day55-regularized-linear-models/ridge-regression-from-scratch.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "99dcba6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_diabetes\n",
    "from sklearn.metrics import r2_score\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "821f3faf",
   "metadata": {},
   "outputs": [],
   "source": [
    "X,y = load_diabetes(return_X_y=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "741b4551",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "id": "5d4d2d1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "30a774ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "5a15c5fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = Ridge(alpha=0.1,solver='cholesky')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "7d66d9f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ridge(alpha=0.1, solver='cholesky')"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "09460a79",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.46930668289209754"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = reg.predict(X_test)\n",
    "r2_score(y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "927958f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[  44.02025512 -241.69666596  452.98163524  332.04993719  -76.34010147\n",
      "  -68.52063199 -164.98817213  149.9687712   431.61985919   58.51762582]\n",
      "150.89050209333024\n"
     ]
    }
   ],
   "source": [
    "print(reg.coef_)\n",
    "print(reg.intercept_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "7ac16788",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MeraRidge:\n",
    "    \n",
    "    def __init__(self,alpha=0.1):\n",
    "        \n",
    "        self.alpha = alpha\n",
    "        self.coef_ = None\n",
    "        self.intercept_ = None\n",
    "        \n",
    "    def fit(self,X_train,y_train):\n",
    "        \n",
    "        X_train = np.insert(X_train,0,1,axis=1)\n",
    "        I = np.identity(X_train.shape[1])\n",
    "        I[0][0] = 0\n",
    "        result = np.linalg.inv(np.dot(X_train.T,X_train) + self.alpha * I).dot(X_train.T).dot(y_train)\n",
    "        self.intercept_ = result[0]\n",
    "        self.coef_ = result[1:]\n",
    "    \n",
    "    def predict(self,X_test):\n",
    "        return np.dot(X_test,self.coef_) + self.intercept_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "0884ea38",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.46930668289209776\n",
      "[  44.02025512 -241.69666596  452.98163524  332.04993719  -76.34010147\n",
      "  -68.52063199 -164.98817213  149.9687712   431.61985919   58.51762582]\n",
      "150.8905020933302\n"
     ]
    }
   ],
   "source": [
    "reg = MeraRidge()\n",
    "reg.fit(X_train,y_train)\n",
    "y_pred = reg.predict(X_test)\n",
    "print(r2_score(y_test,y_pred))\n",
    "print(reg.coef_)\n",
    "print(reg.intercept_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "807e38bc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}