regression_2_.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. 목적\n",
    "\n",
    "조개 나이 예측 -> 혹시 기후 변화 관련 ? \n",
    "\n",
    "### 2. 데이터셋\n",
    "\n",
    "이 데이터셋은 조개의 여러 특성에 관한 정보를 제공합니다. 각 열의 의미는 다음과 같습니다:\n",
    "\n",
    "1. **Sex**: 조개의 성별을 나타냅니다. 'M'은 수컷, 'F'는 암컷, 'I'는 유아기를 나타냅니다.\n",
    "2. **Length**: 조개의 가장 긴 부분의 길이를 나타냅니다.\n",
    "3. **Diameter**: 수직으로 측정한 조개의 직경을 나타냅니다.\n",
    "4. **Height**: 조개의 높이를 나타냅니다(조개껍질 포함).\n",
    "5. **Whole weight**: 조개 전체의 무게를 나타냅니다.\n",
    "6. **Shucked weight**: 조개의 살만을 제거한 무게를 나타냅니다.\n",
    "7. **Viscera weight**: 조개의 내장 무게를 나타냅니다.\n",
    "8. **Shell weight**: 조개껍질의 무게를 나타냅니다.\n",
    "9. **Rings**: 조개의 나이를 나타냅니다. 일반적으로, 링의 개수에 1.5를 더하면 조개의 나이(년)가 됩니다.\n",
    "\n",
    "각 행은 특정 조개의 위 특성을 나타냅니다. 이 데이터를 사용하여 조개의 특성과 나이 사이의 관계를 분석할 수 있습니다. 예를 들어, 회귀 분석을 사용하여 특정 특성(예: 무게, 길이 등)이 조개의 나이에 어떤 영향을 미치는지 예측할 수 있습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sklearn\n",
    "from sklearn.preprocessing import OneHotEncoder, StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_reg = pd.read_csv('./csv/Regression_data.csv')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### EDA 및 set split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Python311\\Lib\\site-packages\\sklearn\\preprocessing\\_encoders.py:972: FutureWarning: `sparse` was renamed to `sparse_output` in version 1.2 and will be removed in 1.4. `sparse_output` is ignored unless you leave `sparse` to its default value.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Length</th>\n",
       "      <th>Diameter</th>\n",
       "      <th>Height</th>\n",
       "      <th>Whole weight</th>\n",
       "      <th>Shucked weight</th>\n",
       "      <th>Viscera weight</th>\n",
       "      <th>Shell weight</th>\n",
       "      <th>Sex_Infant</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4038</th>\n",
       "      <td>0.550</td>\n",
       "      <td>0.445</td>\n",
       "      <td>0.125</td>\n",
       "      <td>0.6720</td>\n",
       "      <td>0.2880</td>\n",
       "      <td>0.1365</td>\n",
       "      <td>0.210</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1272</th>\n",
       "      <td>0.475</td>\n",
       "      <td>0.355</td>\n",
       "      <td>0.100</td>\n",
       "      <td>0.5035</td>\n",
       "      <td>0.2535</td>\n",
       "      <td>0.0910</td>\n",
       "      <td>0.140</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3384</th>\n",
       "      <td>0.305</td>\n",
       "      <td>0.225</td>\n",
       "      <td>0.070</td>\n",
       "      <td>0.1485</td>\n",
       "      <td>0.0585</td>\n",
       "      <td>0.0335</td>\n",
       "      <td>0.045</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3160</th>\n",
       "      <td>0.275</td>\n",
       "      <td>0.200</td>\n",
       "      <td>0.065</td>\n",
       "      <td>0.1165</td>\n",
       "      <td>0.0565</td>\n",
       "      <td>0.0130</td>\n",
       "      <td>0.035</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3894</th>\n",
       "      <td>0.495</td>\n",
       "      <td>0.380</td>\n",
       "      <td>0.135</td>\n",
       "      <td>0.6295</td>\n",
       "      <td>0.2630</td>\n",
       "      <td>0.1425</td>\n",
       "      <td>0.215</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Length  Diameter  Height  Whole weight  Shucked weight  Viscera weight   \n",
