var_sellection_PLS1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Хочу сначала отобрать признаки. Признаки с низкой дисперсией скорее всего менее информативны, поэтому, сначала сделаю отбор по дисперсии. В следующем ноутбуке хочу попробовать SBS алгоритм."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from gzip import open # NB: overrides standard open()\n",
    "import pickle as pkl\n",
    "import tensorly as tl\n",
    "from tensorly.base import tensor_to_vec,  partial_tensor_to_vec\n",
    "import pandas as pd\n",
    "import warnings \n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import cross_validate\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.cross_decomposition import PLSRegression\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "import matplotlib as mpl\n",
    "from glob import glob\n",
    "import os\n",
    "import math\n",
    "import functools\n",
    "from sklearn.base import BaseEstimator\n",
    "from sklearn.base import RegressorMixin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "warnings.simplefilter('ignore') # отключим предупреждения Anaconda"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(206,)\n"
     ]
    }
   ],
   "source": [
    "Xdata_numpy = pkl.load(open('C:/Users/admin/Desktop/PLS2021/N-PLS/X_new.pkl.gz', 'rb'))\n",
    "Ydata = pkl.load(open('C:/Users/admin/Desktop/PLS2021/N-PLS/y.pkl.gz', 'rb'))\n",
    "print(Xdata_numpy['emission'].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_big=Xdata_numpy['X']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {},
   "outputs": [],
   "source": [
    "import  subprocess\n",
    "import sys\n",
    "import os\n",
    "from PLS1_help_scripts.regression import new_PLS1_regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "metadata": {},
   "outputs": [],
   "source": [
    "r2cv_best=np.zeros([10,3])\n",
    "r2c_best=np.zeros([10,3])\n",
    "r2p_best=np.zeros([10,3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(1,11,1):\n",
    "    for j in range(1,4,1):\n",
    "        model=new_PLS1_regression(n_components=5,\n",
    "                          var_excitation=i*0.01,\n",
    "                          var_emission=j*0.002)\n",
    "        result=model.regression(X_big.copy(),Ydata.copy(),Xdata_numpy['excitation'],\n",
    "                          Xdata_numpy['emission'],type_matter=0,\n",
    "                          test_size_=0.1428)\n",
    "        \n",
    "        r2cv_best[i-1,j-1]=max(result.r2_cv)\n",
    "        r2c_best[i-1,j-1]=max(result.r2_c)\n",
    "        r2p_best[i-1,j-1]=result.r2_p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(r2cv_best[:,:], aspect='auto')\n",
    "plt.show();\n",
    "plt.imshow(r2p_best[:,:], aspect='auto')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.52552399, 0.72010522, 0.42860273],\n",
       "       [0.53326221, 0.63906024, 0.44569846],\n",
       "       [0.49888747, 0.66014041, 0.56750065],\n",
       "       [0.17915837, 0.66597225, 0.64557938],\n",
       "       [0.43570484, 0.60211213, 0.08193066],\n",
       "       [0.48804486, 0.63905034, 0.61626462],\n",
       "       [0.70096151, 0.73503927, 0.64690371],\n",
       "       [0.70292254, 0.28645714, 0.46480568],\n",
       "       [0.48902522, 0.09664526, 0.50828242],\n",
       "       [0.72624786, 0.59262507, 0.60637404]])"
      ]
     },
     "execution_count": 242,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r2cv_best"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 333,
   "metadata": {},
   "outputs": [],
   "source": [
    "model=new_PLS1_regression(n_components=5,\n",
    "                          var_excitation=0.058,\n",
    "                          var_emission=0.004)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 334,
   "metadata": {},
   "outputs": [],
   "source": [
    "result=model.regression(X_big,Ydata,Xdata_numpy['excitation'],\n",
    "                          Xdata_numpy['emission'],type_matter=0,\n",
    "                          test_size_=0.15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 335,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4130382118950876"
      ]
     },
     "execution_count": 335,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.r2_p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 336,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.39891416,  0.10236568,  0.73504317,  0.71011938,  0.38158194])"
      ]
     },
     "execution_count": 336,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.r2_cv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 337,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.13876578, 0.4941609 , 0.82546858, 0.86899766, 0.89843168])"
      ]
     },
     "execution_count": 337,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.r2_c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 338,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 338,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.best['n_components']"
   ]
  },
  {
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}