{
 "nbformat": 4,
 "nbformat_minor": 5,
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3"
  }
 },
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Regressão linear — Diabetes\n",
    "\n",
    "Material introdutório — Prof. Alex Souza\n",
    "\n",
    "Pergunta: qual o valor quantitativo de progressão da doença após um ano? Regressão prevê um número. Para tornar a primeira explicação visual, usaremos apenas o IMC (bmi) como entrada. A regressão linear ajusta uma reta minimizando a soma dos erros ao quadrado. É um exemplo didático de associação, sem aplicação clínica."
   ],
   "id": "celula-000"
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Como executar\n",
    "Abra no Jupyter ou Google Colab e execute as células na ordem. As bases vêm no scikit-learn; não há download externo de dados. O CSV correspondente também acompanha este material. Instale as bibliotecas apenas se necessário."
   ],
   "id": "celula-001"
  },
  {
   "cell_type": "code",
   "metadata": {},
   "source": [
    "# Execute esta linha se as bibliotecas não estiverem instaladas.\n",
    "# %pip install scikit-learn pandas matplotlib\n",
    "import sklearn\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "print(\"Versão do scikit-learn:\", sklearn.__version__)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Versão do scikit-learn: 1.8.0\n"
     ]
    }
   ],
   "execution_count": 1,
   "id": "celula-002"
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fonte e dicionário de dados\n",
    "\n",
    "442 pacientes, dez atributos e um alvo. As entradas carregadas por padrão já estão centradas e escaladas pelo scikit-learn; bmi não está em kg/m² neste CSV.\n",
    "\n",
    "| Coluna | Descrição |\n",
    "|---|---|\n",
    "| age | Idade |\n",
    "| sex | Variável de sexo codificada |\n",
    "| bmi | Índice de massa corporal |\n",
    "| bp | Pressão arterial média |\n",
    "| s1 | Colesterol sérico total |\n",
    "| s2 | Lipoproteínas de baixa densidade |\n",
    "| s3 | Lipoproteínas de alta densidade |\n",
    "| s4 | Razão colesterol total/HDL |\n",
    "| s5 | Logaritmo dos triglicerídeos |\n",
    "| s6 | Glicemia |\n",
    "| target | Medida quantitativa de progressão após um ano |\n",
    "\n",
    "A documentação oficial ressalva ambiguidades históricas na descrição de alguns atributos. O alvo não é probabilidade, diagnóstico ou preço."
   ],
   "id": "celula-003"
  },
  {
   "cell_type": "code",
   "metadata": {},
   "source": [
    "from sklearn.datasets import load_diabetes\n",
    "# Carrega os dados locais distribuídos com o scikit-learn.\n",
    "base = load_diabetes(as_frame=True)\n",
    "dados = base.frame.copy()\n",
    "# Consulte também o CSV incluído: pd.read_csv(\"02_regressao_diabetes.csv\")\n",
    "print(\"Linhas e colunas:\", dados.shape)\n",
    "print(dados.head().to_string())\n",
    "print(\"Ausentes por coluna:\\n\", dados.isna().sum())"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Linhas e colunas: (442, 11)\n",
      "        age       sex       bmi        bp        s1        s2        s3        s4        s5        s6  target\n",
      "0  0.038076  0.050680  0.061696  0.021872 -0.044223 -0.034821 -0.043401 -0.002592  0.019907 -0.017646   151.0\n",
      "1 -0.001882 -0.044642 -0.051474 -0.026328 -0.008449 -0.019163  0.074412 -0.039493 -0.068332 -0.092204    75.0\n",
      "2  0.085299  0.050680  0.044451 -0.005670 -0.045599 -0.034194 -0.032356 -0.002592  0.002861 -0.025930   141.0\n",
      "3 -0.089063 -0.044642 -0.011595 -0.036656  0.012191  0.024991 -0.036038  0.034309  0.022688 -0.009362   206.0\n",
      "4  0.005383 -0.044642 -0.036385  0.021872  0.003935  0.015596  0.008142 -0.002592 -0.031988 -0.046641   135.0\n",
      "Ausentes por coluna:\n",
      " age       0\n",
      "sex       0\n",
      "bmi       0\n",
      "bp        0\n",
      "s1        0\n",
      "s2        0\n",
      "s3        0\n",
      "s4        0\n",
      "s5        0\n",
      "s6        0\n",
      "target    0\n",
      "dtype: int64\n"
     ]
    }
   ],
   "execution_count": 2,
   "id": "celula-004"
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Descrição original da fonte\n",
    "A descrição abaixo preserva as referências e observações distribuídas com a base."
