ExtremeRandomTrees.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import discriminant_analysis\n",
    "%matplotlib inline\n",
    "data = pd.read_csv('numeric sequence.csv', sep = ',', engine = 'python')\n",
    "processed_data = data\n",
    "for i in range (1, 15):\n",
    "     processed_data['feature'+str(i)] = data['feature'+str(i*2-1)]*2 + data['feature'+str(i*2)]\n",
    "to_drop=[]\n",
    "for i in range (15, 29):\n",
    "    to_drop.append('feature'+str(i))\n",
    "processed_data = processed_data.drop(to_drop, 1).drop('label', 1)\n",
    "processed_data = processed_data.replace(0, 'A')\\\n",
    "                              .replace(1, 'T')\\\n",
    "                              .replace(2, 'C')\\\n",
    "                              .replace(3, 'G')\n",
    "processed_data = pd.get_dummies(processed_data)\n",
    "processed_data=pd.concat((processed_data, data['label']), axis = 1)\n",
    "data = processed_data\n",
    "X, X_test, y, y_test = train_test_split(data.drop('label', axis=1), data['label'], test_size = 0.1, \n",
    "                                                    random_state = 2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Extremely randomized trees"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Зависимость ошибок от количества деревьев"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.164179104478\n",
      "85\n"
     ]
    }
   ],
   "source": [
    "from sklearn import ensemble\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "est_number = np.arange(1, 100)\n",
    "ert_model = ensemble.ExtraTreesClassifier()\n",
    "grid = GridSearchCV(ert_model, param_grid = {'n_estimators': est_number})\n",
    "grid.fit(X, y)\n",
    "min_err_cv = 1 - grid.best_score_\n",
    "argmin_err_cv = grid.best_estimator_.n_estimators\n",
    "print(min_err_cv)\n",
    "print(\"Число деревьев:\")\n",
    "print(argmin_err_cv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x21a815565f8>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nHPxmZplx8JuZZcbBb2aWGQe/mVlm/j/cXr3mfG0BygAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x21a80039400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "score = 1 - np.array([i for i in grid.cv_results_['mean_test_score']])\n",
    "depth = np.array([i['n_estimators'] for i in grid.cv_results_['params']])\n",
    "plt.scatter(depth, score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}