{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "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>title</th>\n",
       "      <th>source</th>\n",
       "      <th>title_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Iranian President Raisi is killed in helicopte...</td>\n",
       "      <td>nbc</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Kristen Cavallari and Jay Cutler to divorce af...</td>\n",
       "      <td>nbc</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Why Atlanta spa shooters Asian acquaintances c...</td>\n",
       "      <td>nbc</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>The best TV streaming services in 2024</td>\n",
       "      <td>nbc</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Mike Johnson wont commit to bringing House bac...</td>\n",
       "      <td>nbc</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353917</th>\n",
       "      <td>Statins not linked to suicide risk</td>\n",
       "      <td>nbc</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353918</th>\n",
       "      <td>Mono may increase risk of Hodgkins</td>\n",
       "      <td>nbc</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353919</th>\n",
       "      <td>Herpes virus tied to rare lung disease</td>\n",
       "      <td>nbc</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353920</th>\n",
       "      <td>Scan reveals hidden life of fetuses</td>\n",
       "      <td>nbc</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353921</th>\n",
       "      <td>Camp helps kids with cancer live</td>\n",
       "      <td>nbc</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>353922 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    title source  title_length\n",
       "0       Iranian President Raisi is killed in helicopte...    nbc             8\n",
       "1       Kristen Cavallari and Jay Cutler to divorce af...    nbc            11\n",
       "2       Why Atlanta spa shooters Asian acquaintances c...    nbc            14\n",
       "3                  The best TV streaming services in 2024    nbc             7\n",
       "4       Mike Johnson wont commit to bringing House bac...    nbc            15\n",
       "...                                                   ...    ...           ...\n",
       "353917                 Statins not linked to suicide risk    nbc             6\n",
       "353918                 Mono may increase risk of Hodgkins    nbc             6\n",
       "353919             Herpes virus tied to rare lung disease    nbc             7\n",
       "353920                Scan reveals hidden life of fetuses    nbc             6\n",
       "353921                   Camp helps kids with cancer live    nbc             6\n",
       "\n",
       "[353922 rows x 3 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load data\n",
    "df = pd.read_csv('combined_data.csv')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "source\n",
       "foxnews    233726\n",
       "nbc        120196\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check the source balance\n",
    "df['source'].value_counts()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Splitting the data...\n",
      "Vectorizing text using TF-IDF...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Fitting TF-IDF on train set: 100%|██████████| 283137/283137 [00:01<00:00, 262450.64it/s]\n",
      "Transforming TF-IDF on test set: 100%|██████████| 70785/70785 [00:00<00:00, 255003.82it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Encoding target labels...\n"
     ]
    }
   ],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.metrics import classification_report, accuracy_score\n",
    "from tqdm import tqdm\n",
    "import pandas as pd\n",
    "\n",
    "# Enable tqdm to show progress for Pandas operations\n",
    "tqdm.pandas()\n",
    "\n",
    "# Assuming data is already loaded into df\n",
    "X = df['title']\n",
    "y = df['source']\n",
    "\n",
    "# Step 1: Split the data\n",
    "print(\"Splitting the data...\")\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "# Step 2: Vectorize the text using TF-IDF\n",
    "print(\"Vectorizing text using TF-IDF...\")\n",
    "vectorizer = TfidfVectorizer()\n",
    "X_train = vectorizer.fit_transform(tqdm(X_train, desc=\"Fitting TF-IDF on train set\"))\n",
    "X_test = vectorizer.transform(tqdm(X_test, desc=\"Transforming TF-IDF on test set\"))\n",
    "\n",
    "# Step 3: Encode the target labels\n",
    "print(\"Encoding target labels...\")\n",
    "encoder = LabelEncoder()\n",
    "y_train = encoder.fit_transform(y_train)\n",
    "y_test = encoder.transform(y_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((283137, 69008), (283137,), (70785, 69008), (70785,))"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, y_train.shape, X_test.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training and Evaluating Models:   0%|          | 0/3 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=== Logistic Regression ===\n",
      "\n",
      "Training Logistic Regression...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training and Evaluating Models:  33%|███▎      | 1/3 [00:01<00:03,  1.90s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Evaluating Logistic Regression...\n",
      "Classification Report:\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.82      0.90      0.86     46733\n",
      "           1       0.76      0.62      0.68     24052\n",
      "\n",
      "    accuracy                           0.80     70785\n",
      "   macro avg       0.79      0.76      0.77     70785\n",
      "weighted avg       0.80      0.80      0.80     70785\n",
      "\n",
      "Accuracy: 0.8032775305502579\n",
      "\n",
      "=== Random Forest ===\n",
      "\n",
      "Training Random Forest...\n",
      "\n",
      "Evaluating Random Forest...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training and Evaluating Models:  67%|██████▋   | 2/3 [07:46<04:34, 274.03s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report:\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.78      0.93      0.85     46733\n",
      "           1       0.79      0.50      0.61     24052\n",
      "\n",
      "    accuracy                           0.79     70785\n",
      "   macro avg       0.79      0.72      0.73     70785\n",
      "weighted avg       0.79      0.79      0.77     70785\n",
      "\n",
      "Accuracy: 0.7864236773327682\n",
      "\n",
      "=== XGBoost ===\n",
      "\n",
