PCA-CNN-LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7c701c67",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 导包\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import glob\n",
    "import os\n",
    "import tensorflow as tf\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.utils import shuffle\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6075ea34",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_path = \"./\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "199a7a4c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['.ipynb_checkpoints', '0-6.csv', '7-9.csv', 'CNN+LSTM=数据9在第一位 64-0712最新修改.ipynb', 'CNN+LSTM=数据9在第一位 64-0712最新修改导入模型.ipynb', 'data', 'graphviz.ipynb', 'GRU.h5', 'GRU0712.h5', 'GRU071236-15.h5', 'GRU071236-15原始.h5', 'GRU=数据0在第一位-修改0712.ipynb', 'GRU=数据0在第一位.ipynb', 'GRU=数据1在第一位.ipynb', 'GRU=数据2在第一位.ipynb', 'GRU=数据3在第一位.ipynb', 'GRU=数据4在第一位.ipynb', 'GRU=数据5在第一位.ipynb', 'GRU=数据6在第一位.ipynb', 'GRU=数据7在第一位.ipynb', 'GRU=数据8在第一位 -64-0712最新修改.ipynb', 'GRU=数据8在第一位.ipynb', 'GRU=数据9在第一位 - 修改0712.ipynb', 'GRU=数据9在第一位.ipynb', 'lstm.png', 'MLP=数据9在第一位 64-0712最新修改.ipynb', 'model.h5', 'modelnew.h5', 'SMOTE.ipynb', 'Untitled.ipynb', '数据合并.ipynb', '数据拆分.ipynb', '模型保存']\n"
     ]
    }
   ],
   "source": [
    "print(os.listdir(train_path))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1c0169b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>gpsPdop</th>\n",
       "      <th>gpsNumSv</th>\n",
       "      <th>gpsWeek</th>\n",
       "      <th>itow</th>\n",
       "      <th>gpsSec</th>\n",
       "      <th>P_Sol_Status</th>\n",
       "      <th>Pos_Type</th>\n",
       "      <th>AvionicsTemp</th>\n",
       "      <th>MainPowerV</th>\n",
       "      <th>...</th>\n",
       "      <th>rcState</th>\n",
       "      <th>Failure_ad</th>\n",
       "      <th>leftRPM</th>\n",
       "      <th>AirHeight</th>\n",
       "      <th>Failure</th>\n",
       "      <th>WWest</th>\n",
       "      <th>WSouth</th>\n",
       "      <th>Event</th>\n",
       "      <th>Capacity</th>\n",
       "      <th>Jammer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>142</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190057400</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>198.4</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-1.93</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>143</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190058400</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>198.1</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-1.83</td>\n",
       "      <td>0.77</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>150</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190065400</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>201.7</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-0.41</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>160</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190075400</td>\n",
       "      <td>21.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>197.1</td>\n",
       "      <td>3145920</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.19</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>169</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190084400</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>201.3</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-0.11</td>\n",
       "      <td>1.37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12449</th>\n",
       "      <td>90</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190005400</td>\n",
       "      <td>6504.6</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>202.0</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-0.46</td>\n",
       "      <td>1.04</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12450</th>\n",
       "      <td>94</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190009400</td>\n",
       "      <td>6508.6</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.4</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>201.3</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-0.31</td>\n",
       "      <td>1.36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12451</th>\n",
       "      <td>101</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190016400</td>\n",
       "      <td>6515.6</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>198.3</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-1.11</td>\n",
       "      <td>1.26</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12452</th>\n",
       "      <td>119</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190034400</td>\n",
       "      <td>6533.6</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>197.4</td>\n",
       "      <td>3145920</td>\n",
       "      <td>0.30</td>\n",
       "      <td>-0.13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12453</th>\n",
       "      <td>137</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190052400</td>\n",
       "      <td>6551.6</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.4</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>200.9</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-0.67</td>\n",
       "      <td>1.22</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12454 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       count  gpsPdop  gpsNumSv  gpsWeek       itow  gpsSec  P_Sol_Status  \\\n",
       "0        142      1.0        29     2190  190057400     3.0             0   \n",
       "1        143      1.0        29     2190  190058400     4.0             0   \n",
