ssugiyama/neshi.ipynb

## neshi.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('shojisu_kaiten2.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = df.iloc[:, 0:8].values\n",
    "y = df.iloc[:, 8].values.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 315,
   "metadata": {},
   "outputs": [],
   "source": [
    "x0 = x[x[:, 0] == 0]\n",
    "y0 = y[x[:, 0] == 0]\n",
    "x1 = x[x[:, 0] == 1]\n",
    "y1 = y[x[:, 0] == 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 316,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "kfold = KFold(5, shuffle=True)\n",
    "split = kfold.split(x1)\n",
    "split = list(split)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(42, 8)"
      ]
     },
     "execution_count": 281,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x0.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 317,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_dataset(i, _x, _y):\n",
    "    return _x[split[i][0]], _y[split[i][0]], _x[split[i][1]], _y[split[i][1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/site-packages/scipy/linalg/basic.py:223: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number: 9.221851574618809e-20\n",
      "  ' condition number: {}'.format(rcond), RuntimeWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "KernelRidge(alpha=0, coef0=1, degree=3, gamma=1e-07, kernel='linear',\n",
       "      kernel_params=None)"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.kernel_ridge import KernelRidge\n",
    "kr = KernelRidge(alpha=0, kernel='linear', gamma=0.0000001)\n",
    "kr.fit(train_x, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_y = kr.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "147994343.75911847"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean((test_y-pred_y)**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 326,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "668407725.634\n",
      "1079674157.31\n",
      "804429459.959\n",
      "2453905269.86\n",
      "3136958186.14\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "64065937.272935063"
      ]
     },
     "execution_count": 326,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_ys = []\n",
    "for i in range(5):\n",
    "    train_x, train_y, test_x, test_y = get_dataset(i, x1, y1)\n",
    "    kr = KernelRidge(alpha=1000, kernel='poly', degree=2)\n",
    " #   kr = LinearRegression(normalize=False) \n",
    "    kr.fit(train_x, train_y)\n",
    "    pred_y = kr.predict(test_x)\n",
    "    print(np.mean((test_y-pred_y)**2))\n",
    "    pred_y = kr.predict(x1)\n",
    "    pred_ys.append(pred_y)\n",
    "np.mean((np.mean(pred_ys, axis=0) - y1)**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 327,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x11b36a400>"
      ]
     },
     "execution_count": 327,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11b1b4d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y0, (np.mean(pred_ys, axis=0))-y1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 312,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ -260.09582989,  1253.14800558,  -623.23894678, -2100.5579591 ,\n",
       "        2096.04043716, -1370.39462547,    70.471288  ,  2061.18171313,\n",
       "         210.27182744,  -924.13772215, -1939.32669674,   109.78702867,\n",
       "        -215.15916274,   299.37574509,    63.85001686,   -50.66809427,\n",
       "         263.72994123,   671.91824185,   181.97868203,   715.65134855,\n",
       "          36.57555645,   121.47592863,  -280.29311864,   407.80912381,\n",
       "         590.11777997, -2111.80080431,  -845.19197236,   214.94107447,\n",
       "        -721.85117123,  -840.04805598,   197.87427703,  2895.46423512,\n",
       "        -750.12368083,  -536.93128617,  1168.90487671, -1788.80896264,\n",
       "         786.72263237,   -67.72275628, -2891.4098957 ,   661.3327594 ])"
      ]
     },
     "execution_count": 312,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(((np.mean(pred_ys, axis=0))-y0).flatten())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [],
   "source": [
    "y0 = np.delete(y0, 39)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {},
   "outputs": [],
   "source": [
    "x0 = np.delete(x0, 39, axis=0)\n",
    "y0 = np.delete(y0, 39, axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(42, 8)"
      ]
     },
     "execution_count": 287,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x0.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}
	{
	"cells": [
	{
	"cell_type": "code",
	"execution_count": 180,
	"metadata": {},
	"outputs": [],
	"source": [
	"import numpy as np\n",
	"import pandas as pd\n",
	"from sklearn.linear_model import LinearRegression\n",
	"import matplotlib.pyplot as plt\n",
	"%matplotlib inline"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 2,
	"metadata": {},
	"outputs": [],
	"source": [
	"df = pd.read_csv('shojisu_kaiten2.csv')"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 9,
	"metadata": {},
	"outputs": [],
	"source": [
	"x = df.iloc[:, 0:8].values\n",
	"y = df.iloc[:, 8].values.reshape(-1, 1)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 315,
	"metadata": {},
	"outputs": [],
	"source": [
	"x0 = x[x[:, 0] == 0]\n",
	"y0 = y[x[:, 0] == 0]\n",
	"x1 = x[x[:, 0] == 1]\n",
	"y1 = y[x[:, 0] == 1]"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 316,