      "Training XGBoost...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [13:53:23] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "Training and Evaluating Models: 100%|██████████| 3/3 [08:00<00:00, 160.05s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Evaluating XGBoost...\n",
      "Classification Report:\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.74      0.95      0.83     46733\n",
      "           1       0.77      0.35      0.48     24052\n",
      "\n",
      "    accuracy                           0.74     70785\n",
      "   macro avg       0.75      0.65      0.65     70785\n",
      "weighted avg       0.75      0.74      0.71     70785\n",
      "\n",
      "Accuracy: 0.7416966871512326\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import classification_report, accuracy_score\n",
    "from tqdm import tqdm\n",
    "\n",
    "# Initialize Models\n",
    "models = {\n",
    "    \"Logistic Regression\": LogisticRegression(max_iter=1000, random_state=42),\n",
    "    \"Random Forest\": RandomForestClassifier(random_state=42),\n",
    "    \"XGBoost\": XGBClassifier(use_label_encoder=False, eval_metric='logloss', random_state=42),\n",
    "}\n",
    "\n",
    "# Train and Evaluate Models\n",
    "results = []\n",
    "\n",
    "for model_name, model in tqdm(models.items(), desc=\"Training and Evaluating Models\"):\n",
    "    print(f\"\\n=== {model_name} ===\")\n",
    "    \n",
    "    # Training\n",
    "    print(f\"\\nTraining {model_name}...\")\n",
    "    model.fit(X_train, y_train)\n",
    "    \n",
    "    # Evaluation\n",
    "    print(f\"\\nEvaluating {model_name}...\")\n",
    "    y_pred = model.predict(X_test)\n",
    "    report = classification_report(y_test, y_pred)\n",
    "    accuracy = accuracy_score(y_test, y_pred)\n",
    "    \n",
    "    print(\"Classification Report:\\n\", report)\n",
    "    print(\"Accuracy:\", accuracy)\n",
    "    \n",
    "    results.append({\n",
    "        \"Model\": model_name,\n",
    "        \"Classification Report\": report,\n",
    "        \"Accuracy\": accuracy\n",
    "    })\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression\n",
      "Fitting 5 folds for each of 10 candidates, totalling 50 fits\n",
      "[CV] END ...............................C=0.01, solver=lbfgs; total time=   0.4s\n",
      "[CV] END ...............................C=0.01, solver=lbfgs; total time=   0.4s\n",
      "[CV] END ...............................C=0.01, solver=lbfgs; total time=   0.4s\n",
      "[CV] END ...............................C=0.01, solver=lbfgs; total time=   0.5s\n",
      "[CV] END ...........................C=0.01, solver=liblinear; total time=   0.4s\n",
      "[CV] END ...........................C=0.01, solver=liblinear; total time=   0.4s\n",
      "[CV] END ...........................C=0.01, solver=liblinear; total time=   0.4s\n",
      "[CV] END ...........................C=0.01, solver=liblinear; total time=   0.4s\n",
      "[CV] END ...............................C=0.01, solver=lbfgs; total time=   0.5s\n",
      "[CV] END ...........................C=0.01, solver=liblinear; total time=   0.5s\n",
      "[CV] END ................................C=0.1, solver=lbfgs; total time=   0.9s\n",
      "[CV] END ................................C=0.1, solver=lbfgs; total time=   0.9s\n",
      "[CV] END ............................C=0.1, solver=liblinear; total time=   0.8s\n",
      "[CV] END ................................C=0.1, solver=lbfgs; total time=   1.0s\n",
      "[CV] END ................................C=0.1, solver=lbfgs; total time=   1.2s\n",
      "[CV] END ............................C=0.1, solver=liblinear; total time=   0.9s\n",
      "[CV] END ................................C=0.1, solver=lbfgs; total time=   1.1s\n",
      "[CV] END ............................C=0.1, solver=liblinear; total time=   0.9s\n",
      "[CV] END ............................C=0.1, solver=liblinear; total time=   0.9s\n",
      "[CV] END ............................C=0.1, solver=liblinear; total time=   0.9s\n",
      "[CV] END ..............................C=1, solver=liblinear; total time=   1.5s\n",
      "[CV] END ..............................C=1, solver=liblinear; total time=   1.6s\n",
      "[CV] END ..............................C=1, solver=liblinear; total time=   1.7s\n",
      "[CV] END ..............................C=1, solver=liblinear; total time=   1.6s\n",
      "[CV] END ..............................C=1, solver=liblinear; total time=   1.6s\n",
      "[CV] END ..................................C=1, solver=lbfgs; total time=   2.2s\n",
      "[CV] END ..................................C=1, solver=lbfgs; total time=   2.9s\n",
      "[CV] END ..................................C=1, solver=lbfgs; total time=   2.1s\n",
      "[CV] END ..................................C=1, solver=lbfgs; total time=   2.3s\n",
      "[CV] END ..................................C=1, solver=lbfgs; total time=   3.1s\n",
      "[CV] END .................................C=10, solver=lbfgs; total time=   4.6s\n",
      "[CV] END .............................C=10, solver=liblinear; total time=   3.0s\n",
      "[CV] END .............................C=10, solver=liblinear; total time=   3.0s\n",
      "[CV] END .............................C=10, solver=liblinear; total time=   3.0s\n",
      "[CV] END .............................C=10, solver=liblinear; total time=   3.0s\n",
      "[CV] END .................................C=10, solver=lbfgs; total time=   5.0s\n",
      "[CV] END .............................C=10, solver=liblinear; total time=   3.0s\n",
      "[CV] END .................................C=10, solver=lbfgs; total time=   4.8s\n",
      "[CV] END .................................C=10, solver=lbfgs; total time=   5.2s\n",
      "[CV] END .................................C=10, solver=lbfgs; total time=   5.9s\n",
      "[CV] END ................................C=100, solver=lbfgs; total time=   5.3s\n",
      "[CV] END ................................C=100, solver=lbfgs; total time=   3.5s\n",
      "[CV] END ................................C=100, solver=lbfgs; total time=   6.2s\n",
      "[CV] END ............................C=100, solver=liblinear; total time=   4.8s\n",
      "[CV] END ............................C=100, solver=liblinear; total time=   4.9s\n",
      "[CV] END ............................C=100, solver=liblinear; total time=   5.0s\n",
      "[CV] END ................................C=100, solver=lbfgs; total time=   5.4s\n",
      "[CV] END ............................C=100, solver=liblinear; total time=   5.1s\n",
