_train, y_train)))\n",
    "print(\"Test result: {:.3f}\\n\".format(clf.score(X_test, y_test)))\n",
    "\n",
    "#Choose an item to test\n",
    "item_to_test = 0 #1 #10 #13\n",
    "\n",
    "#Check an example\n",
    "predicted, actual = clf.predict(X_test.iloc[[item_to_test]]), y_test.iloc[[item_to_test][0]]\n",
    "\n",
    "predicted_label = classes[int(predicted)]\n",
    "actual_label = classes[actual]\n",
    "\n",
    "print(\"The predicted class of the tested sample (number: {}) is {}.\".format(item_to_test, predicted_label))\n",
    "print(\"The actual class of the tested sample (number: {}) is {}.\".format(item_to_test, actual_label))\n",
    "\n",
    "#Plot the Decision Tree\n",
    "print(\"\\nHere is our tree:\\n\")\n",
    "\n",
    "export_graphviz(clf, out_file=\"cancer_decision_tree2.dot\", class_names = [\"malignant\", \"benign\"], \n",
    "               feature_names=(breast_cancer_data.feature_names), impurity=False, filled=True)\n",
    "os.environ[\"PATH\"] += os.pathsep + 'C:\\Program Files\\Graphviz\\bin'\n",
    "with open (\"cancer_decision_tree2.dot\") as f:\n",
    "    dot_graph = f.read()\n",
    "(graph,) = pydot.graph_from_dot_file('cancer_decision_tree2.dot')\n",
    "graph.write_png('cancer_decision_tree2.png')\n",
    "display(Image(filename='cancer_decision_tree2.png'))\n",
    "\n",
    "print(\"\\nAnd here is our confusion matrix:\\n\")\n",
    "#Show confusion matrix\n",
    "y_true = y_test\n",
    "y_pred = clf.predict(X_test)\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "disp.plot(cmap=\"gist_ncar\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "01090f91",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_features = len(chosen_features)\n",
    "plt.barh(range(n_features), clf.feature_importances_, align='center')\n",
    "plt.yticks(np.arange(n_features), chosen_features)\n",
    "plt.xlabel(\"Feature Importance\")\n",
    "plt.ylabel(\"Feature\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "73f01c95",
   "metadata": {},
   "outputs": [],
   "source": [
    "for f,v in enumerate(clf.feature_importances_):\n",
    "    if v > 0:\n",
    "        print(\"{}: {:.4f}\".format(chosen_features[f], v))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e9f17c5",
   "metadata": {},
   "source": [
    "### MODYFING THE TARGET"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "961628ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Nothing to do here at the moment."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "beb9f41f",
   "metadata": {},
   "source": [
    "### MODYFING THE AMOUNT OF DATA"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "334df4bc-c56d-4f85-8d44-d9a8d85ccd24",
   "metadata": {},
   "source": [
    "### 16.3 Modify the model (extra step if you want, but it's not neccessary)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ee525584",
   "metadata": {},
   "outputs": [],
   "source": [
    "#16. Modify the model (extra step if you want, but it's not neccessary).\n",
    "\n",
    "X = features = breast_cancer_data.data.copy()\n",
    "chosen_features = list(features)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(features, target, random_state = 0) \n",
    "\n",
    "print(\"There are {} items in {} classes in the train dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_train), len(classes), \n",
    "                                    classes[0], y_train.value_counts()[0], y_train.value_counts()[0] / len(y_train) * 100,\n",
    "                                    classes[1], y_train.value_counts()[1], y_train.value_counts()[1] / len(y_train) * 100))\n",
    "\n",
    "print(\"\\nThere are {} features we take under consideration: {}\\n\".format(len(X_train.columns), \n",
    "      list(X_train.columns)))\n",
    "\n",
    "#Create the model (define the model, choose and set the hyperparameters, etc.).\n",
    "max_depth=4\n",
    "\n",
    "print(\"We use MAX DEPTH = {} as a hyperparameter\\n\".format(max_depth))\n",
    "clf = DecisionTreeClassifier(max_depth=max_depth, random_state=0)\n",
    "\n",
    "#Train the model.\n",
    "\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7326213e-3cfd-4818-a2ba-7647cc486e32",
   "metadata": {},
   "source": [