       "4038   0.550     0.445   0.125        0.6720          0.2880          0.1365  \\\n",
       "1272   0.475     0.355   0.100        0.5035          0.2535          0.0910   \n",
       "3384   0.305     0.225   0.070        0.1485          0.0585          0.0335   \n",
       "3160   0.275     0.200   0.065        0.1165          0.0565          0.0130   \n",
       "3894   0.495     0.380   0.135        0.6295          0.2630          0.1425   \n",
       "\n",
       "      Shell weight  Sex_Infant  \n",
       "4038         0.210         0.0  \n",
       "1272         0.140         0.0  \n",
       "3384         0.045         0.0  \n",
       "3160         0.035         0.0  \n",
       "3894         0.215         1.0  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Separate the features and the target\n",
    "X_reg = data_reg.drop(\"Rings\", axis=1)\n",
    "y_reg = data_reg[\"Rings\"]\n",
    "\n",
    "# Apply one-hot encoding to the 'Sex' feature\n",
    "encoder = OneHotEncoder(sparse=False, drop='first')\n",
    "sex_encoded = encoder.fit_transform(X_reg[['Sex']])\n",
    "sex_encoded_df = pd.DataFrame(sex_encoded, columns=['Sex_Female', 'Sex_Infant'])\n",
    "sex_encoded_df = sex_encoded_df.drop(\"Sex_Female\", axis=1)\n",
    "\n",
    "# Replace the 'Sex' column with the encoded columns\n",
    "X_reg = X_reg.drop(\"Sex\", axis=1)\n",
    "X_reg = pd.concat([X_reg, sex_encoded_df], axis=1)\n",
    "\n",
    "# Split the data into training and test sets\n",
    "X_train_reg, X_test_reg, y_train_reg, y_test_reg = train_test_split(X_reg, y_reg, test_size=0.2, random_state=42)\n",
    "\n",
    "X_train_reg.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x900 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Compute the correlation matrix\n",
    "corr = X_train_reg.corr()\n",
    "\n",
    "# Generate a mask for the upper triangle\n",
    "mask = np.triu(np.ones_like(corr, dtype=bool))\n",
    "\n",
    "# Set up the matplotlib figure\n",
    "f, ax = plt.subplots(figsize=(11, 9))\n",
    "\n",
    "# Generate a custom diverging colormap\n",
    "cmap = sns.diverging_palette(230, 20, as_cmap=True)\n",
    "\n",
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "# Here, we will not use the mask, and add the annotation to display the correlation scores\n",
    "sns.heatmap(corr, cmap=cmap, vmax=.3, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5}, annot=True)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 회귀모델학습"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "평균 제곱 오차 (Mean Squared Error, MSE): 5.03023912355076\n",
      "평균 제곱근 오차 (Root Mean Squared Error, RMSE): 2.242819458527761\n",
      "평균 절대 오차 (Mean Absolute Error, MAE): 1.624183637598811\n",
      "결정 계수 (R^2 Score): 0.5353217995418885\n"
     ]
    }
   ],
   "source": [
    "# Initialize a linear regression model\n",
    "model_reg = LinearRegression()\n",
    "\n",
    "# Fit the model to the training data\n",
    "model_reg.fit(X_train_reg, y_train_reg)\n",
    "\n",
    "# Use the model to make predictions on the test data\n",
    "y_pred_reg = model_reg.predict(X_test_reg)\n",
    "\n",
    "# Calculate the performance metrics\n",
    "mse = mean_squared_error(y_test_reg, y_pred_reg)\n",
    "rmse = np.sqrt(mse)\n",
    "mae = mean_absolute_error(y_test_reg, y_pred_reg)\n",
    "r2 = r2_score(y_test_reg, y_pred_reg)\n",