       "2        150      1.0        29     2190  190065400    11.0             0   \n",
       "3        160      1.0        29     2190  190075400    21.0             0   \n",
       "4        169      1.0        29     2190  190084400    30.0             0   \n",
       "...      ...      ...       ...      ...        ...     ...           ...   \n",
       "12449     90      1.0        29     2190  190005400  6504.6             0   \n",
       "12450     94      1.0        29     2190  190009400  6508.6             0   \n",
       "12451    101      1.0        29     2190  190016400  6515.6             0   \n",
       "12452    119      1.0        29     2190  190034400  6533.6             0   \n",
       "12453    137      1.0        29     2190  190052400  6551.6             0   \n",
       "\n",
       "       Pos_Type  AvionicsTemp  MainPowerV  ...  rcState  Failure_ad  leftRPM  \\\n",
       "0            50            28        24.3  ...        1          64        0   \n",
       "1            50            28        24.2  ...        1          64        0   \n",
       "2            50            28        24.3  ...        1          64        0   \n",
       "3            50            28        24.3  ...        1          64        0   \n",
       "4            50            28        24.1  ...        1          64        0   \n",
       "...         ...           ...         ...  ...      ...         ...      ...   \n",
       "12449        50            28        24.3  ...        1          64        0   \n",
       "12450        50            28        24.4  ...        1          64        0   \n",
       "12451        50            28        24.2  ...        1          64        0   \n",
       "12452        50            28        24.3  ...        1          64        0   \n",
       "12453        50            28        24.4  ...        1          64        0   \n",
       "\n",
       "       AirHeight  Failure  WWest  WSouth  Event  Capacity  Jammer  \n",
       "0          198.4  3145920  -1.93    0.96      0         0       0  \n",
       "1          198.1  3145920  -1.83    0.77      0         0       0  \n",
       "2          201.7  3145920  -0.41    0.02      0         0       0  \n",
       "3          197.1  3145920   0.49    0.19      0         0       0  \n",
       "4          201.3  3145920  -0.11    1.37      0         0       0  \n",
       "...          ...      ...    ...     ...    ...       ...     ...  \n",
       "12449      202.0  3145920  -0.46    1.04      0         0       0  \n",
       "12450      201.3  3145920  -0.31    1.36      0         0       0  \n",
       "12451      198.3  3145920  -1.11    1.26      0         0       0  \n",
       "12452      197.4  3145920   0.30   -0.13      0         0       0  \n",
       "12453      200.9  3145920  -0.67    1.22      0         0       0  \n",
       "\n",
       "[12454 rows x 65 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 读取文件\n",
    "\n",
    "\n",
    "#dataset = pd.read_csv('./data/SMOTE数据拆分/0在第一位.csv')\n",
    "#data = pd.read_csv('./data/SMOTE数据拆分/1在第一位.csv')\n",
    "#dataset = pd.read_csv('./data/SMOTE数据拆分/2在第一位.csv')\n",
    "#data = pd.read_csv('./data/SMOTE数据拆分/3在第一位.csv')\n",
    "#data = pd.read_csv('./data/SMOTE数据拆分/4在第一位.csv')\n",
    "#data = pd.read_csv('./data/SMOTE数据拆分/5在第一位.csv')\n",
    "#dataset = pd.read_csv('./data/SMOTE数据拆分/6在第一位.csv')\n",
    "#dataset = pd.read_csv('./data/SMOTE数据拆分/7在第一位.csv')\n",
    "#dataset = pd.read_csv('./data/SMOTE数据拆分/8在第一位.csv')\n",
    "\n",
    "\n",
    "data = pd.read_csv('./data/SMOTE数据拆分7699数据扩充数据合并/9在第一位.csv')\n",
    "#data = pd.read_csv('./data/SMOTE数据拆分7699/0-8.csv')\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4f23d598",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>gpsPdop</th>\n",
       "      <th>gpsNumSv</th>\n",
       "      <th>gpsWeek</th>\n",
       "      <th>itow</th>\n",
       "      <th>gpsSec</th>\n",
       "      <th>P_Sol_Status</th>\n",
       "      <th>Pos_Type</th>\n",
       "      <th>AvionicsTemp</th>\n",
       "      <th>MainPowerV</th>\n",
       "      <th>...</th>\n",
       "      <th>rcState</th>\n",
       "      <th>Failure_ad</th>\n",
       "      <th>leftRPM</th>\n",
       "      <th>AirHeight</th>\n",
       "      <th>Failure</th>\n",
       "      <th>WWest</th>\n",
       "      <th>WSouth</th>\n",
       "      <th>Event</th>\n",
       "      <th>Capacity</th>\n",
       "      <th>Jammer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>142</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190057400</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>198.4</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-1.93</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>143</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190058400</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.2</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>198.1</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-1.83</td>\n",
       "      <td>0.77</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>150</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190065400</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>201.7</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-0.41</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>160</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190075400</td>\n",
       "      <td>21.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.3</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>197.1</td>\n",