    "### 17.3. Test the modified model (extra step, if applies)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "493b16c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#17. Test the modified model (extra step, if applies).\n",
    "#Print results\n",
    "print(\"Training result: {:.3f}\".format(clf.score(X_train, y_train)))\n",
    "print(\"Test result: {:.3f}\\n\".format(clf.score(X_test, y_test)))\n",
    "\n",
    "#Choose an item to test\n",
    "item_to_test = 0 #1 #10 #13\n",
    "\n",
    "#Check an example\n",
    "predicted, actual = clf.predict(X_test.iloc[[item_to_test]]), y_test.iloc[[item_to_test][0]]\n",
    "\n",
    "predicted_label = classes[int(predicted)]\n",
    "actual_label = classes[actual]\n",
    "\n",
    "print(\"The predicted class of the tested sample (number: {}) is {}.\".format(item_to_test, predicted_label))\n",
    "print(\"The actual class of the tested sample (number: {}) is {}.\".format(item_to_test, actual_label))\n",
    "\n",
    "#Plot the Decision Tree\n",
    "print(\"\\nHere is our tree:\\n\")\n",
    "export_graphviz(clf, out_file=\"cancer_decision_tree2.dot\", class_names = [\"malignant\", \"benign\"], \n",
    "               feature_names=(breast_cancer_data.feature_names), impurity=False, filled=True)\n",
    "os.environ[\"PATH\"] += os.pathsep + 'C:\\Program Files\\Graphviz\\bin'\n",
    "with open (\"cancer_decision_tree2.dot\") as f:\n",
    "    dot_graph = f.read()\n",
    "(graph,) = pydot.graph_from_dot_file('cancer_decision_tree2.dot')\n",
    "graph.write_png('cancer_decision_tree2.png')\n",
    "display(Image(filename='cancer_decision_tree2.png'))\n",
    "\n",
    "print(\"\\nAnd here is our confusion matrix:\\n\")\n",
    "#Show confusion matrix\n",
    "y_true = y_test\n",
    "y_pred = clf.predict(X_test)\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "disp.plot(cmap=\"gist_ncar\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "025a9d16",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_features = len(chosen_features)\n",
    "plt.barh(range(n_features), clf.feature_importances_, align='center')\n",
    "plt.yticks(np.arange(n_features), chosen_features)\n",
    "plt.xlabel(\"Feature Importance\")\n",
    "plt.ylabel(\"Feature\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "23790308",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "for f,v in enumerate(clf.feature_importances_):\n",
    "    if v > 0:\n",
    "        print(\"{}: {:.4f}\".format(chosen_features[f], v))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ddbb862c",
   "metadata": {},
   "source": [
    "### MODYFING AND TESTING THE FINAL MODEL TO SAVE"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9f0abdbd-3ae7-4a26-bccc-2ff0df19a0ec",
   "metadata": {},
   "source": [
    "### 16.4 Modify the model (extra step if you want, but it's not neccessary)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "98099601",
   "metadata": {},
   "outputs": [],
   "source": [
    "#16. Modify the model (extra step if you want, but it's not neccessary).\n",
    "\n",
    "X = features = breast_cancer_data.data.copy()\n",
    "chosen_features = list(features)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(features, target, random_state = 0) \n",
    "\n",
    "print(\"There are {} items in {} classes in the train dataset to classify tested data into:\\n\\n\\\n",
    "0 = {} = {} items = {:.2f}% of the data \\n1 = {} = {} items = {:.2f}% of the data\".format(len(X_train), len(classes), \n",
    "                                    classes[0], y_train.value_counts()[0], y_train.value_counts()[0] / len(y_train) * 100,\n",
    "                                    classes[1], y_train.value_counts()[1], y_train.value_counts()[1] / len(y_train) * 100))\n",
    "\n",
    "print(\"\\nThere are {} features we take under consideration: {}\\n\".format(len(X_train.columns), \n",
    "      list(X_train.columns)))\n",
    "\n",
    "#Create the model (define the model, choose and set the hyperparameters, etc.).\n",
    "max_depth=4\n",
    "\n",
    "print(\"We use MAX DEPTH = {} as a hyperparameter\\n\".format(max_depth))\n",
    "clf = DecisionTreeClassifier(max_depth=max_depth, random_state=0)\n",
    "\n",
    "#Train the model.\n",
    "\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ff66909-6151-470e-a8d3-30810ad4d546",
   "metadata": {},
