{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Análise Exploratória de Dados\n",
    "\n",
    "#### SME0221 Introdução à Inferência Estatística\n",
    "\n",
    "\n",
    "\n",
    "por **Cibele Russo** \n",
    "\n",
    "**ICMC/USP - São Carlos SP**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WXcecFLkpgNH",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Programa\n",
    "\n",
    "a. Medidas de posição ou localização\n",
    "\n",
    "b. Medidas de dispersão\n",
    "\n",
    "c. Apresentação tabular\n",
    "\n",
    "d. Representação Gráfica"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rXqJWNMEpgNM",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Análise exploratória de dados\n",
    "\n",
    "\n",
    "**Análise descritiva** ou **análise exploratória de dados** (AED) tem como objetivos básicos:\n",
    "\n",
    "- explorar os dados para descobrir ou identificar aspectos ou padrões de maior interesse,\n",
    "\n",
    "- representar os dados de forma a destacar ou chamar a atenção para aspectos ou padrões que podem ou não se confirmar inferencialmente.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tukey (1977) chama a **análise exploratória de dados** de **trabalho de detetive**, que busca pistas e evidência, e a **análise confirmatória de dados** é um **trabalho judicial ou quase-judicial**, que analisa e avalia a força das provas e da evidência.\n",
    "\n",
    "\n",
    "Tukey também diz que: **\"A análise exploratória de dados nunca conta a história toda, mas nada é tão perfeito para ser considerado a pedra fundamental, um primeiro passo para a análise de dados\"**.\n",
    "\n",
    "\n",
    "É importante salientar que a AED é um trabalho inicial, a pedra fundamental, e os resultados devem ser analisados com uma análise confirmatória. \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TvN46dkKuBVr",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "![AED.png](data:image/png;base64,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)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fC1Yp17YpgNQ",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## A natureza dos dados\n",
    "\n",
    "\n",
    "Nesta aula, e quase sempre neste curso de Estatística para Ciência de Dados, trataremos de **dados retangulares**, que tem nas linhas as **unidades amostrais (exemplos, samples)** e nas colunas as **variáveis (atributos, features)**.\n",
    "\n",
    "\n",
    "\n",
    "### Tipos de variáveis\n",
    "\n",
    "\n",
    "- **Qualitativas (não-numéricas)**\n",
    "\n",
    "    - **Nominais:** sexo, cor da pele, fumante/não-fumante, adimplente/inadimplente\n",
    "    \n",
    "    - **Ordinais:** escolaridade (em categorias), grau de satisfação, idade (em faixas)\n",
    "\n",
    "\n",
    "- **Quantitativas (numéricas)**\n",
    "\n",
    "    - **Discretas:** número de defeitos em uma peça, número de produtos contratados\n",
    "    \n",
    "    - **Contínuas:** peso, idade, pressão sanguínea, valor contratado de um produto\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LWn6OK9-pgNS",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Medidas-resumo\n",
    "\n",
    "- **Medidas de posição**\n",
    "    - **Média:** boas propriedades estatísticas\n",
    "    - **Mediana:** medida resistente\n",
    "    - **Moda:** valor mais frequente\n",
    "    - **Quantis:** caracterização da distribuição dos dados\n",
    "\n",
    "\n",
    "- **Medidas de dispersão**\n",
    "    - **Desvio-padrão**\n",
    "    - **Variância**\n",
    "    - **Amplitude (range)**\n",
    "    - **Coeficiente de variação: medida de dispersão relativa**\n",
    "    \n",
    "    \n",
    "- **Assimetria**: Assimetria da distribuição dos dados\n",
    "\n",
    "\n",
    "- **Curtose**: Achatamento da distribuição\n",
    "\n",
    "\n",
    "- **Medidas de associação**: Covariância, Coeficiente de correlação de Pearson, Coeficiente de correlação de Spearman"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3W91GtSXpgNT",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Medidas de posição\n",
    "\n",
    "Daqui em diante, vamos estabelecer $X_1,\\ldots, X_n$ é uma amostra aleatória e $x_1,\\ldots, x_n$ os **dados observados** dessa amostra. As medidas aqui apresentadas são **amostrais** e são obtidas a partir de $x_1,\\ldots, x_n$.\n",
    "\n",
    "A **média** (amostral observada) é definida como\n",
    "\n",
    "\n",
    "$\\bar{x} = \\displaystyle{\\frac{\\displaystyle\\sum_{i=1}^{n} x_i}{n}}$\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Medidas de posição\n",
    "\n",
    "Considere agora os dados ordenados $x_{(1)},\\ldots, x_{(n)}$, isto é,\n",
    "\n",
    "$x_{(1)} = min(x_1,\\ldots, x_n)$ e $x_{(n)} = max(x_1,\\ldots, x_n)$.\n",
    "\n",
    "Se $n$ é ímpar, a posição central é $c = (n + 1) / 2$. Se $n$ é par, as posições centrais são $c = n / 2$ e $c + 1 = n / 2  + 1$. \n",
    "\n",
    "\n",
    "A **mediana** é definida como\n",
    "\n",
    "\n",
    "$Md = \\bigg \\{\\begin{array}{l}x_{(c)}, \\mbox{se n é ímpar}\\\\\n",
    "\\displaystyle\\frac{x_{(c)}+x_{(c+1)}}{2}, \\mbox{se n é par}\\end{array}$\n",
    "\n",
    "\n",
    "\n",
    "A **moda** é o valor mais frequente da amostra. Não necessariamente existe."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wRE72i6BpgNV",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Medidas de posição\n",
    "\n",
    "Um **quantil** é o valor que provoca uma divisão conveniente nos valores ordenados. O quantil de 10% divide os dados de tal forma que 10% dos menores valores fiquem \"à sua esquerda\". O quantil de 50% é a mediana. \n",
    "\n",
    "\n",
    "\n",
    "Os **quartis** dividem os dados em porções de 25%.\n",
    "\n",
    "\n",
    "\n",
    "Os **decis** dividem os dados em porções de 10%.\n",
    "\n",
    "\n",
    "\n",
    "Os **percentis** dividem os dados em porções de 1%.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "AHJgmYVjpgNW",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Medidas de dispersão\n",
    "\n",
    "A **variância amostral** é dada por\n",
    "\n",
    "$s^2 = \\displaystyle{\\frac{\\displaystyle\\sum_{i=1}^{n}(x_i-\\bar{x})^2}{n}}.$\n",
    "\n",
    "\n",
    "O **desvio padrão** é dado por\n",
    "\n",
    "$s = \\displaystyle\\sqrt{\\frac{\\displaystyle\\sum_{i=1}^{n}(x_i-\\bar{x})^2}{n}}.$\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Medidas de dispersão\n",
    "\n",
    "É comum, entretanto, utilizar as medidas corrigidas:\n",
    "\n",
    "**Variância amostral corrigida:**\n",
    "\n",
    "$s^2 = \\displaystyle{\\frac{\\displaystyle\\sum_{i=1}^{n}(x_i-\\bar{x})^2}{n-1}}$\n",
    "\n",
    "\n",
    "**Desvio padrão corrigido:**\n",