      "[CV] END ............................C=100, solver=liblinear; total time=   4.5s\n",
      "[CV] END ................................C=100, solver=lbfgs; total time=   6.7s\n",
      "Best parameters for Logistic Regression: {'C': 10, 'solver': 'liblinear'}\n",
      "0.8079819170728262\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     foxnews       0.84      0.88      0.86     46733\n",
      "         nbc       0.74      0.66      0.70     24052\n",
      "\n",
      "    accuracy                           0.81     70785\n",
      "   macro avg       0.79      0.77      0.78     70785\n",
      "weighted avg       0.80      0.81      0.81     70785\n",
      "\n",
      "Random Forest\n",
      "Fitting 5 folds for each of 36 candidates, totalling 180 fits\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time=11.4min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/joblib/externals/loky/process_executor.py:752: UserWarning: A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time=12.6min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time=12.5min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time=12.8min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time=13.0min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time=10.9min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time=11.3min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time=24.9min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time=25.1min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time=25.8min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time=26.0min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time=26.3min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time=11.1min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time=10.6min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time=11.9min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time=21.3min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time=21.3min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time=23.3min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time=48.6min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time=23.6min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time=51.0min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time=22.1min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time=10.5min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time= 9.8min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time=48.8min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time= 9.8min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time=49.4min\n",
      "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time=49.5min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time=10.9min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time=10.5min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time=44.5min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time=20.2min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time=21.1min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time=22.5min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time=45.8min\n",
      "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time=  22.8s\n",
      "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time=  20.0s\n",
      "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time=  19.9s\n",
      "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time=  20.2s\n",
      "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time=  23.9s\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time=  39.9s\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time=  42.2s\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time=22.4min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time=22.3min\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time=  39.0s\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time=  42.6s\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time=  42.8s\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 1.2min\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 1.2min\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 1.4min\n",
      "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time=  18.4s\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 1.4min\n",
      "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time=  20.9s\n",
      "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time=  19.8s\n",
      "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time=  21.0s\n",
      "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 1.4min\n",
      "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time=  20.7s\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time=  37.8s\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time=  41.4s\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time=  38.7s\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time=  35.6s\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time=  32.6s\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 1.4min\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 1.3min\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 1.4min\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 1.3min\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time=  21.9s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time=  22.1s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time=  18.4s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time=  20.1s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time=  20.6s\n",
      "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 1.4min\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time=  39.4s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time=  41.9s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time=  40.9s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time=  39.4s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time=  36.8s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 1.4min\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 1.4min\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 1.3min\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 1.4min\n",
      "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time=  41.0s\n",
      "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time=  40.4s\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time=46.4min\n",
      "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time=  42.7s\n",
      "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 1.3min\n",