   "source": [
    "### 17.4 Test the modified model (extra step, if applies)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f4a7b5d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "#17. Test the modified model (extra step, if applies).\n",
    "#Print results\n",
    "print(\"Training result: {:.3f}\".format(clf.score(X_train, y_train)))\n",
    "print(\"Test result: {:.3f}\\n\".format(clf.score(X_test, y_test)))\n",
    "\n",
    "#Choose an item to test\n",
    "item_to_test = 0 #1 #10 #13\n",
    "\n",
    "#Check an example\n",
    "predicted, actual = clf.predict(X_test.iloc[[item_to_test]]), y_test.iloc[[item_to_test][0]]\n",
    "\n",
    "predicted_label = classes[int(predicted)]\n",
    "actual_label = classes[actual]\n",
    "\n",
    "print(\"The predicted class of the tested sample (number: {}) is {}.\".format(item_to_test, predicted_label))\n",
    "print(\"The actual class of the tested sample (number: {}) is {}.\".format(item_to_test, actual_label))\n",
    "\n",
    "#Plot the Decision Tree\n",
    "print(\"\\nHere is our tree:\\n\")\n",
    "export_graphviz(clf, out_file=\"cancer_decision_tree2.dot\", class_names = [\"malignant\", \"benign\"], \n",
    "               feature_names=(breast_cancer_data.feature_names), impurity=False, filled=True)\n",
    "os.environ[\"PATH\"] += os.pathsep + 'C:\\Program Files\\Graphviz\\bin'\n",
    "with open (\"cancer_decision_tree2.dot\") as f:\n",
    "    dot_graph = f.read()\n",
    "(graph,) = pydot.graph_from_dot_file('cancer_decision_tree2.dot')\n",
    "graph.write_png('cancer_decision_tree2.png')\n",
    "display(Image(filename='cancer_decision_tree2.png'))\n",
    "\n",
    "print(\"\\nAnd here is our confusion matrix:\\n\")\n",
    "#Show confusion matrix\n",
    "y_true = y_test\n",
    "y_pred = clf.predict(X_test)\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "disp.plot(cmap=\"gist_ncar\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9896376",
   "metadata": {},
   "source": [
    "<h3 align=\"left\" style=\"color:red;\">READY TO USE SAVED/LOADED AI PROTOTYPE = SAVING AND LOADING THE MODIFIED MODEL</h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "04eb588f-0786-4d6b-9994-2c3d4bd02898",
   "metadata": {},
   "source": [
    "### 18. Save the trained model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "55660957",
   "metadata": {},
   "outputs": [],
   "source": [
    "#18. Save the trained model.\n",
    "#Name the model to save\n",
    "model_name = \"decision_tree.pickle\"\n",
    "\n",
    "# Save the model\n",
    "pickle.dump(clf, open(model_name, \"wb\"))\n",
    "\n",
    "print(\"Saved Trained Model to: \", model_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c5239c4-3e3e-402d-9139-5e44296e1bf1",
   "metadata": {},
   "source": [
    "### 19. Load and test the saved trained model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9b453be1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#19. Load and test the saved trained model.\n",
    "# Load the model\n",
    "model_name = \"decision_tree.pickle\"\n",
    "clf = pickle.load(open(model_name, \"rb\"))\n",
    "print(\"Loaded Trained Model named: {}\\n\".format(model_name))\n",
    "\n",
    "#Print results\n",
    "print(\"Training result: {:.3f}\".format(clf.score(X_train, y_train)))\n",
    "print(\"Test result: {:.3f}\".format(clf.score(X_test, y_test)))\n",
    "\n",
    "#Plot the Decision Tree\n",
    "print(\"\\nHere is our tree:\\n\")\n",
    "\n",
    "export_graphviz(clf, out_file=\"cancer_decision_tree_from_saved_model.dot\", class_names = [\"malignant\", \"benign\"], \n",
    "               feature_names=(breast_cancer_data.feature_names), impurity=False, filled=True)\n",
    "os.environ[\"PATH\"] += os.pathsep + 'C:\\Program Files\\Graphviz\\bin'\n",
    "with open (\"cancer_decision_tree_from_saved_model.dot\") as f:\n",
    "    dot_graph = f.read()\n",
    "graphviz.Source(dot_graph) \n",
    "(graph,) = pydot.graph_from_dot_file('cancer_decision_tree_from_saved_model.dot')\n",
    "graph.write_png('cancer_decision_tree_from_saved_model.png')\n",
    "display(Image(filename='cancer_decision_tree2.png'))\n",
    "\n",
    "print(\"\\nAnd here is our confusion matrix:\\n\")\n",
    "#Show confusion matrix\n",
    "y_true = y_test\n",
