Silvia l Random Forest

Solutions.ipynb

@@ -16,44 +16,148 @@
     "- For this lab, you will be using the CSV files provided in the files_for_lab folder."
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "b3286904",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "pd.set_option('display.max_columns', None)\n",
+    "import warnings\n",
+    "warnings.filterwarnings('ignore')\n",
+    "from sklearn.preprocessing import OneHotEncoder\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from imblearn.over_sampling import SMOTE\n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "from sklearn.model_selection import cross_val_score"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "40546e90",
    "metadata": {},
    "source": [
-    "### 1. Apply the Random Forests algorithm but this time only by upscaling the data using SMOTE."
+    "### 1. Apply the Random Forests algorithm but this time only by upscaling the data using SMOTE. Note that since SMOTE works on numerical data only, we will first encode the categorical variables in this case."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "id": "64274b33",
    "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0    90569\n",
+       "1     4843\n",
+       "Name: TARGET_B, dtype: int64"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "numerical = pd.read_csv('./files_for_lab/numerical.csv')\n",
+    "categorical = pd.read_csv('./files_for_lab/categorical.csv')\n",
+    "targets = pd.read_csv('./files_for_lab/target.csv')\n",
+    "data = pd.concat([numerical, categorical, targets], axis = 1)\n",
+    "data['TARGET_B'].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "6f71a011",
+   "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "y = data['TARGET_B']\n",
+    "X = data.drop(['TARGET_B'], axis = 1)"
+   ]
   },
   {
-   "cell_type": "markdown",
-   "id": "9100309b",
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "2c120e30",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "numericalX = X.select_dtypes(np.number)\n",
+    "categorcalX = X.select_dtypes(np.object)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "86bdc530",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "encoder = OneHotEncoder(drop='first').fit(categorcalX)\n",
+    "encoded_categorical = encoder.transform(categorcalX).toarray()\n",
+    "encoded_categorical = pd.DataFrame(encoded_categorical)\n",
+    "X = pd.concat([numericalX, encoded_categorical], axis = 1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "fe7eb394",
    "metadata": {},
+   "outputs": [],
    "source": [
-    "### 2. Note that since SMOTE works on numerical data only, we will first encode the categorical variables in this case."
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=0)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "id": "0fedb9b1",
+   "execution_count": 7,
+   "id": "af3ca41b",
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "smote = SMOTE()\n",
+    "\n",
+    "X_train_sm, y_train_sm = smote.fit_resample(X_train, y_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "a2161028",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.9474279964784303\n",
+      "0.9498455836369397\n"
+     ]
+    }
+   ],
+   "source": [
+    "clf = RandomForestClassifier(max_depth=2, random_state=0)\n",
+    "clf.fit(X_train, y_train)\n",
+    "print(clf.score(X_test, y_test))\n",
+    "\n",
+    "clf = RandomForestClassifier(max_depth=2, random_state=0)\n",
+    "cross_val_scores = cross_val_score(clf, X_train, y_train, cv=10)\n",
+    "print(np.mean(cross_val_scores))"
+   ]
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "ironhac",
+   "display_name": "Python 3",
    "language": "python",
-   "name": "ironhac"
+   "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
@@ -65,7 +169,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.2"
+   "version": "3.8.8"
   }
  },
  "nbformat": 4,