      "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time=  40.9s\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time=46.8min\n",
      "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time=  38.5s\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 1.4min\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 1.4min\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 1.4min\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 1.3min\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 1.3min\n",
      "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time=  37.4s\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 2.5min\n",
      "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time=48.8min\n",
      "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time=  39.3s\n",
      "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time=  33.6s\n",
      "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time=  38.6s\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 2.6min\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 2.7min\n",
      "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time=  38.7s\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 2.7min\n",
      "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 2.5min\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 1.2min\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 1.3min\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 1.2min\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 1.3min\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 1.3min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time=  38.7s\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time=  39.1s\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time=  34.2s\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 2.6min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time=  38.7s\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 2.6min\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 2.3min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time=  35.7s\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 2.5min\n",
      "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 2.6min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 1.2min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 1.3min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 1.3min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 1.1min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 1.2min\n",
      "[CV] END .max_depth=30, min_samples_split=2, n_estimators=50; total time=  50.0s\n",
      "[CV] END .max_depth=30, min_samples_split=2, n_estimators=50; total time=  55.3s\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 2.3min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 2.4min\n",
      "[CV] END .max_depth=30, min_samples_split=2, n_estimators=50; total time=  58.0s\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 2.3min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 2.3min\n",
      "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 2.3min\n",
      "[CV] END .max_depth=30, min_samples_split=2, n_estimators=50; total time=  53.3s\n",
      "[CV] END .max_depth=30, min_samples_split=2, n_estimators=50; total time=  57.7s\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=100; total time= 1.8min\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=100; total time= 1.6min\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=100; total time= 1.9min\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=100; total time= 1.9min\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=100; total time= 1.9min\n",
      "[CV] END .max_depth=30, min_samples_split=5, n_estimators=50; total time=  52.5s\n",
      "[CV] END .max_depth=30, min_samples_split=5, n_estimators=50; total time=  56.7s\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=200; total time= 3.7min\n",
      "[CV] END .max_depth=30, min_samples_split=5, n_estimators=50; total time=  54.9s\n",
      "[CV] END .max_depth=30, min_samples_split=5, n_estimators=50; total time=  55.8s\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=200; total time= 3.5min\n",
      "[CV] END .max_depth=30, min_samples_split=5, n_estimators=50; total time=  51.0s\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=200; total time= 3.6min\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=200; total time= 3.5min\n",
      "[CV] END max_depth=30, min_samples_split=2, n_estimators=200; total time= 3.5min\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=100; total time= 1.7min\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=100; total time= 1.8min\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=100; total time= 1.8min\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=100; total time= 1.7min\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=100; total time= 1.8min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time=44.3min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=50; total time=  54.1s\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=50; total time=  54.0s\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time=43.4min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=50; total time=  57.6s\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=50; total time=  54.2s\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=200; total time= 3.3min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=50; total time=  50.9s\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=200; total time= 3.8min\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=200; total time= 3.5min\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=200; total time= 3.7min\n",
      "[CV] END max_depth=30, min_samples_split=5, n_estimators=200; total time= 3.7min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=100; total time= 1.7min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=100; total time= 1.8min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=100; total time= 1.7min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=100; total time= 1.7min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=100; total time= 1.7min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=200; total time= 2.8min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=200; total time= 3.2min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=200; total time= 3.0min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=200; total time= 3.0min\n",