    "\n",
    "$s = \\displaystyle\\sqrt{\\frac{\\displaystyle\\sum_{i=1}^{n}(x_i-\\bar{x})^2}{n-1}}$\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Medidas de dispersão\n",
    "\n",
    "\n",
    "A **amplitude** é dada por\n",
    "\n",
    "$A = x_{(n)} -x_{(1)}.$\n",
    "\n",
    "\n",
    "\n",
    "O **coeficiente de variação** (amostral) é dado pela razão entre o desvio-padrão e a média\n",
    "\n",
    "$CV = \\displaystyle{\\frac{s}{\\bar{x}}}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Dk6Zu4J-pgNY",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Assimetria \n",
    "\n",
    "- Distribuição simétrica: média = mediana = moda\n",
    "- Distribuição assimétrica à direita: moda < mediana < média\n",
    "- Distribuição assimétrica à esquerda: média < mediana < moda \n",
    "\n",
    "\n",
    "### Curtose\n",
    "\n",
    "- Distribuições mesocúrticas: achatamento da distribuição normal\n",
    "- Distribuições leptocúrticas: distribuição mais concentrada \n",
    "- Distribuições platicúrticas: distribuição mais achatada\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 368
    },
    "executionInfo": {
     "elapsed": 1786,
     "status": "ok",
     "timestamp": 1618514359564,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "zGJjujpPpgNa",
    "outputId": "d2318dd9-a62d-4b4b-c8e5-4313cdb4a1b5",
    "scrolled": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Ilustração das medidas média, moda, mediana para dados simétricos\n",
    "# Adaptado de https://stackoverflow.com/questions/51417483/mean-median-mode-lines-showing-only-in-last-graph-in-seaborn/51417635\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "df = pd.DataFrame({\"rating\": [5, 6, 6, 7, 7, 7, 7, 8, 8, 9]})\n",
    "\n",
    "f, (ax_box, ax_hist) = plt.subplots(2, sharex=True, gridspec_kw= {\"height_ratios\": (0.2, 1)})\n",
    "mean=df['rating'].mean()\n",
    "median=df['rating'].median()\n",
    "mode=df['rating'].mode().values[0]\n",
    "\n",
    "sns.boxplot(data=df, x=\"rating\", ax=ax_box)\n",
    "ax_box.axvline(mean, color='r', linestyle='--')\n",
    "ax_box.axvline(median, color='g', linestyle='-')\n",
    "ax_box.axvline(mode, color='b', linestyle='-')\n",
    "\n",
    "sns.histplot(data=df, x=\"rating\", ax=ax_hist, kde=True)\n",
    "ax_hist.axvline(mean, color='r', linestyle='--', label=\"Média\")\n",
    "ax_hist.axvline(median, color='g', linestyle='-', label=\"Mediana\")\n",
    "ax_hist.axvline(mode, color='b', linestyle='-', label=\"Moda\")\n",
    "\n",
    "plt.legend()\n",
    "\n",
    "ax_box.set(xlabel='')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 368
    },
    "executionInfo": {
     "elapsed": 2436,
     "status": "ok",
     "timestamp": 1618514360227,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "9yhDkV4opgNe",
    "outputId": "c1c08e51-00f6-4029-a168-e5051f7e5ec7",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Ilustração das medidas média, moda, mediana para dados assimétricos à direita ou assimetria positiva\n",
    "# Adaptado de https://stackoverflow.com/questions/51417483/mean-median-mode-lines-showing-only-in-last-graph-in-seaborn/51417635\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import statistics\n",
    "\n",
    "df = pd.DataFrame({\"rating\": [1,1,1,2,2,3,4,5,5,10]})\n",
    "\n",
    "f, (ax_box, ax_hist) = plt.subplots(2, sharex=True, gridspec_kw= {\"height_ratios\": (0.2, 1)})\n",
    "mean=df['rating'].mean()\n",
    "median=df['rating'].median()\n",
    "mode=df['rating'].mode().values[0]\n",
    "\n",
    "sns.boxplot(data=df, x=\"rating\", ax=ax_box)\n",
    "ax_box.axvline(mean, color='r', linestyle='--')\n",
    "ax_box.axvline(median, color='g', linestyle='-')\n",
    "ax_box.axvline(mode, color='b', linestyle='-')\n",
    "\n",
    "sns.histplot(data=df, x=\"rating\", ax=ax_hist, kde=True)\n",
    "ax_hist.axvline(mean, color='r', linestyle='--', label=\"Média\")\n",
    "ax_hist.axvline(median, color='g', linestyle='-', label=\"Mediana\")\n",
    "ax_hist.axvline(mode, color='b', linestyle='-', label=\"Moda\")\n",
    "\n",
    "plt.legend()\n",
    "\n",
    "\n",
    "ax_box.set(xlabel='')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 368
    },
    "executionInfo": {
     "elapsed": 3266,
     "status": "ok",
     "timestamp": 1618514361069,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "uq6ngft-pgNf",
    "outputId": "ad10d8a0-739c-4329-94e9-b6c1d82c918e",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Ilustração das medidas média, moda, mediana para dados assimétricos à esquerda ou com assimetria negativa\n",
    "# Adaptado de: https://stackoverflow.com/questions/51417483/mean-median-mode-lines-showing-only-in-last-graph-in-seaborn/51417635\n",
    "\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import statistics\n",
    "\n",
    "df = pd.DataFrame({\"rating\": [1, 4, 6, 8, 8, 9, 10, 10, 10, 10]})\n",
    "\n",
    "f, (ax_box, ax_hist) = plt.subplots(2, sharex=True, gridspec_kw= {\"height_ratios\": (0.2, 1)})\n",
    "mean=df['rating'].mean()\n",
    "median=df['rating'].median()\n",
    "mode = statistics.mode(df['rating'])\n",
    "\n",
    "sns.boxplot(data=df, x=\"rating\", ax=ax_box)\n",
    "ax_box.axvline(mean, color='b', linestyle='--')\n",
    "ax_box.axvline(median, color='r', linestyle='-')\n",
    "ax_box.axvline(mode, color='g', linestyle='-')\n",
    "\n",
    "sns.histplot(data=df, x=\"rating\", ax=ax_hist, kde=True)\n",
    "ax_hist.axvline(mean, color='b', linestyle='--', label=\"Média\")\n",
    "ax_hist.axvline(median, color='r', linestyle='-', label=\"Mediana\")\n",
    "ax_hist.axvline(mode, color='g', linestyle='-', label=\"Moda\")\n",
    "\n",
    "plt.legend()\n",
    "\n",
    "ax_box.set(xlabel='')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "GC3eCa9NpgNg",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Curtose\n",
    "\n",
    "Referências: \n",
    "- https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.kurtosis.html\n",
    "- https://pt.wikipedia.org/wiki/Curtose\n",
    "\n",
    "Medida que caracteriza o achatamento da curva.\n",
    "\n",
    "- Curtose $\\approx 0$: achatamento da curva normal\n",
    "- Curtose $>0$: leptocúrtica, distribuição mais afunilada\n",
    "- Curtose $<0$: platicúrtica, distribuição mais achatada\n",
    "\n",
    "Obs: Distribuição normal https://www.spss-tutorials.com/normal-distribution/\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3256,
     "status": "ok",
     "timestamp": 1618514361072,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "yQUNjzZfpgNg",
    "outputId": "c8aab2f6-bc2b-48d3-dc4e-892291fdcf52",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.018503562462675482"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from scipy.stats import norm, kurtosis\n",