       "      <td>3145920</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.19</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>169</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>2190</td>\n",
       "      <td>190084400</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>28</td>\n",
       "      <td>24.1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>201.3</td>\n",
       "      <td>3145920</td>\n",
       "      <td>-0.11</td>\n",
       "      <td>1.37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   count  gpsPdop  gpsNumSv  gpsWeek       itow  gpsSec  P_Sol_Status  \\\n",
       "0    142      1.0        29     2190  190057400     3.0             0   \n",
       "1    143      1.0        29     2190  190058400     4.0             0   \n",
       "2    150      1.0        29     2190  190065400    11.0             0   \n",
       "3    160      1.0        29     2190  190075400    21.0             0   \n",
       "4    169      1.0        29     2190  190084400    30.0             0   \n",
       "\n",
       "   Pos_Type  AvionicsTemp  MainPowerV  ...  rcState  Failure_ad  leftRPM  \\\n",
       "0        50            28        24.3  ...        1          64        0   \n",
       "1        50            28        24.2  ...        1          64        0   \n",
       "2        50            28        24.3  ...        1          64        0   \n",
       "3        50            28        24.3  ...        1          64        0   \n",
       "4        50            28        24.1  ...        1          64        0   \n",
       "\n",
       "   AirHeight  Failure  WWest  WSouth  Event  Capacity  Jammer  \n",
       "0      198.4  3145920  -1.93    0.96      0         0       0  \n",
       "1      198.1  3145920  -1.83    0.77      0         0       0  \n",
       "2      201.7  3145920  -0.41    0.02      0         0       0  \n",
       "3      197.1  3145920   0.49    0.19      0         0       0  \n",
       "4      201.3  3145920  -0.11    1.37      0         0       0  \n",
       "\n",
       "[5 rows x 65 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 查看前5行数据\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e80698ea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=12454, step=1)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a3f65a12",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 12454 entries, 0 to 12453\n",
      "Data columns (total 65 columns):\n",
      " #   Column          Non-Null Count  Dtype  \n",
      "---  ------          --------------  -----  \n",
      " 0   count           12454 non-null  int64  \n",
      " 1   gpsPdop         12454 non-null  float64\n",
      " 2   gpsNumSv        12454 non-null  int64  \n",
      " 3   gpsWeek         12454 non-null  int64  \n",
      " 4   itow            12454 non-null  int64  \n",
      " 5   gpsSec          12454 non-null  float64\n",
      " 6   P_Sol_Status    12454 non-null  int64  \n",
      " 7   Pos_Type        12454 non-null  int64  \n",
      " 8   AvionicsTemp    12454 non-null  int64  \n",
      " 9   MainPowerV      12454 non-null  float64\n",
      " 10  ServPowerV      12454 non-null  float64\n",
      " 11  powerV          12454 non-null  float64\n",
      " 12  APGlobalStatus  12454 non-null  int64  \n",
      " 13  GPSStatus       12454 non-null  int64  \n",
      " 14  mpStatus        12454 non-null  int64  \n",
      " 15  cmd             12454 non-null  int64  \n",
      " 16  Lat             12454 non-null  float64\n",
      " 17  Lon             12454 non-null  float64\n",
      " 18  H               12454 non-null  float64\n",
      " 19  tas             12454 non-null  float64\n",
      " 20  Eu[0]           12454 non-null  float64\n",
      " 21  Eu[1]           12454 non-null  float64\n",
      " 22  Eu[2]           12454 non-null  float64\n",
      " 23  acc[0]          12454 non-null  float64\n",
      " 24  acc[1]          12454 non-null  float64\n",
      " 25  acc[2]          12454 non-null  float64\n",
      " 26  pqr[0]          12454 non-null  float64\n",
      " 27  pqr[1]          12454 non-null  float64\n",
      " 28  pqr[2]          12454 non-null  float64\n",
      " 29  InertVNED[0]    12454 non-null  float64\n",
      " 30  InertVNED[1]    12454 non-null  float64\n",
      " 31  InertVNED[2]    12454 non-null  float64\n",
      " 32  StaticP         12454 non-null  int64  \n",
      " 33  DynPressure     12454 non-null  float64\n",
      " 34  XMagFiled       12454 non-null  int64  \n",
      " 35  YMagFiled       12454 non-null  int64  \n",
      " 36  ZMagFiled       12454 non-null  int64  \n",
      " 37  Compass         12454 non-null  int64  \n",
      " 38  aileronLeftCmd  12454 non-null  float64\n",
      " 39  elevatorCmd     12454 non-null  float64\n",
      " 40  rudderCmd       12454 non-null  float64\n",
      " 41  throttleCmd     12454 non-null  float64\n",
      " 42  latCmd          12454 non-null  float64\n",
      " 43  lonCmd          12454 non-null  float64\n",
      " 44  pedCmd          12454 non-null  float64\n",
      " 45  throtCmd        12454 non-null  float64\n",
      " 46  Aileron         12454 non-null  float64\n",
      " 47  Elevator        12454 non-null  float64\n",
      " 48  Throttle        12454 non-null  float64\n",
      " 49  Rudder          12454 non-null  float64\n",
      " 50  Action          12454 non-null  int64  \n",
      " 51  photoNum        12454 non-null  int64  \n",
      " 52  XaccBias        12454 non-null  float64\n",
      " 53  YaccBias        12454 non-null  float64\n",