      "[CV] END max_depth=30, min_samples_split=10, n_estimators=200; total time= 3.0min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time=41.6min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time=40.4min\n",
      "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time=40.6min\n",
      "Best parameters for Random Forest: {'max_depth': None, 'min_samples_split': 5, 'n_estimators': 200}\n",
      "0.7878646605919333\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     foxnews       0.79      0.93      0.85     46733\n",
      "         nbc       0.79      0.51      0.62     24052\n",
      "\n",
      "    accuracy                           0.79     70785\n",
      "   macro avg       0.79      0.72      0.74     70785\n",
      "weighted avg       0.79      0.79      0.77     70785\n",
      "\n",
      "XGBoost\n",
      "Fitting 5 folds for each of 27 candidates, totalling 135 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.01, max_depth=3, n_estimators=50; total time=   9.0s\n",
      "[CV] END ...learning_rate=0.01, max_depth=3, n_estimators=50; total time=   9.1s\n",
      "[CV] END ...learning_rate=0.01, max_depth=3, n_estimators=50; total time=   9.2s\n",
      "[CV] END ...learning_rate=0.01, max_depth=3, n_estimators=50; total time=   9.2s\n",
      "[CV] END ...learning_rate=0.01, max_depth=3, n_estimators=50; total time=   9.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=100; total time=  16.5s\n",
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=100; total time=  16.6s\n",
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=100; total time=  16.7s\n",
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=100; total time=  16.8s\n",
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=100; total time=  16.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:23] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:24] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:24] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:24] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:24] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.01, max_depth=5, n_estimators=50; total time=  20.1s\n",
      "[CV] END ...learning_rate=0.01, max_depth=5, n_estimators=50; total time=  20.3s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:36] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:36] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=200; total time=  31.3s\n",
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=200; total time=  31.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:38] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:38] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.01, max_depth=5, n_estimators=50; total time=  20.3s\n",
      "[CV] END ...learning_rate=0.01, max_depth=5, n_estimators=50; total time=  20.6s\n",
      "[CV] END ...learning_rate=0.01, max_depth=5, n_estimators=50; total time=  20.9s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:44] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:44] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:45] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=200; total time=  31.8s\n",
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=200; total time=  32.0s\n",
      "[CV] END ..learning_rate=0.01, max_depth=3, n_estimators=200; total time=  32.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:48] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:48] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:22:48] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=100; total time=  39.5s\n",
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=100; total time=  39.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:03] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:03] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=100; total time=  39.7s\n",
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=100; total time=  39.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=100; total time=  39.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:18] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.01, max_depth=7, n_estimators=50; total time=  46.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:34] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.01, max_depth=7, n_estimators=50; total time=  47.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:35] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.01, max_depth=7, n_estimators=50; total time=  47.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:50] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.01, max_depth=7, n_estimators=50; total time=  48.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:52] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=200; total time= 1.3min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:23:56] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=200; total time= 1.3min\n",
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=200; total time= 1.3min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:01] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=200; total time= 1.3min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:02] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:02] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=5, n_estimators=200; total time= 1.3min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:05] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.01, max_depth=7, n_estimators=50; total time=  53.9s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:10] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=3, n_estimators=50; total time=   9.9s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:15] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=3, n_estimators=50; total time=   9.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:20] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=3, n_estimators=50; total time=   9.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:25] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=3, n_estimators=50; total time=   9.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:29] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=3, n_estimators=50; total time=   9.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:34] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=100; total time=  18.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:47] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=100; total time=  17.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:24:52] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=100; total time=  17.