    "\n",
    "data = norm.rvs(size=100000)\n",
    "\n",
    "kurtosis(data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "executionInfo": {
     "elapsed": 3248,
     "status": "ok",
     "timestamp": 1618514361073,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "Br6G9JBtpgNh",
    "outputId": "e42067ff-016f-4091-ec53-608dcdaae997",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Referência: https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.kurtosis.html\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.stats as stats\n",
    "from scipy.stats import kurtosis\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "x = np.linspace(-5, 5, 100)\n",
    "ax = plt.subplot()\n",
    "distnames = ['laplace', 'norm', 'uniform']\n",
    "\n",
    "\n",
    "\n",
    "for distname in distnames:\n",
    "    if distname == 'uniform':\n",
    "        dist = getattr(stats, distname)(loc=-2, scale=4)\n",
    "    else:\n",
    "        dist = getattr(stats, distname)\n",
    "    data = dist.rvs(size=1000)\n",
    "    kur = kurtosis(data, fisher=True)\n",
    "    y = dist.pdf(x)\n",
    "    ax.plot(x, y, label=\"{}, {}\".format(distname, round(kur, 3)))\n",
    "    ax.legend()\n",
    "    \n",
    "    \n",
    "    \n",
    "# Normal: mesocúrtica\n",
    "# Laplace: leptocúrtica\n",
    "# Uniforme: platicúrtica"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tyTI3VU8pgNi",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Medidas de associação entre variáveis quantitativas\n",
    "\n",
    "\n",
    "Sejam $X$ e $Y$ variáveis quantitativas de interesse e as amostras aleatórias observadas $x_1,\\ldots,x_n$ e $y_1,\\ldots,y_n$, respectivamente. As medidas de associação mais utilizadas são:\n",
    "\n",
    "\n",
    "### Covariância (amostral)\n",
    "\n",
    "$s_{XY} = \\displaystyle{\\frac{\\displaystyle\\sum_{i=1}^{n}(x_i-\\bar{x})(y_i-\\bar{y})}{n-1}}$\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### Coeficiente de correlação linear (amostral) de Pearson\n",
    "\n",
    "Referência: https://pt.wikipedia.org/wiki/Coeficiente_de_correla%C3%A7%C3%A3o_de_Pearson\n",
    "\n",
    "\n",
    "$r = \\displaystyle{\\frac{s_{XY}}{\\sqrt{s^2_X s^2_Y }}}$\n",
    "\n",
    "\n",
    "Propriedade: \n",
    "\n",
    " $-1 \\leq r \\leq 1$\n",
    "\n",
    "\n",
    "É comum usar as seguintes classificações:\n",
    "\n",
    "1. $r=1$ indica uma correlação perfeita e positiva\n",
    "\n",
    "2. $r=-1$ indica uma correlação perfeita e negativa\n",
    "\n",
    "3. $0.7 \\leq |r| \\leq 1$ indica uma correlação forte\n",
    "\n",
    "4. $0.5 \\leq |r| \\leq 0.69$ indica uma correlação moderada\n",
    "\n",
    "5. $0 \\leq |r| \\leq 0.49$ indica uma correlação fraca\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### Coeficiente de correlação de Spearman\n",
    "\n",
    "Avalia relações monótonas entre duas variáveis\n",
    "\n",
    "Referência: https://pt.wikipedia.org/wiki/Coeficiente_de_correla%C3%A7%C3%A3o_de_postos_de_Spearman\n",
    "\n",
    "\n",
    "\n",
    "## Associação entre variáveis qualitativas e quantitativas\n",
    "\n",
    "\n",
    "Alguns casos que veremos mais adiante:\n",
    "\n",
    "- Associação entre variáveis quantitativas e qualitativas: Testes para comparação de médias em duas populações.\n",
    "\n",
    "- Associação entre variáveis qualitativas: Teste qui-quadrado, teste exato de Fisher, entre outros.\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ivudn59MpgNj",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Representação gráfica e tabular de dados\n",
    "\n",
    "- **Variáveis qualitativas:**\n",
    "\n",
    "    - **Tabela de frequência:** Resume a informação dos dados de forma a possibilitar a observação de frequências absolutas ou relativas de cada categoria das variáveis qualitativas (ou valores assumidos pelas variáveis quantitativas discretas).\n",
    "    \n",
    "    - **Gráfico de barras** Representação gráfica das frequências de cada categoria das variáveis qualitativas (ou valores assumidos pelas variáveis quantitativas discretas). As barras são separadas.\n",
    "    \n",
    "    - **Gráfico de Pareto**: Gráfico de barras + frequências acumuladas das categorias.\n",
    "    \n",
    "    - **Gráfico de setores (pizza)**: Representação gráfica das proporções das categorias das variáveis quantitativas discretas.\n",
    "    \n",
    "    \n",
    "- **Variáveis quantitativas discretas:**\n",
    "    - **Tabelas de frequências**\n",
    "       \n",
    "    - **Gráficos de barras**\n",
    "    \n",
    "    - **Gráficos de pontos**\n",
    "    \n",
    "    \n",
    "    \n",
    "- **Variáveis quantitativas contínuas:**\n",
    "\n",
    "    - **Histogramas**: Representação gráfica para uma aproximação da distribuição de uma variável quantitativa contínua, discretizada em classes de tamanhos convenientes. As barras são adjacentes. Permitem observar a localização, dispersão, assimetria, número de picos, curtose dos dados.\n",
    "    \n",
    "    - **Gráficos de linhas (dados coletados ao longo do tempo)**\n",
    "        \n",
    "    - **Boxplots (gráficos de caixas)**: Representação gráfica inteligente que permite a observação da localização, dispersão, assimetria, pontos discrepantes (outliers). Além disso, permite comparar visualemente a distribuição de dados em dois grupos. Pode indicar evidências sobre a igualdade das médias entre os dados de dois grupos, pendente de análise confirmatória inferencial.    \n",
    "    \n",
    "  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ADz3TbI1pgNk",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "![boxplot.png](data:image/png;base64,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)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8o-JyjQLpgNl",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Representação tabular\n",
    "\n",
    "Tabelas que resumem a informação da base completa de dados.\n",
    "\n",
    "- **Tabelas de frequências:** Resumo dos dados originais considerando as frequências observadas na amostra, de variáveis qualitativas ou variáveis que foram categorizadas\n",
    "\n",
    "\n",
    "- **Tabelas de dupla entrada:** Avaliação da associação entre variáveis qualitativas ou que foram categorizadas."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "J9UkYDXcpgNm",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Aplicação com visualização e exploração de dados\n",
    "\n",
    "Considere os dados de 100 mil clientes de um banco no arquivo dados_banco.csv. Estão disponíveis as variáveis:\n",
    "\n",
    "- Cliente: Identificador do cliente.\n",