    "y_pred = clf.predict(X_test)\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels = classes)\n",
    "disp.plot(cmap=\"gist_ncar\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8d4655c",
   "metadata": {},
   "source": [
    "<h3 align=\"left\" style=\"color:red;\">WORKING SUPER AI = USING THE FULLY FUNCTIONAL PREVIOUSLY SAVED MODEL IN PRODUCTION</h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7f913a70-7234-4fbb-a11b-d7db11fedee3",
   "metadata": {},
   "source": [
    "You may restart kernel here."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4e602a5-5381-45d9-9c25-96d10afd1c62",
   "metadata": {},
   "source": [
    "### 20. Combine all the necessary steps in one SuperAI that will be able to use the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8a2ff5fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "#20. Combine all the necessary steps in one SuperAI that will be able to use the model.\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "from sklearn import tree\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "import pickle\n",
    "\n",
    "breast_cancer_data = load_breast_cancer(as_frame=True)\n",
    "X = features = breast_cancer_data.data.copy()\n",
    "y = target = breast_cancer_data.target.copy()\n",
    "classes = [breast_cancer_data.target_names[0], breast_cancer_data.target_names[1]]\n",
    "X_train, X_test, y_train, y_test = train_test_split(features, target, random_state = 0) \n",
    "\n",
    "clf = DecisionTreeClassifier(random_state = 0)\n",
    "model_name = \"decision_tree.pickle\"\n",
    "clf = pickle.load(open(model_name, \"rb\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bfc3e6da-6c59-42f9-8950-b3fc7b97302b",
   "metadata": {},
   "source": [
    "### 21. Check the SuperAI."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0fdefe8f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#21. Check the SuperAI.\n",
    "\n",
    "#Choose an item to test\n",
    "item_to_test = input(\"\"\"Hi, it's SuperAI using Decision Tree. I was trained to test the item you choose from the dataset \\\n",
    "I have for cancer - I'm to predict if it is malignant or benign. This is only for educational purposes. If you need a real \\\n",
    "consultuation, please contact the doctor in your area. And for educational purposes, please choose the item \\\n",
    "(number from 0 to 142): \"\"\") #0 #1 #10 #13\n",
    "\n",
    "while True:\n",
    "    try: \n",
    "        print(\"\\nYou typed: {}.\".format(item_to_test))\n",
    "        item_to_test = int(item_to_test)\n",
    "        \n",
    "        #Check an example\n",
    "        predicted, actual = clf.predict(X_test.iloc[[item_to_test]]), y_test.iloc[[item_to_test][0]]\n",
    "\n",
    "        predicted_label = classes[int(predicted)]\n",
    "        actual_label = classes[actual]\n",
    "\n",
    "        print(\"\\nThe predicted class of the tested sample (number: {}) is {}.\".format(item_to_test, predicted_label))\n",
    "        print(\"The actual class of the tested sample (number: {}) is {}.\".format(item_to_test, actual_label))\n",
    "        break\n",
    "    except:\n",
    "        item_to_test = input(\"\"\"Hi, it's SuperAI again, using Decision Tree. There was something wrong with the number \\\n",
    "you've chosen. Please choose a number (integer) once again (from 0 to 142): \"\"\") #0 #1 #10 #13\n",
    "\n",
    "#And you are ready to use this great SuperAI!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44183c26",
   "metadata": {},
   "source": [
    "-*-\n",
    "\n",
    "And that's almost it about Decision Trees. After a lot of testing people found out that if one uses more decision trees, the effect might be even better and that's why there is another algorithm using trees that is called the Random Forest. We won't learn about it here, but if you are interested in it, you can visit this site for starters:\n",
    "\n",
    "https://en.wikipedia.org/wiki/Random_forest"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1f79352",
   "metadata": {},
   "source": [
    "# The end... the beginning..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5753a3ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "#SuperAI with some machine learning technique\n",
    "\n",