      " 54  ZaccBias        12454 non-null  float64\n",
      " 55  rcState         12454 non-null  int64  \n",
      " 56  Failure_ad      12454 non-null  int64  \n",
      " 57  leftRPM         12454 non-null  int64  \n",
      " 58  AirHeight       12454 non-null  float64\n",
      " 59  Failure         12454 non-null  int64  \n",
      " 60  WWest           12454 non-null  float64\n",
      " 61  WSouth          12454 non-null  float64\n",
      " 62  Event           12454 non-null  int64  \n",
      " 63  Capacity        12454 non-null  int64  \n",
      " 64  Jammer          12454 non-null  int64  \n",
      "dtypes: float64(40), int64(25)\n",
      "memory usage: 6.2 MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "print(data.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "15ca9bd3",
   "metadata": {},
   "outputs": [],
   "source": [
    "## 去除缺失数据\n",
    "#A = data.dropna(axis=0, how='any', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "32b92e4d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       count  gpsPdop  gpsNumSv  gpsWeek       itow  gpsSec  P_Sol_Status  \\\n",
      "0        142      1.0        29     2190  190057400     3.0             0   \n",
      "1        143      1.0        29     2190  190058400     4.0             0   \n",
      "2        150      1.0        29     2190  190065400    11.0             0   \n",
      "3        160      1.0        29     2190  190075400    21.0             0   \n",
      "4        169      1.0        29     2190  190084400    30.0             0   \n",
      "...      ...      ...       ...      ...        ...     ...           ...   \n",
      "12449     90      1.0        29     2190  190005400  6504.6             0   \n",
      "12450     94      1.0        29     2190  190009400  6508.6             0   \n",
      "12451    101      1.0        29     2190  190016400  6515.6             0   \n",
      "12452    119      1.0        29     2190  190034400  6533.6             0   \n",
      "12453    137      1.0        29     2190  190052400  6551.6             0   \n",
      "\n",
      "       Pos_Type  AvionicsTemp  MainPowerV  ...  rcState  Failure_ad  leftRPM  \\\n",
      "0            50            28        24.3  ...        1          64        0   \n",
      "1            50            28        24.2  ...        1          64        0   \n",
      "2            50            28        24.3  ...        1          64        0   \n",
      "3            50            28        24.3  ...        1          64        0   \n",
      "4            50            28        24.1  ...        1          64        0   \n",
      "...         ...           ...         ...  ...      ...         ...      ...   \n",
      "12449        50            28        24.3  ...        1          64        0   \n",
      "12450        50            28        24.4  ...        1          64        0   \n",
      "12451        50            28        24.2  ...        1          64        0   \n",
      "12452        50            28        24.3  ...        1          64        0   \n",
      "12453        50            28        24.4  ...        1          64        0   \n",
      "\n",
      "       AirHeight  Failure  WWest  WSouth  Event  Capacity  Jammer  \n",
      "0          198.4  3145920  -1.93    0.96      0         0       0  \n",
      "1          198.1  3145920  -1.83    0.77      0         0       0  \n",
      "2          201.7  3145920  -0.41    0.02      0         0       0  \n",
      "3          197.1  3145920   0.49    0.19      0         0       0  \n",
      "4          201.3  3145920  -0.11    1.37      0         0       0  \n",
      "...          ...      ...    ...     ...    ...       ...     ...  \n",
      "12449      202.0  3145920  -0.46    1.04      0         0       0  \n",
      "12450      201.3  3145920  -0.31    1.36      0         0       0  \n",
      "12451      198.3  3145920  -1.11    1.26      0         0       0  \n",
      "12452      197.4  3145920   0.30   -0.13      0         0       0  \n",
      "12453      200.9  3145920  -0.67    1.22      0         0       0  \n",
      "\n",
      "[12454 rows x 65 columns]\n"
     ]
    }
   ],
   "source": [
    "print(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "4a6fb86e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\SUN\\anaconda3\\envs\\LSTM\\lib\\site-packages\\seaborn\\_decorators.py:43: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  FutureWarning\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Jammer')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "# 样本分布\n",
    "# 以图方式表示\n",
    "sns.countplot(data[\"Jammer\"])\n",
    "plt.xlabel(\"Jammer\")\n",
    "plt.title(\"Jammer\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1d11fa0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 重新排序\n",
    "# df = data.sample(frac=1).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "b2d1c15b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#A = data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "228d2e4a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       gpsPdop  gpsNumSv  gpsSec  P_Sol_Status  AvionicsTemp        Lat  \\\n",
      "0          1.0        29     3.0             0            28  40.072993   \n",
      "1          1.0        29     4.0             0            28  40.072850   \n",
      "2          1.0        29    11.0             0            28  40.072293   \n",
      "3          1.0        29    21.0             0            28  40.073505   \n",
      "4          1.0        29    30.0             0            28  40.074564   \n",