    "- Sexo: Feminino (F) ou Masculino (M)\n",
    "- Idade: Idade do cliente, em anos completos.\n",
    "- Empresa: Tipo da empresa em que trabalha: Pública, Privada ou Autônomo\n",
    "- Salário: Salário declarado pelo cliente na abertura da conta, em reais.\n",
    "- Saldo_cc: Saldo em conta corrente, em reais.\n",
    "- Saldo_poupança: Saldo em poupança, em reais.\n",
    "- Saldo_investimento: Saldo em investimentos, em reais.\n",
    "- Devedor_cartao: Valor em atraso no cartão de crédito, em reais.\n",
    "- Inadimplente: Se o cliente é considerado inadimplente atualmente (1) ou não (0), de acordo com critérios preestabelecidos.\n",
    "\n",
    "\n",
    "Desenvolva a exploração e visualização dos dados. Verifique possíveis associações entre variáveis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 444
    },
    "executionInfo": {
     "elapsed": 3791,
     "status": "ok",
     "timestamp": 1618514361628,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "LfxBGxsapgNm",
    "outputId": "80d1cc9e-f968-4d00-c98e-441538e017e8",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "import os.path\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "from scipy import stats\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "# Modifique o diretório para fazer a leitura dos dados em dados_banco.csv\n",
    "\n",
    "pkgdir = '/hdd/Disciplinas/!2022 Introdução à Inferência Estatística/1. Visualização e Exploração de Dados'\n",
    "\n",
    "# Dados banco - Leitura dos dados\n",
    "dados = pd.read_csv('dados_banco.csv', index_col=0)\n",
    "\n",
    "#dados = pd.read_csv('dados_banco.csv', index_col=0, decimal=',')\n",
    "\n",
    "dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "executionInfo": {
     "elapsed": 3784,
     "status": "ok",
     "timestamp": 1618514361630,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "3VAXk_bgpgNn",
    "outputId": "8fb617a0-b07d-448c-f2a3-4d7c7747c3e0",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "dados.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zmF_eJCGpgNo",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Classificação das variáveis por tipo\n",
    "\n",
    "- Sexo: qualitativa nominal\n",
    "- Idade: quantitativa contínua\n",
    "- Empresa: qualitativa nominal\n",
    "- Salário: quantitativa contínua\n",
    "- Saldo_cc: quantitativa contínua\n",
    "- Saldo_poupança: quantitativa contínua\n",
    "- Saldo_investimento: quantitativa contínua\n",
    "- Devedor_cartão: quantitativa contínua\n",
    "- Inadimplente: qualitativa nominal (embora numérica)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5T5OtgwapgNp",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Tabela de frequências (absolutas e relativas)\n",
    "(para a Empresa, repetir para outras variáveis qualitativas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 175
    },
    "executionInfo": {
     "elapsed": 3781,
     "status": "ok",
     "timestamp": 1618514361632,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "Nf2LHRFhpgNp",
    "outputId": "78e97feb-1dd6-438b-ee08-50e44f0526ba",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "# Tabela de frequências absolutas\n",
    "\n",
    "tab = pd.crosstab(index=dados['Empresa'], columns='count')\n",
    "\n",
    "tab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 175
    },
    "executionInfo": {
     "elapsed": 3777,
     "status": "ok",
     "timestamp": 1618514361634,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "TYoiStFNpgNq",
    "outputId": "0c90b714-610b-411f-c25e-9e82224e09c8",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "tab = pd.crosstab(index=dados['Empresa'], columns='count')\n",
    "\n",
    "# Tabela de frequências relativas\n",
    "tab/tab.sum()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "sN9bRIfSpgNq",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "**Análise:** Na base de dados, cerca de ___ dos clientes trabalham em empresas privadas, ___ em empresas públicas e ___ são autônomos."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "i5NZizHApgNr",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Medidas resumo\n",
    "(para a idade, poderia repetir para as outras variáveis quantitativas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3772,
     "status": "ok",
     "timestamp": 1618514361636,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "ktCMIGF4pgNr",
    "outputId": "28552ea0-d34d-47f9-a9ff-222e566a43cf"
   },
   "outputs": [],
   "source": [
    "# Média\n",
    "\n",
    "dados['Idade'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3767,
     "status": "ok",
     "timestamp": 1618514361637,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "oZMgHQaZpgNr",
    "outputId": "a994f034-c670-44c8-fd38-282dbcd681b9"
   },
   "outputs": [],
   "source": [
    "# Mediana\n",
    "\n",
    "dados['Idade'].median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3761,
     "status": "ok",
     "timestamp": 1618514361638,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "bwYFNDwTpgNs",
    "outputId": "4119ee5c-dfc9-4827-e1c1-929b06cfde74"
   },
   "outputs": [],
   "source": [
    "# Desvio-padrão\n",
    "\n",
    "dados['Idade'].std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3754,
     "status": "ok",
     "timestamp": 1618514361638,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "bMVflD_OpgNs",
    "outputId": "2c5e0e50-5053-4ba8-e73d-955568633287"
   },
   "outputs": [],
   "source": [
    "# Média de idade por grupos\n",
    "\n",
    "dados.groupby('Sexo')['Idade'].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Xbk75PIipgNt"
   },
   "source": [
    "**Análise:** A média de idade nos dados é ___ anos, a mediana é ___ anos. O desvio-padrão da idade na base de dados geral é ___ anos. Entre mulheres, a média de idade é ___ anos e entre homens, ___ anos. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3749,
     "status": "ok",
     "timestamp": 1618514361640,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "lbGgCI2ppgNt",
    "outputId": "c7b0cf84-5745-4dc9-c6cf-4ee6f89f0a8c"
   },
   "outputs": [],
   "source": [
    "# Média de idade por grupos\n",
    "\n",
    "dados.groupby('Empresa')['Idade'].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "SY7TgaZ2pgNu"
   },
   "source": [
    "**Análise:** A média de idade entre os clientes autônomos é de 29.3 anos, entre clientes que trabalham em empresas privadas é 32.9 anos e para clientes que trabalham em empresas públicas é 30.7 anos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "executionInfo": {
     "elapsed": 3742,
     "status": "ok",
     "timestamp": 1618514361641,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "WGFswn7rpgNu",