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:06] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=100; total time=  18.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:10] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=100; total time= 1.9min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:11] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=100; total time=  18.9s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:24] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=100; total time= 1.9min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:30] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=100; total time= 2.3min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:34] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=200; total time=  33.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:44] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=200; total time=  33.3s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:44] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=200; total time=  31.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:56] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=100; total time= 2.4min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:25:57] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=200; total time=  31.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:01] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=5, n_estimators=50; total time=  18.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:02] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=5, n_estimators=50; total time=  18.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:03] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=3, n_estimators=200; total time=  31.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:05] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=5, n_estimators=50; total time=  18.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:15] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=5, n_estimators=50; total time=  18.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=100; total time= 2.4min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:17] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=5, n_estimators=50; total time=  19.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:20] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=100; total time=  34.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:36] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=100; total time=  34.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:37] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=100; total time=  34.3s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:39] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=100; total time=  33.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:48] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=100; total time=  33.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:26:50] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=200; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:22] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=7, n_estimators=50; total time=  37.7s\n",
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=200; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:26] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:26] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=7, n_estimators=50; total time=  38.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:28] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=200; total time= 3.7min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:34] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=200; total time= 3.7min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:40] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=200; total time= 3.7min\n",
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=200; total time= 3.7min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:41] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:41] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ..learning_rate=0.01, max_depth=7, n_estimators=200; total time= 3.7min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:44] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=200; total time= 1.1min\n",
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=200; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:45] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:46] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=5, n_estimators=200; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:27:47] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=7, n_estimators=50; total time=  57.9s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:28:24] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=7, n_estimators=50; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:28:26] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.1, max_depth=7, n_estimators=50; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:28:32] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=3, n_estimators=50; total time=   9.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:28:35] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=3, n_estimators=50; total time=   9.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:28:41] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=3, n_estimators=50; total time=   9.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:28:44] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=3, n_estimators=50; total time=   9.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:28:50] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=3, n_estimators=50; total time=   8.9s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:28:53] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=100; total time= 1.6min\n",
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=100; total time= 1.5min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:03] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:04] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=100; total time=  17.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:07] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=100; total time=  16.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:10] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=100; total time= 1.6min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:16] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=100; total time=  15.9s\n",