      "...        ...       ...     ...           ...           ...        ...   \n",
      "12449      1.0        29  6504.6             0            28  40.074574   \n",
      "12450      1.0        29  6508.6             0            28  40.074253   \n",
      "12451      1.0        29  6515.6             0            28  40.073133   \n",
      "12452      1.0        29  6533.6             0            28  40.073304   \n",
      "12453      1.0        29  6551.6             0            28  40.073830   \n",
      "\n",
      "              Lon      H    tas      Eu[0]  ...  elevatorCmd  rudderCmd  \\\n",
      "0      116.103066  198.0  19.92 -13.590559  ...    -2.160051  -1.105809   \n",
      "1      116.103176  198.1  19.65 -12.014925  ...    -1.873572  -1.054242   \n",
      "2      116.104514  201.1  20.85 -18.684154  ...    -1.512609  -1.518338   \n",
      "3      116.105935  197.0  19.82 -16.953821  ...    -2.028271  -1.243318   \n",
      "4      116.104542  201.3  19.98 -15.527156  ...    -1.546986  -1.575634   \n",
      "...           ...    ...    ...        ...  ...          ...        ...   \n",
      "12449  116.104241  201.3  20.54 -17.400728  ...    -1.420935  -1.306344   \n",
      "12450  116.103367  200.7  21.05 -16.724638  ...    -1.398017  -1.266237   \n",
      "12451  116.102998  198.2  19.97 -14.690638  ...    -2.154321  -0.962569   \n",
      "12452  116.105935  197.3  19.20 -15.848013  ...    -2.291831  -1.111538   \n",
      "12453  116.103025  200.2  20.91 -16.604317  ...    -1.249048  -1.243318   \n",
      "\n",
      "       lonCmd  XaccBias  YaccBias  ZaccBias  AirHeight  WWest  WSouth  Jammer  \n",
      "0         0.0       0.0       0.0       0.0      198.4  -1.93    0.96       0  \n",
      "1         0.0       0.0       0.0       0.0      198.1  -1.83    0.77       0  \n",
      "2         0.0       0.0       0.0       0.0      201.7  -0.41    0.02       0  \n",
      "3         0.0       0.0       0.0       0.0      197.1   0.49    0.19       0  \n",
      "4         0.0       0.0       0.0       0.0      201.3  -0.11    1.37       0  \n",
      "...       ...       ...       ...       ...        ...    ...     ...     ...  \n",
      "12449     0.0       0.0       0.0       0.0      202.0  -0.46    1.04       0  \n",
      "12450     0.0       0.0       0.0       0.0      201.3  -0.31    1.36       0  \n",
      "12451     0.0       0.0       0.0       0.0      198.3  -1.11    1.26       0  \n",
      "12452     0.0       0.0       0.0       0.0      197.4   0.30   -0.13       0  \n",
      "12453     0.0       0.0       0.0       0.0      200.9  -0.67    1.22       0  \n",
      "\n",
      "[12454 rows x 37 columns]\n"
     ]
    }
   ],
   "source": [
    "# 数据清洗\n",
    "A = data.loc[:, (data != 0).any(axis=0)]#删除数据全部为0的列\n",
    "del A['count']\n",
    "del A['gpsWeek']\n",
    "del A['itow']\n",
    "del A['Pos_Type']\n",
    "del A['MainPowerV']\n",
    "del A['ServPowerV']\n",
    "del A['powerV']\n",
    "del A['APGlobalStatus']\n",
    "del A['GPSStatus']\n",
    "del A['cmd']\n",
    "del A['StaticP']\n",
    "del A['throttleCmd']\n",
    "del A['latCmd']\n",
    "del A['pedCmd']\n",
    "del A['throtCmd']\n",
    "del A['Aileron']\n",
    "del A['Elevator']\n",
    "del A['Throttle']\n",
    "del A['Rudder']\n",
    "#del A['photoNum']\n",
    "#del A['rcState']\n",
    "del A['mpStatus']\n",
    "del A['Action']\n",
    "del A['rcState']\n",
    "del A['Failure_ad']\n",
    "del A['Failure']\n",
    "data = A\n",
    "print(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "71f95a1e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "5de9e1fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 空值检查\n",
    "# df[df.isnull().values==True]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "e80e8402",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = df.iloc[:,:-1]\n",
    "y = df.iloc[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "eb1804e0",
   "metadata": {},
   "outputs": [],
   "source": [
    "## 划分数据集\n",
    "#from sklearn.model_selection import train_test_split\n",
    "#x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "1ac5ec0e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(12454, 36)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train = x\n",
    "y_train = y\n",
    "x_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "98a0c262",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gpsPdop</th>\n",
       "      <th>gpsNumSv</th>\n",
       "      <th>gpsSec</th>\n",
       "      <th>P_Sol_Status</th>\n",
       "      <th>AvionicsTemp</th>\n",
       "      <th>Lat</th>\n",
       "      <th>Lon</th>\n",
       "      <th>H</th>\n",
       "      <th>tas</th>\n",
       "      <th>Eu[0]</th>\n",
       "      <th>...</th>\n",
       "      <th>aileronLeftCmd</th>\n",
       "      <th>elevatorCmd</th>\n",
       "      <th>rudderCmd</th>\n",
       "      <th>lonCmd</th>\n",
       "      <th>XaccBias</th>\n",
       "      <th>YaccBias</th>\n",
       "      <th>ZaccBias</th>\n",
       "      <th>AirHeight</th>\n",
       "      <th>WWest</th>\n",
       "      <th>WSouth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "      <td>40.072993</td>\n",
       "      <td>116.103066</td>\n",
       "      <td>198.0</td>\n",
       "      <td>19.92</td>\n",
       "      <td>-13.590559</td>\n",
       "      <td>...</td>\n",
       "      <td>2.652795</td>\n",
       "      <td>-2.160051</td>\n",
       "      <td>-1.105809</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>198.4</td>\n",