    "outputId": "eddd7ef2-593d-4bbd-932b-2515ed5a4649"
   },
   "outputs": [],
   "source": [
    "# Moda - para a Empresa\n",
    "import statistics\n",
    "\n",
    "statistics.mode(dados['Empresa'])\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "63P2oJQzpgNv"
   },
   "source": [
    "**Análise:** Na base de dados, o tipo de empresa mais comum é a empresa privada."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3737,
     "status": "ok",
     "timestamp": 1618514361643,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "5U9VEZKjpgNv",
    "outputId": "65acff5f-fab4-4be1-e180-930023cba6a4"
   },
   "outputs": [],
   "source": [
    "# Ordenação dos dados\n",
    "\n",
    "np.sort(dados['Idade'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 4134,
     "status": "ok",
     "timestamp": 1618514362047,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "VQEYdvovpgNv",
    "outputId": "abdda860-c800-4d1d-8802-97a1b9b9fbb7"
   },
   "outputs": [],
   "source": [
    "# Quantis de 95% e 25%\n",
    "\n",
    "np.percentile(dados['Idade'],95)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 4130,
     "status": "ok",
     "timestamp": 1618514362048,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "qi8eOZPlpgNw",
    "outputId": "3043fa4b-58dd-43c5-d8d6-f98b5827517e"
   },
   "outputs": [],
   "source": [
    "np.percentile(dados['Idade'],25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 351
    },
    "executionInfo": {
     "elapsed": 4944,
     "status": "ok",
     "timestamp": 1618514362867,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "Y3pw0lvPpgNw",
    "outputId": "c70664c7-fa31-446f-8506-dd6ccf2f6324"
   },
   "outputs": [],
   "source": [
    "sns.displot(dados['Idade'], height=4, aspect=2, kind='hist', rug=True, bins=10);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 144
    },
    "executionInfo": {
     "elapsed": 5254,
     "status": "ok",
     "timestamp": 1618514363182,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "FrQ8PC-dpgNw",
    "outputId": "f31555f4-8461-4de5-e14e-d68e4ce68b2e"
   },
   "outputs": [],
   "source": [
    "tab1 = pd.crosstab(index=dados['Sexo'], columns='count')\n",
    "tab1/tab1.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ffNfKzPOpgNx"
   },
   "source": [
    "### Estatísticas descritivas dos dados com describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 320
    },
    "executionInfo": {
     "elapsed": 5249,
     "status": "ok",
     "timestamp": 1618514363183,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "o3YFIPLPpgNx",
    "outputId": "857d8b1c-0fce-40ce-c05f-753feb1360ea"
   },
   "outputs": [],
   "source": [
    "dados.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 300
    },
    "executionInfo": {
     "elapsed": 5246,
     "status": "ok",
     "timestamp": 1618514363184,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "_oLixNB7pgNy",
    "outputId": "b4c7405a-6979-4e41-cb2b-1aa8f2af56f7"
   },
   "outputs": [],
   "source": [
    "dados.loc[:,dados.columns != 'Cliente'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1UqS3IqzpgNz"
   },
   "source": [
    "### Gráfico de setores (pizza)\n",
    "\n",
    "- https://blog.algorexhealth.com/2018/03/almost-10-pie-charts-in-10-python-libraries/\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 175
    },
    "executionInfo": {
     "elapsed": 5241,
     "status": "ok",
     "timestamp": 1618514363185,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "lIoXrRxRpgNz",
    "outputId": "719ec36c-a3fe-452c-d4f6-7dbbfb2ee273"
   },
   "outputs": [],
   "source": [
    "tab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 248
    },
    "executionInfo": {
     "elapsed": 5237,
     "status": "ok",
     "timestamp": 1618514363186,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "1dzuNAOVpgN0",
    "outputId": "c32cd3da-7f4d-484c-dc5a-8977baad4df2"
   },
   "outputs": [],
   "source": [
    "plot = tab.plot.pie(y='count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 144
    },
    "executionInfo": {
     "elapsed": 5652,
     "status": "ok",
     "timestamp": 1618514363605,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "QBpVoYQOpgN2",
    "outputId": "132bfb8c-f289-4363-99fa-ad25031fcd93"
   },
   "outputs": [],
   "source": [
    "# Tabela de frequências absolutas\n",
    "\n",
    "tab = pd.crosstab(index=dados['Sexo'], columns='count')\n",
    "\n",
    "tab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 248
    },
    "executionInfo": {
     "elapsed": 5645,
     "status": "ok",
     "timestamp": 1618514363606,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "DtRuD6-8pgN2",
    "outputId": "d841b07b-fd73-4e3a-fd03-032106717193"
   },
   "outputs": [],
   "source": [
    "plot = tab.plot.pie(y='count')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NSWCSBljpgN2"
   },
   "source": [
    "### Gráfico de barras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 278
    },
    "executionInfo": {
     "elapsed": 5642,
     "status": "ok",
     "timestamp": 1618514363608,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "L_YnmBOHpgN3",
    "outputId": "e3319adb-ba38-4c72-c72e-d3e436af64c5"
   },
   "outputs": [],
   "source": [
    "tab.plot.bar()\n",
    "plt.legend(title='Sexo')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tWGYJyoMpgN4"
   },
   "source": [
    "### Boxplot\n",
    "\n",
    "- Posição\n",
    "- Dispersão\n",
    "- Outliers\n",
    "- Assimetria\n",
    "\n",
    "https://seaborn.pydata.org/generated/seaborn.boxplot.html\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dados['Salario']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 296
    },
    "executionInfo": {
     "elapsed": 6018,
     "status": "ok",
     "timestamp": 1618514363989,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "vX7X-MUqpgN4",
    "outputId": "0d1abfdb-e442-41eb-ae90-479b32130436"
   },
   "outputs": [],
   "source": [
    "sns.boxplot(y=dados['Salario'], palette='pastel')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "6cxS74qhpgN5"
   },
   "source": [
    "**Histograma**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 351
    },
    "executionInfo": {
     "elapsed": 6015,
     "status": "ok",
     "timestamp": 1618514363990,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "Dn_x1Tp1pgN5",
    "outputId": "0e38be47-9678-498d-d36b-13da65552afc",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "sns.displot(dados['Salario'],kde=False, bins=10, height=5, aspect=2);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 355
    },
    "executionInfo": {
     "elapsed": 7131,
     "status": "ok",
     "timestamp": 1618514365111,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "u2PXJTc4pgN5",