   ],
   "id": "celula-005"
  },
  {
   "cell_type": "code",
   "metadata": {},
   "source": [
    "print(base.DESCR)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      ".. _diabetes_dataset:\n",
      "\n",
      "Diabetes dataset\n",
      "----------------\n",
      "\n",
      "Ten baseline variables, age, sex, body mass index, average blood\n",
      "pressure, and six blood serum measurements were obtained for each of n =\n",
      "442 diabetes patients, as well as the response of interest, a\n",
      "quantitative measure of disease progression one year after baseline.\n",
      "\n",
      "**Data Set Characteristics:**\n",
      "\n",
      ":Number of Instances: 442\n",
      "\n",
      ":Number of Attributes: First 10 columns are numeric predictive values\n",
      "\n",
      ":Target: Column 11 is a quantitative measure of disease progression one year after baseline\n",
      "\n",
      ":Attribute Information:\n",
      "    - age     age in years\n",
      "    - sex\n",
      "    - bmi     body mass index\n",
      "    - bp      average blood pressure\n",
      "    - s1      tc, total serum cholesterol\n",
      "    - s2      ldl, low-density lipoproteins\n",
      "    - s3      hdl, high-density lipoproteins\n",
      "    - s4      tch, total cholesterol / HDL\n",
      "    - s5      ltg, possibly log of serum triglycerides level\n",
      "    - s6      glu, blood sugar level\n",
      "\n",
      "Note: Each of these 10 feature variables have been mean centered and scaled by the standard deviation times the square root of `n_samples` (i.e. the sum of squares of each column totals 1).\n",
      "\n",
      "Source URL:\n",
      "https://www4.stat.ncsu.edu/~boos/var.select/diabetes.html\n",
      "\n",
      "For more information see:\n",
      "Bradley Efron, Trevor Hastie, Iain Johnstone and Robert Tibshirani (2004) \"Least Angle Regression,\" Annals of Statistics (with discussion), 407-499.\n",
      "(https://web.stanford.edu/~hastie/Papers/LARS/LeastAngle_2002.pdf)\n",
      "\n"
     ]
    }
   ],
   "execution_count": 3,
   "id": "celula-006"
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Treinar e comparar\n",
    "MAE mede o erro absoluto médio, na escala do alvo. R² compara o ajuste com uma referência baseada na média e pode ser negativo. O modelo de referência usa somente a média do treino."
   ],
   "id": "celula-007"
  },
  {
   "cell_type": "code",
   "metadata": {},
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.dummy import DummyRegressor\n",
    "from sklearn.metrics import mean_absolute_error, r2_score\n",
    "# Colchetes duplos mantêm X como tabela com uma coluna.\n",
    "X, y = base.data[['bmi']], base.target\n",
    "X_treino, X_teste, y_treino, y_teste = train_test_split(X,y,test_size=0.2,random_state=42)\n",
    "modelo = LinearRegression()\n",
    "modelo.fit(X_treino,y_treino)\n",
    "previsoes = modelo.predict(X_teste)\n",
    "baseline = DummyRegressor(strategy='mean').fit(X_treino,y_treino)\n",
    "print('MAE da média:', mean_absolute_error(y_teste,baseline.predict(X_teste)))\n",
    "print('MAE da regressão:', mean_absolute_error(y_teste,previsoes))\n",
    "print('R²:', r2_score(y_teste,previsoes))\n",
    "print('Intercepto:',modelo.intercept_, 'Coeficiente:',modelo.coef_[0])"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "MAE da média: 64.00646146990485\n",
      "MAE da regressão: 52.259976445345536\n",
      "R²: 0.23335039815872138\n",
      "Intercepto: 152.00335421448167 Coeficiente: 998.5776891375598\n"
     ]
    }
   ],
   "execution_count": 4,
   "id": "celula-008"
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualizar a reta\n",
    "Pontos representam resultados observados; a reta representa previsões. O coeficiente se refere ao IMC na escala transformada da base. Dispersão ao redor da reta indica que o IMC sozinho não explica toda a variação. Associação não comprova causalidade.\n",
    "\n",
    "Extensão: substitua a entrada por todas as dez colunas e compare o MAE mantendo a divisão. Se repetir ajustes para escolher o melhor modelo, use validação cruzada no treino e preserve um teste final."
   ],
   "id": "celula-009"
  },
  {
   "cell_type": "code",
   "metadata": {},
   "source": [
    "# Ordena o eixo para desenhar uma linha contínua.\n",
    "ordem = X_teste['bmi'].argsort()\n",
    "plt.figure(figsize=(8,5))\n",
    "plt.scatter(X_teste['bmi'],y_teste,alpha=0.7,label='Observado')\n",
    "plt.plot(X_teste['bmi'].iloc[ordem],previsoes[ordem],color='red',label='Previsão')\n",
    "plt.xlabel('IMC transformado'); plt.ylabel('Progressão após um ano')\n",
    "plt.legend(); plt.title('Regressão linear no conjunto de teste')"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "metadata": {},
     "data": {
      "image/png": "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"
     }
    }
   ],
   "execution_count": 5,
   "id": "celula-010"
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Discussão em sala\n",
    "Explique com suas palavras a pergunta respondida, as entradas utilizadas e o significado da saída. Que informação adicional seria necessária para usar esse modelo em uma organização?\n",
    "\n",
    "## Referências\n",
    "- [Base e documentação oficial](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_diabetes.html)\n",
    "- [Guia de aprendizado de máquina](https://scikit-learn.org/stable/user_guide.html)\n",
    "\n",
    "Dados pequenos servem para ensino; resultados não garantem desempenho em produção."
   ],
   "id": "celula-011"
  }
 ]
}