       "      <td>-1.93</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "      <td>40.072850</td>\n",
       "      <td>116.103176</td>\n",
       "      <td>198.1</td>\n",
       "      <td>19.65</td>\n",
       "      <td>-12.014925</td>\n",
       "      <td>...</td>\n",
       "      <td>2.223076</td>\n",
       "      <td>-1.873572</td>\n",
       "      <td>-1.054242</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>198.1</td>\n",
       "      <td>-1.83</td>\n",
       "      <td>0.77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "      <td>40.072293</td>\n",
       "      <td>116.104514</td>\n",
       "      <td>201.1</td>\n",
       "      <td>20.85</td>\n",
       "      <td>-18.684154</td>\n",
       "      <td>...</td>\n",
       "      <td>1.879302</td>\n",
       "      <td>-1.512609</td>\n",
       "      <td>-1.518338</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>201.7</td>\n",
       "      <td>-0.41</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>21.0</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "      <td>40.073505</td>\n",
       "      <td>116.105935</td>\n",
       "      <td>197.0</td>\n",
       "      <td>19.82</td>\n",
       "      <td>-16.953821</td>\n",
       "      <td>...</td>\n",
       "      <td>2.681442</td>\n",
       "      <td>-2.028271</td>\n",
       "      <td>-1.243318</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>197.1</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>29</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "      <td>40.074564</td>\n",
       "      <td>116.104542</td>\n",
       "      <td>201.3</td>\n",
       "      <td>19.98</td>\n",
       "      <td>-15.527156</td>\n",
       "      <td>...</td>\n",
       "      <td>2.119944</td>\n",
       "      <td>-1.546986</td>\n",
       "      <td>-1.575634</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>201.3</td>\n",
       "      <td>-0.11</td>\n",
       "      <td>1.37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   gpsPdop  gpsNumSv  gpsSec  P_Sol_Status  AvionicsTemp        Lat  \\\n",
       "0      1.0        29     3.0             0            28  40.072993   \n",
       "1      1.0        29     4.0             0            28  40.072850   \n",
       "2      1.0        29    11.0             0            28  40.072293   \n",
       "3      1.0        29    21.0             0            28  40.073505   \n",
       "4      1.0        29    30.0             0            28  40.074564   \n",
       "\n",
       "          Lon      H    tas      Eu[0]  ...  aileronLeftCmd  elevatorCmd  \\\n",
       "0  116.103066  198.0  19.92 -13.590559  ...        2.652795    -2.160051   \n",
       "1  116.103176  198.1  19.65 -12.014925  ...        2.223076    -1.873572   \n",
       "2  116.104514  201.1  20.85 -18.684154  ...        1.879302    -1.512609   \n",
       "3  116.105935  197.0  19.82 -16.953821  ...        2.681442    -2.028271   \n",
       "4  116.104542  201.3  19.98 -15.527156  ...        2.119944    -1.546986   \n",
       "\n",
       "   rudderCmd  lonCmd  XaccBias  YaccBias  ZaccBias  AirHeight  WWest  WSouth  \n",
       "0  -1.105809     0.0       0.0       0.0       0.0      198.4  -1.93    0.96  \n",
       "1  -1.054242     0.0       0.0       0.0       0.0      198.1  -1.83    0.77  \n",
       "2  -1.518338     0.0       0.0       0.0       0.0      201.7  -0.41    0.02  \n",
       "3  -1.243318     0.0       0.0       0.0       0.0      197.1   0.49    0.19  \n",
       "4  -1.575634     0.0       0.0       0.0       0.0      201.3  -0.11    1.37  \n",
       "\n",
       "[5 rows x 36 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.head()\n",
    "#x_train.to_csv('./data/SMOTE数据拆分7699数据扩充数据合并/训练集8在第一位降维.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "82ef8ca4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(769, 65)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset = pd.read_csv('./data/SMOTE数据拆分7699/8.csv')\n",
    "dataset.shape\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "c7b63829",
   "metadata": {},
   "outputs": [],
   "source": [
    "#from sklearn.utils import shuffle  \n",
    "#data= shuffle(data) \n",
    "#data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e2a2cb9d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     gpsPdop  gpsNumSv  gpsSec  P_Sol_Status  AvionicsTemp        Lat  \\\n",
      "0        1.0        31  1210.2             0            28  40.071637   \n",
      "1        1.0        31  1223.2             0            28  40.072354   \n",
      "2        1.0        31  1272.2             0            29  40.072341   \n",
      "3        1.0        32  2793.6             0            31  40.073017   \n",
      "4        1.0        30  2207.4             0            30  40.071729   \n",
      "..       ...       ...     ...           ...           ...        ...   \n",
      "764      5.0        18   595.8            13            47  40.136965   \n",
      "765      1.0        31   786.2             0            21  40.070674   \n",
      "766      2.5         8  5969.2             0            37  40.140445   \n",
      "767      5.0        18   421.0            13            46  40.137040   \n",
      "768      1.0        32  2874.6             0            31  40.069540   \n",
      "\n",
      "            Lon      H    tas      Eu[0]  ...  elevatorCmd  rudderCmd  lonCmd  \\\n",
      "0    116.108722  177.0  19.58 -12.393077  ...    -2.102755  -0.108862  0.0000   \n",
      "1    116.106714  179.7  20.01 -11.378942  ...    -0.538580  -0.716197  0.0000   \n",