    "outputId": "23fc9acd-033e-4ea5-dcef-e6634a9bfefa"
   },
   "outputs": [],
   "source": [
    "sns.displot(dados['Salario']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 355
    },
    "executionInfo": {
     "elapsed": 8058,
     "status": "ok",
     "timestamp": 1618514366042,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "kPgHbaIapgN6",
    "outputId": "669453f4-7de5-4264-dd08-23a859542f5c"
   },
   "outputs": [],
   "source": [
    "sns.displot(dados['Salario'], bins=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "shd1zeJEpgN6"
   },
   "source": [
    "**Densidade alisada**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 355
    },
    "executionInfo": {
     "elapsed": 9280,
     "status": "ok",
     "timestamp": 1618514367269,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "6h0ObSxBpgN6",
    "outputId": "954494ae-194a-4de5-9308-d237114d80fb"
   },
   "outputs": [],
   "source": [
    "sns.displot(dados['Salario'], kind='kde')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "POudmh5XpgN7"
   },
   "source": [
    "### Associação entre duas variáveis qualitativas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 175
    },
    "executionInfo": {
     "elapsed": 9276,
     "status": "ok",
     "timestamp": 1618514367271,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "RuyYMVhApgN7",
    "outputId": "78931444-70b2-4248-ca3d-90760d0122ba"
   },
   "outputs": [],
   "source": [
    "# Tabela de dupla entrada\n",
    "\n",
    "tabela_dupla = pd.crosstab(index=dados['Empresa'], columns=dados['Sexo'])\n",
    "\n",
    "tabela_dupla"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 320
    },
    "executionInfo": {
     "elapsed": 9856,
     "status": "ok",
     "timestamp": 1618514367856,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "AAeoKPMrpgN8",
    "outputId": "3cb9df07-bcae-4712-a8ad-19c611fe9a16"
   },
   "outputs": [],
   "source": [
    "tabela_dupla.plot.bar()\n",
    "\n",
    "plt.legend(title='Sexo')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 320
    },
    "executionInfo": {
     "elapsed": 9851,
     "status": "ok",
     "timestamp": 1618514367857,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "-vfpOZpmpgN8",
    "outputId": "590a7042-d47a-4446-cecb-c53a57ca8b7b",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "tabela_dupla.plot.bar(stacked=True)\n",
    "\n",
    "plt.legend(title='Sexo')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Gráfico de mosaico"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.graphics.mosaicplot import mosaic\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [10, 5]\n",
    "\n",
    "mosaic(dados,['Sexo', 'Empresa'] );"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "USQs9aydpgN8"
   },
   "source": [
    "### Associação entre variáveis quantitativas e qualitativas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "executionInfo": {
     "elapsed": 10240,
     "status": "ok",
     "timestamp": 1618514368252,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "3ua78SINpgN9",
    "outputId": "3305efa7-bd1f-4499-fad3-b29d5aca1639"
   },
   "outputs": [],
   "source": [
    "ax = sns.boxplot(x='Sexo', y='Salario', data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "executionInfo": {
     "elapsed": 10795,
     "status": "ok",
     "timestamp": 1618514368812,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "RwP5wN51pgN9",
    "outputId": "202f4143-74ea-4ec9-a1f3-11ee7e483de9"
   },
   "outputs": [],
   "source": [
    "ax = sns.boxplot(x='Sexo', y='Salario', hue='Empresa', data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "executionInfo": {
     "elapsed": 10793,
     "status": "ok",
     "timestamp": 1618514368814,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "7w6IMEAUpgN-",
    "outputId": "ee9f1058-8ae2-485d-c48f-2984b6fa78d2"
   },
   "outputs": [],
   "source": [
    "ax = sns.boxplot(x='Empresa', y='Salario', hue='Sexo', data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 383
    },
    "executionInfo": {
     "elapsed": 11330,
     "status": "ok",
     "timestamp": 1618514369356,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "xDe6RVuzpgN-",
    "outputId": "8333239d-f4ea-4ddd-c1cd-d70ab842569a"
   },
   "outputs": [],
   "source": [
    "ax = sns.catplot(x='Sexo', y='Salario', kind='boxen', data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 383
    },
    "executionInfo": {
     "elapsed": 12299,
     "status": "ok",
     "timestamp": 1618514370330,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "vIWT7G59pgN_",
    "outputId": "d3c446d9-1d6f-48ee-d041-bdf67ebae12b"
   },
   "outputs": [],
   "source": [
    "ax = sns.catplot(x='Sexo', y='Salario', kind='violin', data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 383
    },
    "executionInfo": {
     "elapsed": 13408,
     "status": "ok",
     "timestamp": 1618514371444,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "B5bQWQlXpgOA",
    "outputId": "dffa1a44-f56f-499d-cedd-c5432312fb0d"
   },
   "outputs": [],
   "source": [
    "ax = sns.catplot(x='Sexo', y='Salario', hue='Empresa', kind='violin', data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 400
    },
    "executionInfo": {
     "elapsed": 14525,
     "status": "ok",
     "timestamp": 1618514372565,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "nsxRSiAWpgOA",
    "outputId": "64806d9d-efef-449c-fd42-ea37e8a4ac62"
   },
   "outputs": [],
   "source": [
    "sns.catplot(x='Empresa', y='Salario', hue='Sexo', kind='violin', split=True, data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "executionInfo": {
     "elapsed": 15064,
     "status": "ok",
     "timestamp": 1618514373110,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "fDxhR1M2pgOA",
    "outputId": "8d4b1335-9fa0-4a80-d72e-275d0332b708"
   },
   "outputs": [],
   "source": [
    "ax = sns.boxplot(x=\"Sexo\", y=\"Salario\", hue=\"Empresa\",\n",
    "\n",
    "                 data=dados, palette=\"Set3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 312
    },
    "executionInfo": {
     "elapsed": 16462,
     "status": "ok",
     "timestamp": 1618514374514,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "HOzusxbSpgOB",
    "outputId": "f039d4ad-a938-43d4-9af9-8ea5d35a93eb"
   },
   "outputs": [],
   "source": [
    "# Salário médio por tipo de empresa\n",
    "\n",
    "\n",
    "# Estabelecendo o tamanho do gráfico\n",
    "plt.figure(figsize=(8,4))\n",
    "\n",
    "# Título\n",
    "plt.title(\"Salário médio por tipo de empresa\")\n",
    "\n",
    "# Gráfico de barras com salário médio por tipo de empresa\n",
    "sns.barplot(x=dados['Empresa'], y=dados['Salario'])\n",
    "\n",
    "# Label para eixo vertical\n",
