class 32 is done

11_dummy_variable_trap.ipynb

+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 99,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "from jupyterthemes import jtplot\n",
+    "jtplot.style(theme='gruvboxd')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 100,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "N = 10\n",
+    "D = 3"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 101,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([[1., 1., 0.],\n",
+       "       [1., 1., 0.],\n",
+       "       [1., 1., 0.],\n",
+       "       [1., 1., 0.],\n",
+       "       [1., 1., 0.],\n",
+       "       [1., 0., 1.],\n",
+       "       [1., 0., 1.],\n",
+       "       [1., 0., 1.],\n",
+       "       [1., 0., 1.],\n",
+       "       [1., 0., 1.]])"
+      ]
+     },
+     "execution_count": 101,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "X = np.zeros((N, D))\n",
+    "X[:,0] = 1\n",
+    "X[:5,1] = 1\n",
+    "X[5:,2] = 1\n",
+    "X"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 102,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([0, 0, 0, 0, 0, 1, 1, 1, 1, 1])"
+      ]
+     },
+     "execution_count": 102,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Y = np.array([0]*5 + [1]*5)\n",
+    "Y"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 103,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# w = np.linalg.solve(X.T.dot(X), X.T.dot(Y))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 104,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "epochs = 1000\n",
+    "w = np.random.randn(D) / np.sqrt(D)\n",
+    "lr = 0.001"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 105,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def get_w_der(X, delta):\n",
+    "    return X.T.dot(delta)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 106,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def get_MSE(X, Y, w):\n",
+    "    MSE = []\n",
+    "    for t in range(epochs):\n",
+    "        Yhat = X.dot(w)\n",
+    "        delta = Yhat - Y\n",
+    "        w = w - lr*get_w_der(X, delta)\n",
+    "        mse = delta.dot(delta) / N\n",
+    "        MSE.append(mse)\n",
+    "    return MSE"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 107,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 460.8x403.2 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "last mse 9.20931783766317e-05\n"
+     ]
+    }
+   ],
+   "source": [
+    "MSE = get_MSE(X, Y, w)\n",
+    "plt.plot(MSE)\n",
+    "plt.show()\n",
+    "print('last mse', MSE[len(MSE)-1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 108,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 460.8x403.2 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.plot(Y, label='target')\n",
+    "plt.plot(Yhat, label='prediction')\n",
+    "plt.legend()\n",
+    "plt.show()"
+   ]
+  }
+ ],
+ "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": "3.6.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}