      "2    116.106017  181.0  20.31  -3.724226  ...    -1.472502  -0.412530  0.0000   \n",
      "3    116.106085  179.6  20.86 -20.586374  ...    -1.896490  -1.369369  0.0000   \n",
      "4    116.105539  181.3  19.83 -16.587128  ...    -1.730333  -1.294885  0.0000   \n",
      "..          ...    ...    ...        ...  ...          ...        ...     ...   \n",
      "764  116.239083   22.6   4.35  22.643292  ...   -13.905686   0.074485  0.0000   \n",
      "765  116.105402  140.0  20.90  -0.160428  ...     1.260507   0.045837  0.0000   \n",
      "766  116.237375   10.1  -0.02   7.683364  ...   -10.147083   0.000000  0.0000   \n",
      "767  116.238850   15.8   4.06  20.769720  ...   -14.318215   0.097403  0.0000   \n",
      "768  116.106837  119.1   4.06  -5.958761  ...     0.126051  -0.544310 -0.0601   \n",
      "\n",
      "     XaccBias  YaccBias  ZaccBias  AirHeight  WWest  WSouth  Jammer  \n",
      "0        0.00      0.03      0.00      177.6   3.86   -3.80       0  \n",
      "1        0.00      0.03      0.00      180.5   1.97   -2.36       0  \n",
      "2        0.00      0.03      0.00      182.6   1.58   -0.36       0  \n",
      "3        0.00      0.00      0.00      180.2   2.78   -1.51       0  \n",
      "4        0.00      0.00      0.00      181.6   5.51   -4.38       0  \n",
      "..        ...       ...       ...        ...    ...     ...     ...  \n",
      "764      0.00      0.00      0.00       30.6  -4.74    4.17       1  \n",
      "765     -0.02     -0.07      0.00      145.2   5.76   -1.15       0  \n",
      "766      0.00      0.00      0.02      -12.7  -0.06   -0.56       0  \n",
      "767      0.00      0.00      0.00       28.2  -4.76    2.34       1  \n",
      "768      0.00      0.00      0.00      115.1   4.16   -0.50       0  \n",
      "\n",
      "[769 rows x 37 columns]\n"
     ]
    }
   ],
   "source": [
    "# 数据清洗\n",
    "\n",
    "A = dataset.loc[:, (dataset != 0).any(axis=0)]#删除数据全部为0的列\n",
    "\n",
    "del A['count']\n",
    "del A['gpsWeek']\n",
    "del A['itow']\n",
    "del A['Pos_Type']\n",
    "del A['MainPowerV']\n",
    "del A['ServPowerV']\n",
    "del A['powerV']\n",
    "del A['APGlobalStatus']\n",
    "del A['GPSStatus']\n",
    "del A['cmd']\n",
    "del A['StaticP']\n",
    "del A['throttleCmd']\n",
    "del A['latCmd']\n",
    "del A['pedCmd']\n",
    "del A['throtCmd']\n",
    "del A['Aileron']\n",
    "del A['Elevator']\n",
    "del A['Throttle']\n",
    "##del A['Rudder']\n",
    "#del A['photoNum']\n",
    "#del A['rcState']\n",
    "del A['mpStatus']\n",
    "del A['Action']\n",
    "del A['rcState']\n",
    "del A['Failure_ad']\n",
    "del A['Failure']\n",
    "\n",
    "\n",
    "dataset = A\n",
    "\n",
    "print(dataset)\n",
    "#dataset.to_csv('./data/SMOTE数据拆分7699数据扩充数据合并/测试集7在第一位降维.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "49f9e87c",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_test = dataset.iloc[:,:-1]\n",
    "y_test = dataset.iloc[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "0e06d0e5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(769, 36)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "1f5865fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 数据归一化处理\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(x_train)\n",
    "\n",
    "X_train = scaler.transform(x_train)\n",
    "X_test = scaler.transform(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "7e3c1495",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(12454, 36)\n",
      "(12454,)\n",
      "(769, 36)\n",
      "(769,)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print(y_train.shape)\n",
    "print(X_test.shape)\n",
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "897e6bc9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-0.18820559  1.01249942 -1.39601634 ...  1.3522794  -0.9176386\n",
      "   0.1539243 ]\n",
      " [-0.18820559  1.01249942 -1.39561921 ...  1.34876113 -0.88442678\n",
      "   0.09571758]\n",
      " [-0.18820559  1.01249942 -1.39283927 ...  1.39098032 -0.41281897\n",
      "  -0.13404578]\n",
      " ...\n",
      " [-0.18820559  1.01249942  1.19035969 ...  1.35110664 -0.64530169\n",
      "   0.24582964]\n",
      " [-0.18820559  1.01249942  1.19750811 ...  1.34055184 -0.17701506\n",
      "  -0.17999845]\n",
      " [-0.18820559  1.01249942  1.20465653 ...  1.38159828 -0.4991697\n",
      "   0.23357559]]\n",
      "[1.45785117e+01 3.28364749e+00 2.79498400e+00 1.86774861e+00\n",
      " 1.61720314e+00 1.39185004e+00 1.11025441e+00 1.02309966e+00\n",
      " 1.00845330e+00 9.36447528e-01 8.30592385e-01 7.39130118e-01\n",
      " 6.85772956e-01 5.43796315e-01 4.85894858e-01 4.49483759e-01\n",
      " 4.31396583e-01 4.16276437e-01 3.64983226e-01 3.28958236e-01\n",
      " 2.52992916e-01 2.04787916e-01 1.67223281e-01 1.16111270e-01\n",
      " 8.65197088e-02 7.75388201e-02 5.34981406e-02 4.52821493e-02\n",
      " 2.72485867e-02 2.60595105e-02 2.06776554e-02 1.53469384e-02\n",
      " 8.72101746e-03 6.32321656e-03 4.00922321e-03 2.06576155e-03]\n",
      "[4.04926142e-01 9.12051062e-02 7.76322105e-02 5.18777399e-02\n",
      " 4.49187024e-02 3.86593968e-02 3.08379239e-02 2.84171532e-02\n",
      " 2.80103424e-02 2.60103426e-02 2.30701581e-02 2.05297436e-02\n",
      " 1.90477192e-02 1.51042403e-02 1.34959956e-02 1.24846574e-02\n",
      " 1.19822762e-02 1.15623059e-02 1.01376089e-02 9.13699506e-03\n",
      " 7.02701671e-03 5.68809645e-03 4.64471816e-03 3.22505407e-03\n",
      " 2.40313227e-03 2.15368317e-03 1.48594014e-03 1.25773648e-03\n",