    "plt.ylabel(\"Salário\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "g9cqnsmUpgOB"
   },
   "source": [
    "## Associação entre variáveis quantitativas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Bb3qJesqpgOC"
   },
   "source": [
    "**Gráfico de dispersão**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 400
    },
    "executionInfo": {
     "elapsed": 17313,
     "status": "ok",
     "timestamp": 1618514375369,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "uI_ejrFGpgOC",
    "outputId": "3f97a1c2-faeb-4727-ec95-5c2fd290a71e"
   },
   "outputs": [],
   "source": [
    "sns.set_palette('colorblind')\n",
    "sns.relplot(x='Idade', y='Salario', data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 364
    },
    "executionInfo": {
     "elapsed": 27724,
     "status": "ok",
     "timestamp": 1618514385786,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "hdPJc---pgOD",
    "outputId": "d558e47a-8f2e-4532-9a61-3d146621c6c3"
   },
   "outputs": [],
   "source": [
    "sns.set_palette('colorblind')\n",
    "sns.relplot(x='Idade', y='Salario', hue='Sexo', col='Empresa',  data=dados)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "52yCD_IYpgOE"
   },
   "source": [
    "**Gráficos com Regressão**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 364
    },
    "executionInfo": {
     "elapsed": 37705,
     "status": "ok",
     "timestamp": 1618514395773,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "K_cKhwGrpgOE",
    "outputId": "928d4784-fdad-40f5-a6bd-4dd13b1d0b79"
   },
   "outputs": [],
   "source": [
    "sns.lmplot(x='Idade', y='Salario', hue='Sexo', col='Empresa',  data=dados)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "r1CGWhjUpgOF"
   },
   "source": [
    "**Joint plot**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 458
    },
    "executionInfo": {
     "elapsed": 39740,
     "status": "ok",
     "timestamp": 1618514397813,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "1KoD-idwpgOG",
    "outputId": "5c96aedd-5698-4166-e1fa-f680c82aa753"
   },
   "outputs": [],
   "source": [
    "sns.jointplot(x='Idade', y='Salario',   data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 458
    },
    "executionInfo": {
     "elapsed": 41756,
     "status": "ok",
     "timestamp": 1618514399834,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "3shJT3g6pgOH",
    "outputId": "68462297-9a6a-4afa-fda3-e3004939e9ce"
   },
   "outputs": [],
   "source": [
    "sns.jointplot(x='Idade', y='Salario', kind='hex',  data=dados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 458
    },
    "executionInfo": {
     "elapsed": 194735,
     "status": "ok",
     "timestamp": 1618514552817,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "2F2x51vLpgOH",
    "outputId": "1203d7cc-4048-42d0-fc3a-83e01a3a27ed"
   },
   "outputs": [],
   "source": [
    "sns.jointplot(x='Idade', y='Salario', kind='kde',  data=dados)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "V0pyYiGOpgOI"
   },
   "source": [
    "**Coeficiente de correlação de Pearson**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 194726,
     "status": "ok",
     "timestamp": 1618514552820,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "7imJ-qMLpgOJ",
    "outputId": "33474d2f-d38f-4aa6-f01b-484355564a79"
   },
   "outputs": [],
   "source": [
    "from scipy.stats import pearsonr\n",
    "\n",
    "pearsonr(dados['Idade'], dados['Salario'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cEvKzJxnpgOK"
   },
   "source": [
    "## Heatmap (mapa de calor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 408
    },
    "executionInfo": {
     "elapsed": 195158,
     "status": "ok",
     "timestamp": 1618514553257,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "iqmGuj73pgOK",
    "outputId": "861028ff-9355-40dc-e021-8d62990df53c"
   },
   "outputs": [],
   "source": [
    "dados_heatmap = dados.loc[:,dados.columns != 'Cliente'].groupby(['Idade']).mean()\n",
    "dados_heatmap.head()\n",
    "\n",
    "# Estabelecendo o tamanho do gráfico\n",
    "plt.figure(figsize=(10,5))\n",
    "\n",
    "ax = sns.heatmap(dados_heatmap)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "a5e0cleDpgOK"
   },
   "source": [
    "### Gráficos multivariados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 920
    },
    "executionInfo": {
     "elapsed": 215311,
     "status": "ok",
     "timestamp": 1618514573415,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "p5Xemqn0pgOK",
    "outputId": "b8d0b42d-c373-4f99-fd4c-546cbefaea5e"
   },
   "outputs": [],
   "source": [
    "sns.pairplot(dados[['Salario','Saldo_cc', 'Saldo_poupança', 'Saldo_investimento', 'Devedor_cartao']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.pairplot(dados[['Salario','Saldo_cc', 'Saldo_poupança', 'Saldo_investimento', 'Devedor_cartao']], kind='reg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "executionInfo": {
     "elapsed": 242748,
     "status": "ok",
     "timestamp": 1618514600870,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "Xkudz-X4pgOL"
   },
   "outputs": [],
   "source": [
    "dados_nozeros = dados[dados['Saldo_investimento']*dados['Saldo_poupança']!=0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 920
    },
    "executionInfo": {
     "elapsed": 277923,
     "status": "ok",
     "timestamp": 1618514636051,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "YMNm3eODpgOM",
    "outputId": "2f45523f-2e4f-4c7d-d6f0-aa79d3403f3c"
   },
   "outputs": [],
   "source": [
    "sns.pairplot(dados_nozeros[['Salario','Saldo_cc', 'Saldo_poupança', 'Saldo_investimento', 'Devedor_cartao']], kind='reg')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oGfruSg1pgOM"
   },
   "source": [
    "## Agrupamento de dados\n",
    "\n",
    "\n",
    "- Agrupamento hierárquico (dendrograma)\n",
    "\n",
    "- Agrupamento não-hierárquico (k-médias)\n",
    "\n",
    "\n",
    "**Referências:**\n",
    "\n",
    "Aulas no contexto de Análise Multivariada e Aprendizado Não-supervisionado (Profa. Cibele Russo):\n",
    "\n",
    "- Análise de Agrupamentos: https://youtu.be/zyLDAnQMnbo\n",
    "\n",
    "- Análise de Agrupamentos - Um exemplo passo a passo: https://youtu.be/Re97VX6ZhPA\n",
    "\n",
    "- Análise de Agrupamentos - Aplicação em Python: https://youtu.be/d_CJGaAbC7o \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NlfdmVSbpgOR"
   },
   "source": [
    "**Exercício**\n",
    "\n",
    "Analise as possíveis associações entre o sexo, idade, empresa, salário, saldo em conta corrente, saldo em conta poupança, saldo em investimento e devedor no cartão com a variável Inadimplente."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 920
    },
    "executionInfo": {
     "elapsed": 242749,
     "status": "ok",