	"metadata": {},
	"outputs": [],
	"source": [
	"from sklearn.model_selection import KFold\n",
	"kfold = KFold(5, shuffle=True)\n",
	"split = kfold.split(x1)\n",
	"split = list(split)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 281,
	"metadata": {},
	"outputs": [
	{
	"data": {
	"text/plain": [
	"(42, 8)"
	]
	},
	"execution_count": 281,
	"metadata": {},
	"output_type": "execute_result"
	}
	],
	"source": [
	"x0.shape"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 317,
	"metadata": {},
	"outputs": [],
	"source": [
	"def get_dataset(i, _x, _y):\n",
	" return _x[split[i][0]], _y[split[i][0]], _x[split[i][1]], _y[split[i][1]]"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 275,
	"metadata": {},
	"outputs": [
	{
	"name": "stderr",
	"output_type": "stream",
	"text": [
	"/usr/local/lib/python3.6/site-packages/scipy/linalg/basic.py:223: RuntimeWarning: scipy.linalg.solve\n",
	"Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
	"Reciprocal condition number: 9.221851574618809e-20\n",
	" ' condition number: {}'.format(rcond), RuntimeWarning)\n"
	]
	},
	{
	"data": {
	"text/plain": [
	"KernelRidge(alpha=0, coef0=1, degree=3, gamma=1e-07, kernel='linear',\n",
	" kernel_params=None)"
	]
	},
	"execution_count": 275,
	"metadata": {},
	"output_type": "execute_result"
	}
	],
	"source": [
	"from sklearn.kernel_ridge import KernelRidge\n",
	"kr = KernelRidge(alpha=0, kernel='linear', gamma=0.0000001)\n",
	"kr.fit(train_x, train_y)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 96,
	"metadata": {},
	"outputs": [],
	"source": [
	"pred_y = kr.predict(test_x)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 97,
	"metadata": {},
	"outputs": [
	{
	"data": {
	"text/plain": [
	"147994343.75911847"
	]
	},
	"execution_count": 97,
	"metadata": {},
	"output_type": "execute_result"
	}
	],
	"source": [
	"np.mean((test_y-pred_y)**2)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 326,
	"metadata": {},
	"outputs": [
	{
	"name": "stdout",
	"output_type": "stream",
	"text": [
	"668407725.634\n",
	"1079674157.31\n",
	"804429459.959\n",
	"2453905269.86\n",
	"3136958186.14\n"
	]
	},
	{
	"data": {
	"text/plain": [
	"64065937.272935063"
	]
	},
	"execution_count": 326,
	"metadata": {},
	"output_type": "execute_result"
	}
	],
	"source": [
	"pred_ys = []\n",
	"for i in range(5):\n",
	" train_x, train_y, test_x, test_y = get_dataset(i, x1, y1)\n",
	" kr = KernelRidge(alpha=1000, kernel='poly', degree=2)\n",
	" # kr = LinearRegression(normalize=False) \n",
	" kr.fit(train_x, train_y)\n",
	" pred_y = kr.predict(test_x)\n",
	" print(np.mean((test_y-pred_y)**2))\n",
	" pred_y = kr.predict(x1)\n",
	" pred_ys.append(pred_y)\n",
	"np.mean((np.mean(pred_ys, axis=0) - y1)**2)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 327,
	"metadata": {},
	"outputs": [
	{
	"data": {
	"text/plain": [
	"<matplotlib.collections.PathCollection at 0x11b36a400>"
	]
	},
	"execution_count": 327,
	"metadata": {},
	"output_type": "execute_result"
	},
	{
	"data": {
	"image/png": "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\n",
	"text/plain": [
	"<matplotlib.figure.Figure at 0x11b1b4d68>"
	]
	},
	"metadata": {},
	"output_type": "display_data"
	}
	],
	"source": [
	"plt.scatter(y0, (np.mean(pred_ys, axis=0))-y1)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 312,
	"metadata": {},
	"outputs": [
	{
	"data": {
	"text/plain": [
	"array([ -260.09582989, 1253.14800558, -623.23894678, -2100.5579591 ,\n",
	" 2096.04043716, -1370.39462547, 70.471288 , 2061.18171313,\n",
	" 210.27182744, -924.13772215, -1939.32669674, 109.78702867,\n",
	" -215.15916274, 299.37574509, 63.85001686, -50.66809427,\n",
	" 263.72994123, 671.91824185, 181.97868203, 715.65134855,\n",
	" 36.57555645, 121.47592863, -280.29311864, 407.80912381,\n",
	" 590.11777997, -2111.80080431, -845.19197236, 214.94107447,\n",
	" -721.85117123, -840.04805598, 197.87427703, 2895.46423512,\n",
	" -750.12368083, -536.93128617, 1168.90487671, -1788.80896264,\n",
	" 786.72263237, -67.72275628, -2891.4098957 , 661.3327594 ])"
	]
	},
	"execution_count": 312,
	"metadata": {},
	"output_type": "execute_result"
	}
	],
	"source": [
	"(((np.mean(pred_ys, axis=0))-y0).flatten())"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 270,
	"metadata": {},
	"outputs": [],
	"source": [
	"y0 = np.delete(y0, 39)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 296,
	"metadata": {},
	"outputs": [],
	"source": [
	"x0 = np.delete(x0, 39, axis=0)\n",
	"y0 = np.delete(y0, 39, axis=0)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 287,
	"metadata": {},
	"outputs": [
	{
	"data": {
	"text/plain": [
	"(42, 8)"
	]
	},
	"execution_count": 287,
	"metadata": {},
	"output_type": "execute_result"
	}
	],
	"source": [
	"x0.shape"
	]
	},
	{
	"cell_type": "code",
	"execution_count": null,
	"metadata": {},
	"outputs": [],
	"source": []
	},
	{
	"cell_type": "code",
	"execution_count": null,
	"metadata": {},
	"outputs": [],
	"source": []
	}
	],
	"metadata": {
	"kernelspec": {
	"display_name": "Python 3",
	"language": "python",
	"name": "python3"
	},
	"language_info": {
	"codemirror_mode": {
	"name": "ipython",
	"version": 3
	},
	"file_extension": ".py",
	"mimetype": "text/x-python",
	"name": "python",
	"nbconvert_exporter": "python",
	"pygments_lexer": "ipython3",
	"version": "3.6.4"
	}
	},
	"nbformat": 4,
	"nbformat_minor": 2
	}