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=100; total time=  15.9s\n",
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=100; total time= 1.6min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:19] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:19] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:20] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=100; total time= 1.6min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:20] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=100; total time=  15.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:23] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=5, n_estimators=50; total time=  17.3s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:37] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=200; total time=  29.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:40] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=5, n_estimators=50; total time=  17.3s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:40] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=200; total time=  29.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:45] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=200; total time=  29.6s\n",
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=200; total time=  29.4s\n",
      "[CV] END ...learning_rate=0.2, max_depth=3, n_estimators=200; total time=  29.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:49] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:49] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:49] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=5, n_estimators=50; total time=  17.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:55] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=5, n_estimators=50; total time=  17.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:57] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=5, n_estimators=50; total time=  18.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:29:58] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=200; total time= 2.5min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:11] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=200; total time= 2.5min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:14] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=200; total time= 2.5min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:14] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=200; total time= 2.5min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:15] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=100; total time=  33.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:19] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=100; total time=  33.1s\n",
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=100; total time=  33.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:22] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:22] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=100; total time=  34.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:23] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.1, max_depth=7, n_estimators=200; total time= 2.1min\n",
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=100; total time=  34.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:29] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:29] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=7, n_estimators=50; total time=  35.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:51] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=7, n_estimators=50; total time=  36.3s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:55] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=7, n_estimators=50; total time=  35.0s\n",
      "[CV] END ....learning_rate=0.2, max_depth=7, n_estimators=50; total time=  35.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:57] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n",
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:57] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ....learning_rate=0.2, max_depth=7, n_estimators=50; total time=  36.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:30:59] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=200; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:31:02] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=200; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:31:03] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=200; total time= 1.1min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:31:17] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=200; total time= 1.1min\n",
      "[CV] END ...learning_rate=0.2, max_depth=5, n_estimators=200; total time= 1.1min\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=100; total time= 1.1min\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=100; total time= 1.1min\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=100; total time= 1.0min\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=100; total time=  59.2s\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=100; total time=  59.9s\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=200; total time= 1.6min\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=200; total time= 1.6min\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=200; total time= 1.6min\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=200; total time= 1.6min\n",
      "[CV] END ...learning_rate=0.2, max_depth=7, n_estimators=200; total time= 1.5min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [16:32:44] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters for XGBoost: {'learning_rate': 0.2, 'max_depth': 7, 'n_estimators': 200}\n",
      "0.7528854983400438\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     foxnews       0.75      0.94      0.83     46733\n",
      "         nbc       0.77      0.39      0.51     24052\n",
      "\n",
      "    accuracy                           0.75     70785\n",
      "   macro avg       0.76      0.66      0.67     70785\n",
      "weighted avg       0.76      0.75      0.73     70785\n",
      "\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>model</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>best_params</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logistic Regression</td>\n",
       "      <td>0.807982</td>\n",
       "      <td>{'C': 10, 'solver': 'liblinear'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Random Forest</td>\n",
       "      <td>0.787865</td>\n",
       "      <td>{'max_depth': None, 'min_samples_split': 5, 'n...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>XGBoost</td>\n",