    "\n",
    "print(\"평균 제곱 오차 (Mean Squared Error, MSE):\", mse)\n",
    "print(\"평균 제곱근 오차 (Root Mean Squared Error, RMSE):\", rmse)\n",
    "print(\"평균 절대 오차 (Mean Absolute Error, MAE):\", mae)\n",
    "print(\"결정 계수 (R^2 Score):\", r2)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 평균 제곱 오차 (MSE)와 평균 제곱근 오차 (RMSE)는 실제 값과 예측 값의 차이를 제곱하여 평균한 값입니다. 이 값이 작을수록 모델의 성능이 좋다고 할 수 있습니다.\n",
    "\n",
    "- 평균 절대 오차 (MAE)는 실제 값과 예측 값의 차이를 절대값으로 변환하여 평균한 값입니다. 이 값도 작을수록 모델의 성능이 좋다고 할 수 있습니다.\n",
    "\n",
    "- 결정 계수 (R^2 Score)는 회귀 모델이 데이터를 얼마나 잘 설명하는지를 나타내는 지표입니다. 이 값이 1에 가까울수록 모델이 데이터를 잘 설명하고 있다고 할 수 있습니다."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 라쏘 회귀모델 학습"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "평균 제곱 오차 (Mean Squared Error, MSE): 5.044325431130134\n",
      "평균 제곱근 오차 (Root Mean Squared Error, RMSE): 2.242819458527761\n",
      "평균 절대 오차 (Mean Absolute Error, MAE): 1.627586516815485\n",
      "결정 계수 (R^2 Score): 0.5340205492640563\n"
     ]
    }
   ],
   "source": [
    "# Initialize a lasso regression model\n",
    "lasso_reg = Lasso()\n",
    "\n",
    "parameters = {'alpha': [0.001, 0.01, 0.1, 1, 10, 100, 1000]}\n",
    "\n",
    "\n",
    "# Perform grid search\n",
    "grid_search = GridSearchCV(lasso_reg, parameters, cv=5)\n",
    "grid_search.fit(X_train_reg, y_train_reg)\n",
    "\n",
    "# Get the best model\n",
    "best_model = grid_search.best_estimator_\n",
    "\n",
    "# Use the model to make predictions on the test data\n",
    "y_pred_reg = best_model.predict(X_test_reg)\n",
    "\n",
    "# Calculate the performance metrics\n",
    "lasso_mse = mean_squared_error(y_test_reg, y_pred_reg)\n",
    "lasso_rmse = np.sqrt(mse)\n",
    "lasso_mae = mean_absolute_error(y_test_reg, y_pred_reg)\n",
    "lasso_r2 = r2_score(y_test_reg, y_pred_reg)\n",
    "\n",
    "print(\"평균 제곱 오차 (Mean Squared Error, MSE):\", lasso_mse)\n",
    "print(\"평균 제곱근 오차 (Root Mean Squared Error, RMSE):\", lasso_rmse)\n",
    "print(\"평균 절대 오차 (Mean Absolute Error, MAE):\", lasso_mae)\n",
    "print(\"결정 계수 (R^2 Score):\", lasso_r2)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 특성 스케일링 - 표준화 후 성능 메트릭"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5.03144393389456, 2.2430880352528653, 1.624675245513548, 0.5352105028244365)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Initialize a standard scaler\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Fit the scaler to the training data and transform the data\n",
    "X_train_scaled = scaler.fit_transform(X_train_reg)\n",
    "\n",
    "# Transform the test data\n",
    "X_test_scaled = scaler.transform(X_test_reg)\n",
    "\n",
    "# Perform grid search with the scaled data\n",
    "grid_search = GridSearchCV(lasso_reg, parameters, cv=5)\n",
    "grid_search.fit(X_train_scaled, y_train_reg)\n",
    "\n",
    "# Get the best model\n",
    "best_model = grid_search.best_estimator_\n",
    "\n",
    "# Use the model to make predictions on the test data\n",
    "y_pred_reg = best_model.predict(X_test_scaled)\n",
    "\n",
    "# Calculate the performance metrics\n",
    "mse = mean_squared_error(y_test_reg, y_pred_reg)\n",
    "rmse = np.sqrt(mse)\n",
    "mae = mean_absolute_error(y_test_reg, y_pred_reg)\n",
    "r2 = r2_score(y_test_reg, y_pred_reg)\n",
    "\n",
    "mse, rmse, mae, r2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}