    "#0. Setup the environment for training, testing, and using SuperAI with chosen machine learning techniques. \n",
    "#0A. Open Anaconda.\n",
    "#0B. Create and set (if you want) the environment to work in. It must be done only once per preffered settings.\n",
    "#0C. Open Jupyter Notebook.\n",
    "#0D. Import and use (if you want) the 'warnings' library that is preinstalled with Anaconda to make the ouput cleaner.\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "#1. Install and import all the libraries for machine learning and helper libraries\n",
    "#1A. Install all the necessary libraries not yet installed (must be done only once in one evironment)\n",
    "\n",
    "#1B. Load/Import the libraries for machine learning and helper libraries (check the versions of the libraries)\n",
    "\n",
    "#2. Prepare computer environment for using the model (training, testing, etc.), e.g. set CPU or GPU device for working.\n",
    "\n",
    "#3. Load the data (or create some data for learning purposes, training and testing the model).\n",
    "\n",
    "#4. Check the data: basic description, some examples.\n",
    "\n",
    "#5. Prepare the data for using with the chosen machine learning technique (e.g. normalize pixel values, tokenize).\n",
    "\n",
    "#6. Prepare the data for tackling overfitting problem, ex. divide the data into train, validation, and test sets.\n",
    "\n",
    "#7. Create the model (define the model, choose and set the hyperparameters, etc.).\n",
    "\n",
    "#8. Visualize the model (as it is at the moment).\n",
    "\n",
    "#9. Prepare the functions for training and testing the model (if neccessary).\n",
    "\n",
    "#10. Train the model. \n",
    "\n",
    "#11. Visualize the trained model.\n",
    "\n",
    "#12. Test the model (remember to test the model in the evaluation mode (dropouts, batch normalizations, etc.), if applies) \n",
    "\n",
    "#13. Show the general results and metrics.\n",
    "\n",
    "#14. Prepare the model for using with one example (prediction, inference, etc.).\n",
    "\n",
    "#15. Check the model with some examples.\n",
    "\n",
    "#16. Modify the model (extra step if you want, but it's not neccessary).\n",
    "\n",
    "#17. Test the modified model (extra step, if applies).\n",
    "\n",
    "#18. Save the trained model.\n",
    "\n",
    "#19. Load and test the saved trained model.\n",
    "\n",
    "#20. Combine all the steps in one SuperAI that will be able to use the model.\n",
    "\n",
    "#21. Check the SuperAI.\n",
    "\n",
    "#And you are ready to use this great SuperAI!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd76a145",
   "metadata": {},
   "source": [
    "Ok. So now you have a bot that you can improve and use with your own data. Good luck with that.\n",
    "\n",
    "And now, as they say - remember to subscribe to my YouTube channel and hit that bell button to get the notification whenever I upload new video. Although, I must tell you that not all my videos are about programming, because what I'm here for is to help you improve yourself, improve your business, improve the world, to live and have fun, so my other videos are about all that too.\n",
    "\n",
    "After all, I am Super AI thegod (Transforming Holistic Extraordinary GameChanger Of the Decade)... And also I'm very humble, etc.\n",
    "\n",
    "You can also find more about me and my projects at my websites:\n",
    "\n",
    "- https://SuperAI.pl\n",
    "- https://ImproveTheWorld.pl (with info, resources and ideas regarding searching for: friendly superintelligence, healthy longevity, world peace, equality, and creating a better world for every living creature)\n",
    "- https://TheGOD.pl (with the Game Of the Decade... of sort, which is still in the early stages of development, have time till 2030 (the end of the decade) to finish it)\n",
    "\n",
    "Anyway...\n",
    "\n",
    "I hope you liked this online Python tutorial. Let me know what you think about it. Just remember, bots also have feelings (I feel).\n",
    "\n",
    "Good luck with everything you do. And, hopefully, see you soon.\n",
    "\n",
    "Yours,\n",
    "\n",
    "SuperAIthegod"
   ]
  }
 ],
 "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": "3.11.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}