       "      <td>0.752885</td>\n",
       "      <td>{'learning_rate': 0.2, 'max_depth': 7, 'n_esti...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 model  accuracy  \\\n",
       "0  Logistic Regression  0.807982   \n",
       "1        Random Forest  0.787865   \n",
       "2              XGBoost  0.752885   \n",
       "\n",
       "                                         best_params  \n",
       "0                   {'C': 10, 'solver': 'liblinear'}  \n",
       "1  {'max_depth': None, 'min_samples_split': 5, 'n...  \n",
       "2  {'learning_rate': 0.2, 'max_depth': 7, 'n_esti...  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Initialize results summary\n",
    "results_summary = []\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# Define parameter grids for the models\n",
    "param_grids = {\n",
    "    'Logistic Regression': {\n",
    "        'C': [0.01, 0.1, 1, 10, 100],\n",
    "        'solver': ['lbfgs', 'liblinear']\n",
    "    },\n",
    "    'Random Forest': {\n",
    "        'n_estimators': [50, 100, 200],\n",
    "        'max_depth': [None, 10, 20, 30],\n",
    "        'min_samples_split': [2, 5, 10]\n",
    "    },\n",
    "    'XGBoost': {\n",
    "        'n_estimators': [50, 100, 200],\n",
    "        'learning_rate': [0.01, 0.1, 0.2],\n",
    "        'max_depth': [3, 5, 7]\n",
    "    }\n",
    "}\n",
    "\n",
    "# Train and evaluate models\n",
    "for model_name, model in models.items():\n",
    "    print(model_name)\n",
    "    # Perform Grid Search for sklearn or XGBoost model\n",
    "    grid_search = GridSearchCV(model, param_grids[model_name], cv=5, n_jobs=-1, verbose=2)\n",
    "    grid_search.fit(X_train, y_train)\n",
    "    best_model = grid_search.best_estimator_\n",
    "    print(f\"Best parameters for {model_name}: {grid_search.best_params_}\")\n",
    "        \n",
    "    # Predict with the best model\n",
    "    y_pred = best_model.predict(X_test)\n",
    "\n",
    "    # Evaluate\n",
    "    accuracy = accuracy_score(y_test, y_pred)\n",
    "    results_summary.append({'model': model_name, 'accuracy': accuracy, 'best_params': grid_search.best_params_})\n",
    "    print(accuracy)\n",
    "    print(classification_report(y_test, y_pred, target_names=encoder.classes_))\n",
    "\n",
    "# Convert results summary to DataFrame\n",
    "results_df = pd.DataFrame(results_summary)\n",
    "results_df\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "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>model</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>best_params</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logistic Regression</td>\n",
       "      <td>0.807982</td>\n",
       "      <td>{'C': 10, 'solver': 'liblinear'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Random Forest</td>\n",
       "      <td>0.787865</td>\n",
       "      <td>{'max_depth': None, 'min_samples_split': 5, 'n_estimators': 200}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>XGBoost</td>\n",
       "      <td>0.752885</td>\n",
       "      <td>{'learning_rate': 0.2, 'max_depth': 7, 'n_estimators': 200}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 model  accuracy  \\\n",
       "0  Logistic Regression  0.807982   \n",
       "1        Random Forest  0.787865   \n",
       "2              XGBoost  0.752885   \n",
       "\n",
       "                                                        best_params  \n",
       "0                                  {'C': 10, 'solver': 'liblinear'}  \n",
       "1  {'max_depth': None, 'min_samples_split': 5, 'n_estimators': 200}  \n",
       "2       {'learning_rate': 0.2, 'max_depth': 7, 'n_estimators': 200}  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# show full content of the column\n",
    "pd.set_option('display.max_colwidth', None)\n",
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.12/site-packages/xgboost/core.py:158: UserWarning: [17:03:17] WARNING: /Users/runner/work/xgboost/xgboost/src/learner.cc:740: \n",
      "Parameters: { \"use_label_encoder\" } are not used.\n",
      "\n",
      "  warnings.warn(smsg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score, roc_curve\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# plt with size 20,10\n",
    "plt.figure(figsize=(20, 10))\n",
    "\n",
    "# Calculate AUC-ROC for each model with best parameters\n",
    "for model_name, model in models.items():\n",
    "    best_params = results_df[results_df['model'] == model_name]['best_params'].values[0]\n",
    "    best_model = models[model_name].set_params(**best_params)\n",
    "    best_model.fit(X_train, y_train)\n",
    "    y_pred = best_model.predict_proba(X_test)[:, 1]\n",
    "    fpr, tpr, _ = roc_curve(y_test, y_pred, pos_label=1)\n",
    "    auc = roc_auc_score(y_test, y_pred)\n",
    "    plt.plot(fpr, tpr, label=f\"{model_name} (AUC = {auc:.2f})\")\n",
    "\n",
    "plt.plot([0, 1], [0, 1], 'k--')\n",
    "plt.xlabel('False positive rate')\n",
    "plt.ylabel('True positive rate')\n",
    "plt.title('ROC curve')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save the model\n",
    "import joblib\n",
    "\n",
    "# Save the best model\n",
    "best_model = models['Random Forest']\n",
    "best_model.set_params(**results_df[results_df['model'] == 'Random Forest']['best_params'].values[0])\n",
    "best_model.fit(X_train, y_train)\n",
    "joblib.dump(best_model, 'best_model.pkl')\n",
    "\n",
    "# Save the encoder\n",
    "joblib.dump(encoder, 'encoder.pkl')\n",
    "\n",
    "# Save the vectorizer\n",
    "joblib.dump(vectorizer, 'vectorizer.pkl')\n",
    "\n",
    "# Save the results summary\n",
    "results_df.to_csv('results_summary.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'foxnews'"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# deploy the best model\n",
    "# given the title, predict the source\n",
    "def predict_source(title):\n",
    "    # Load the model\n",
    "    model = joblib.load('best_model.pkl')\n",
    "    encoder = joblib.load('encoder.pkl')\n",
    "    vectorizer = joblib.load('vectorizer.pkl')\n",
    "    \n",
    "    # Vectorize the input\n",
    "    X = vectorizer.transform([title])\n",
    "    \n",
    "    # Predict\n",
    "    y_pred = model.predict(X)\n",
    "    return encoder.inverse_transform(y_pred)[0]\n",
    "\n",
    "# Test the function\n",
    "predict_source('Wife of murdered Minnesota pastor hired 3 men to kill husband after affair: police')\n",
    "\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}