     "timestamp": 1618514600867,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "NxO7--xZpgOL",
    "outputId": "fc00d7cf-e829-46e0-ce57-0e6bceefe7f3"
   },
   "outputs": [],
   "source": [
    "ax = sns.boxplot(x='Sexo', y='Salario', hue='Inadimplente', data=dados, palette='muted')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 920
    },
    "executionInfo": {
     "elapsed": 242749,
     "status": "ok",
     "timestamp": 1618514600867,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "NxO7--xZpgOL",
    "outputId": "fc00d7cf-e829-46e0-ce57-0e6bceefe7f3"
   },
   "outputs": [],
   "source": [
    "ax = sns.boxplot(x='Sexo', y='Saldo_cc', hue='Inadimplente', data=dados, palette='muted')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ax = sns.boxplot(x='Sexo', y='Devedor_cartao', hue='Inadimplente', data=dados, palette='muted')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dados.loc[dados['Inadimplente']==0, 'Inadimplente']= 'Não'\n",
    "dados.loc[dados['Inadimplente']==1, 'Inadimplente']= 'Sim'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.displot(dados, x='Devedor_cartao', col='Sexo', hue='Inadimplente', bins=30);\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pacote plotly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import plotly.express as px\n",
    "\n",
    "fig = px.histogram(dados, x='Salario')\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = px.histogram(dados, x='Salario', marginal='box')\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = px.box(dados, y='Salario', x='Sexo', color='Empresa')\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = px.scatter(dados, x='Saldo_cc', y='Salario', color='Empresa')\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Hm5JeQCKpgON"
   },
   "source": [
    "## AED de forma automatizada\n",
    "\n",
    "**Usar somente se estritamente necessário ou como uma análise inicial!!**\n",
    "\n",
    "Referência: https://medium.datadriveninvestor.com/10-python-automatic-eda-libraries-which-makes-data-scientist-life-easier-825d0a928570\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3711,
     "status": "ok",
     "timestamp": 1618515602070,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "8jZoos39pgON",
    "outputId": "19a90fa2-6589-465f-ae01-0d899693e7ad"
   },
   "outputs": [],
   "source": [
    "#!pip install pandas_profiling==3.0.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import pandas_profiling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#pandas_profiling.ProfileReport(dados)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_sb9jez5pgOQ"
   },
   "source": [
    "#### Exemplo: pacote sweetviz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 6439,
     "status": "ok",
     "timestamp": 1618514704961,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "No7r_mAQpgOQ",
    "outputId": "8e818796-c2d0-48a0-82b9-7330e3fc0ff2"
   },
   "outputs": [],
   "source": [
    "#!pip install sweetviz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 86,
     "referenced_widgets": [
      "28ec056a6ef44cdca59b6af57e4547ef",
      "b5a86d6fbfe247d6ae126bda5b2b047c",
      "55538afef7af477aa08d315a77d58f89",
      "be65d61ff842471f8afcd4bd19440277",
      "be6da7395cfd4ac090ed18f44622c39c",
      "d22d1245f7ff4cee884686ca66df708f",
      "4f5cd3fe1e02417b806d19456f727ea7",
      "12bb4bb32aa74bf7a636bea757e424fd",
      "52777e048c974aaabae1c75388eb6d31",
      "9c656e2fcc06466b8f82b61659c8ab24",
      "4be4c9aca3c54ebd867403b252dde372"
     ]
    },
    "executionInfo": {
     "elapsed": 9893,
     "status": "ok",
     "timestamp": 1618514929896,
     "user": {
      "displayName": "Cibele Russo",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gji-jMhjQu9TaA43mxVFn6ggWlzNZl0J-f8OMlZmw=s64",
      "userId": "07272141402608887263"
     },
     "user_tz": 180
    },
    "id": "dhinqsvppgOR",
    "outputId": "883f1228-1ca4-4201-d088-586685a3fb4c"
   },
   "outputs": [],
   "source": [
    "# import sweetviz library\n",
    "#import sweetviz as xx\n",
    "\n",
    "# analisando os dados do banco\n",
    "#study_report = xx.analyze(dados)\n",
    "\n",
    "# Gerar relatório\n",
    "#study_report.show_html('Dados_banco.html')\n"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "colab": {
   "collapsed_sections": [
    "8o-JyjQLpgNl",
    "zmF_eJCGpgNo"
   ],
   "name": "ECD-01-Material base - Visualização de dados.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "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": "3.9.5"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "12bb4bb32aa74bf7a636bea757e424fd": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "28ec056a6ef44cdca59b6af57e4547ef": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_55538afef7af477aa08d315a77d58f89",
       "IPY_MODEL_be65d61ff842471f8afcd4bd19440277",
       "IPY_MODEL_be6da7395cfd4ac090ed18f44622c39c"
      ],
      "layout": "IPY_MODEL_b5a86d6fbfe247d6ae126bda5b2b047c"
     }
    },
    "4be4c9aca3c54ebd867403b252dde372": {
     "model_module": "@jupyter-widgets/base",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4f5cd3fe1e02417b806d19456f727ea7": {
     "model_module": "@jupyter-widgets/base",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "52777e048c974aaabae1c75388eb6d31": {
     "model_module": "@jupyter-widgets/base",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": "2",
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "55538afef7af477aa08d315a77d58f89": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_4f5cd3fe1e02417b806d19456f727ea7",
      "placeholder": "​",
      "style": "IPY_MODEL_d22d1245f7ff4cee884686ca66df708f",
      "value": "Done! Use &#x27;show&#x27; commands to display/save.   "
     }
    },
    "9c656e2fcc06466b8f82b61659c8ab24": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "b5a86d6fbfe247d6ae126bda5b2b047c": {
     "model_module": "@jupyter-widgets/base",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": "inline-flex",
      "flex": null,
      "flex_flow": "row wrap",
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "100%"
     }
    },
    "be65d61ff842471f8afcd4bd19440277": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_52777e048c974aaabae1c75388eb6d31",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_12bb4bb32aa74bf7a636bea757e424fd",
      "value": 1
     }
    },
    "be6da7395cfd4ac090ed18f44622c39c": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_4be4c9aca3c54ebd867403b252dde372",
      "placeholder": "​",
      "style": "IPY_MODEL_9c656e2fcc06466b8f82b61659c8ab24",
      "value": " [100%]   00:00 -&gt; (00:00 left)"
     }
    },
    "d22d1245f7ff4cee884686ca66df708f": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}