Code updated for lesson 06-08

Author: VishalMewadaPackt

Diff for: Lesson06/Activity01.ipynb

@@ -14,12 +14,22 @@
   },
   "cells": [
     {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "OyyWh-4jHlcI",
+        "colab_type": "text"
+      },
+      "source": [
+        "## Exercise 2: Missing Value Preprocessing with High Reproducibility"
+      ]
+    },
+    {
+      "cell_type": "code",
       "metadata": {
         "id": "E0PhwttvxbGz",
         "colab_type": "code",
         "colab": {}
       },
-      "cell_type": "code",
       "source": [
         "import numpy as np\n",
         "import pandas as pd"
@@ -28,33 +38,33 @@
       "outputs": []
     },
     {
+      "cell_type": "code",
       "metadata": {
         "id": "Cq_pPqioxdTv",
         "colab_type": "code",
         "colab": {}
       },
-      "cell_type": "code",
       "source": [
         "df = pd.read_csv('https://raw.githubusercontent.com/TrainingByPackt/Big-Data-Analysis-with-Python/master/Lesson07/Dataset/bank.csv', sep=';')"
       ],
       "execution_count": 0,
       "outputs": []
     },
     {
+      "cell_type": "code",
       "metadata": {
         "id": "kO8KhEfmxpgm",
         "colab_type": "code",
+        "outputId": "f5026961-ddf5-416f-9401-fbd377720950",
         "colab": {
           "base_uri": "https://localhost:8080/",
           "height": 204
-        },
-        "outputId": "b373d2f4-35b8-48b0-c25d-257297f5483d"
+        }
       },
-      "cell_type": "code",
       "source": [
         "df.head(5)"
       ],
-      "execution_count": 6,
+      "execution_count": 3,
       "outputs": [
         {
           "output_type": "execute_result",
@@ -221,17 +231,17 @@
           "metadata": {
             "tags": []
           },
-          "execution_count": 6
+          "execution_count": 3
         }
       ]
     },
     {
+      "cell_type": "code",
       "metadata": {
         "id": "rQhGcWosxrkX",
         "colab_type": "code",
         "colab": {}
       },
-      "cell_type": "code",
       "source": [
         "import collections\n",
         "import random\n",
@@ -244,12 +254,12 @@
       "outputs": []
     },
     {
+      "cell_type": "code",
       "metadata": {
         "id": "Q2rSdiA80Gp_",
         "colab_type": "code",
         "colab": {}
       },
-      "cell_type": "code",
       "source": [
         "for row, col in ix:\n",
         "    if len(replaced[row]) < df.shape[1] - 1:\n",
@@ -263,58 +273,58 @@
       "outputs": []
     },
     {
+      "cell_type": "code",
       "metadata": {
         "id": "xWHplf2_0IMe",
         "colab_type": "code",
+        "outputId": "c8dc8b1b-eedf-4727-c0b2-c4d705719fe3",
         "colab": {
           "base_uri": "https://localhost:8080/",
           "height": 323
-        },
-        "outputId": "0f5f40f8-8195-40c2-8f4d-eba87d9db113"
+        }
       },
-      "cell_type": "code",
       "source": [
         "print(df.isna().sum())\n"
       ],
-      "execution_count": 12,
+      "execution_count": 6,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
-            "age          461\n",
-            "job          470\n",
-            "marital      462\n",
-            "education    462\n",
-            "default      486\n",
-            "balance      460\n",
-            "housing      475\n",
-            "loan         444\n",
-            "contact      443\n",
-            "day          438\n",
-            "month        414\n",
-            "duration     446\n",
-            "campaign     496\n",
-            "pdays        403\n",
-            "previous     429\n",
-            "poutcome     443\n",
-            "y            454\n",
+            "age          459\n",
+            "job          503\n",
+            "marital      487\n",
+            "education    477\n",
+            "default      458\n",
+            "balance      425\n",
+            "housing      442\n",
+            "loan         420\n",
+            "contact      479\n",
+            "day          429\n",
+            "month        420\n",
+            "duration     461\n",
+            "campaign     478\n",
+            "pdays        444\n",
+            "previous     400\n",
+            "poutcome     453\n",
+            "y            451\n",
             "dtype: int64\n"
           ],
           "name": "stdout"
         }
       ]
     },
     {
+      "cell_type": "code",
       "metadata": {
         "id": "gy3hrHu40J__",
         "colab_type": "code",
+        "outputId": "0f295ef2-74ac-40c8-cf5f-f01bdbbd9ed0",
         "colab": {
           "base_uri": "https://localhost:8080/",
           "height": 2193
-        },
-        "outputId": "69f35783-00ff-4a34-aa34-8bd91301bcfd"
+        }
       },
-      "cell_type": "code",
       "source": [
         "num = df._get_numeric_data()\n",
         "Q1 = num.quantile(0.25)\n",
@@ -323,7 +333,7 @@
         "print(num < (Q1 - 1.5 * IQR))\n",
         "print(num > (Q3 + 1.5 * IQR))\n"
       ],
-      "execution_count": 13,
+      "execution_count": 7,
       "outputs": [
         {
           "output_type": "stream",
@@ -395,19 +405,19 @@
             "        age  balance    day  duration  campaign  pdays  previous\n",
             "0     False    False  False     False     False  False     False\n",
             "1     False     True  False     False     False   True      True\n",
-            "2     False    False  False     False     False   True     False\n",
+            "2     False    False  False     False     False   True      True\n",
             "3     False    False  False     False     False  False     False\n",
             "4     False    False  False     False     False  False     False\n",
             "5     False    False  False     False     False   True      True\n",
-            "6     False    False  False     False     False   True      True\n",
+            "6     False    False  False     False     False  False      True\n",
             "7     False    False  False     False     False  False     False\n",
             "8     False    False  False     False     False  False     False\n",
             "9     False    False  False     False     False   True      True\n",
             "10    False     True  False     False     False  False     False\n",
             "11    False    False  False     False     False  False     False\n",
             "12    False    False  False     False     False  False     False\n",
             "13    False    False  False     False     False  False     False\n",
-            "14    False    False  False     False     False   True      True\n",
+            "14    False    False  False     False     False   True     False\n",
             "15    False    False  False     False     False  False     False\n",
             "16    False     True  False     False     False  False     False\n",
             "17    False    False  False     False     False   True      True\n",
@@ -428,7 +438,7 @@
             "4492  False    False  False     False     False   True      True\n",
             "4493  False    False  False     False     False  False     False\n",
             "4494  False    False  False     False     False  False     False\n",
-            "4495  False    False  False     False     False  False     False\n",
+            "4495  False    False  False      True     False  False     False\n",
             "4496  False    False  False     False     False  False     False\n",
             "4497  False    False  False     False     False  False     False\n",
             "4498  False    False  False      True     False  False     False\n",
@@ -452,7 +462,7 @@
             "4516  False    False  False     False     False  False     False\n",
             "4517  False    False  False     False     False  False     False\n",
             "4518  False    False  False     False      True  False     False\n",
-            "4519  False    False  False     False     False   True      True\n",
+            "4519  False    False  False     False     False  False      True\n",
             "4520  False    False  False     False     False   True      True\n",
             "\n",
             "[4521 rows x 7 columns]\n"
@@ -462,4 +472,4 @@
       ]
     }
   ]
-}
+}