      " 7.56844409e-04 7.23817167e-04 5.74333197e-04 4.26269613e-04\n",
      " 2.42231033e-04 1.75630801e-04 1.11358369e-04 5.73776579e-05]\n",
      "0.9999999999999999\n"
     ]
    }
   ],
   "source": [
    "#PCA分析\n",
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(n_components=36)\n",
    "\n",
    "\n",
    "#数据的标准化\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "X_norm = StandardScaler().fit_transform(X_train)\n",
    "print(X_norm)\n",
    "\n",
    "X_train_pca = pca.fit_transform(X_norm)\n",
    "#计算分析后各成分的方差以及方差比例\n",
    "var = pca.explained_variance_\n",
    "var_ratio = pca.explained_variance_ratio_\n",
    "print(var)\n",
    "print(var_ratio)#各个成分方差值\n",
    "print(sum(var_ratio))#成分方差总和"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "8617af3c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(12454, 15)\n",
      "[0.40492614 0.09120511 0.07763221 0.05187774 0.0449187  0.0386594\n",
      " 0.03083792 0.02841715 0.02801034 0.02601034 0.02307016 0.02052974\n",
      " 0.01904772 0.01510424 0.013496  ]\n",
      "<class 'numpy.ndarray'>\n",
      "0.9136\n"
     ]
    },
    {
     "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": [
    "# 训练集数据降维到25维\n",
    "pca = PCA(n_components=15)\n",
    "X_train_pca = pca.fit_transform(X_norm)\n",
    "#var_ratio2 = pca.fit_transform(X_norm)\n",
    "print(X_train_pca.shape)\n",
    "\n",
    "\n",
    "#print(X_test_pca.shape)\n",
    "\n",
    "\n",
    "# 计算方差比例\n",
    "var_ratio2 = pca.explained_variance_ratio_\n",
    "print(var_ratio2)\n",
    "print(type(var_ratio2))\n",
    "\n",
    "var = list(range(1, 16))\n",
    "\n",
    "#var_ratio2='{:.4f}'.format(var_ratio2[0])\n",
    "\n",
    "#var_ratio2=int(var_ratio2)\n",
    "\n",
    "var_ratio2 = np.around(var_ratio2,decimals=4)# 保留4位小数\n",
    "A=np.sum(var_ratio2)\n",
    "print(A)\n",
    "\n",
    "# 绘制图像\n",
    "\n",
    "#plt.bar(var,var_ratio2)\n",
    "#plt.bar(var,var_ratio2,tick_label=[\"1\",\"2\",\"3\",\"4\",\"5\",\"6\",\"7\",\"8\",\"9\",\"10\",\"11\",\"12\",\"13\",\"14\",\"15\",\n",
    "#                                   \"16\",\"17\",\"18\",\"19\",\"20\",\"21\",\"22\",\"23\",\"24\",\"25\"])\n",
    "plt.bar(var,var_ratio2,tick_label=[\"1\",\"2\",\"3\",\"4\",\"5\",\"6\",\"7\",\"8\",\"9\",\"10\",\"11\",\"12\",\"13\",\"14\",\"15\"])\n",
    "\n",
    "plt.xlabel('Components')\n",
    "plt.ylabel('Variance ratio')\n",
    "\n",
    "\n",
    "#for i in range(len(var)):\n",
    "#    plt.text(x = i-0.1, y = var_ratio2[i]+1,s = var_ratio2[i] ) # s表示注释内容\n",
    "\n",
    "for var,var_ratio2 in zip(var,var_ratio2):   #柱子上的数字显示\n",
    " plt.text(var,var_ratio2,'%.4f'%var_ratio2,ha='center',va='bottom',fontsize=7);\n",
    "\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "04be00af",
   "metadata": {},
   "outputs": [],
   "source": [
    "#X_norm = StandardScaler().fit_transform(X_test)\n",
    "#X_test_pca = pca.fit_transform(X_norm)\n",
    "##X_test_pca = pca.fit_transform(X_test)\n",
    "#\n",
    "#print(X_test_pca)\n",
    "#print(X_test_pca.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "98355012",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(769, 15)\n"
     ]
    }
   ],
   "source": [
    "X_test_pca = pca.transform(X_test)\n",
    "print(X_test_pca.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "d114a85e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 对数据处理\n",
    "from keras.utils import np_utils\n",
    "\n",
    "X_train = X_train_pca.reshape((-1,1,15))\n",
    "Y_train = np_utils.to_categorical(y_train)\n",
    "X_test = X_test_pca.reshape((-1,1,15))\n",
    "Y_test = np_utils.to_categorical(y_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "5a6a5548",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(12454, 1, 15)\n",
      "(12454, 2)\n",
      "(769, 1, 15)\n",
      "(769, 2)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print(Y_train.shape)\n",
    "print(X_test.shape)\n",
    "print(Y_test.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "d9423511",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(12454, 1, 1, 15)\n",
      "(12454, 1, 2)\n",
      "(769, 1, 1, 15)\n",
      "(769, 1, 2)\n"
     ]
    }
   ],
   "source": [
    "X_train = np.expand_dims(X_train, axis=2)\n",
    "Y_train = np.expand_dims(Y_train, axis=1)\n",
    "X_test = np.expand_dims(X_test, axis=2)\n",
    "Y_test = np.expand_dims(Y_test, axis=1)\n",
    "print(X_train.shape)\n",
    "print(Y_train.shape)\n",
    "print(X_test.shape)\n",
    "print(Y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "962d20dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import Sequential\n",
    "from keras.layers import LSTM,Activation,Dense,Dropout,Input,Embedding,BatchNormalization,Add,concatenate,Flatten\n",
    "from keras.layers.recurrent import GRU\n",
    "from keras.models import Sequential\n",
    "from tensorflow.keras.layers import LSTM, Dense\n",
    "from tensorflow.keras.layers import RepeatVector, TimeDistributed,Conv1D,MaxPooling1D\n",
    "\n",
    "\n",
    "\n",
    "model = Sequential()\n",
    "model.add(TimeDistributed(Conv1D(filters=64, kernel_size=1, activation='relu', input_shape=(None,1,15))))\n",
    "model.add(TimeDistributed(MaxPooling1D(pool_size=1)))\n",
    "model.add(TimeDistributed(Flatten()))\n",
    "\n",
    "\n",
    "model.add(LSTM(units=50,return_sequences=True))\n",
    "#model.add(Dropout(0.25))\n",
    "\n",
    "model.add(LSTM(units=50,return_sequences=True))\n",
    "#model.add(Dropout(0.25))\n",
    "\n",
    "model.add(LSTM(units=50,return_sequences=True))\n",
    "#model.add(Dropout(0.25))\n",
    "model.add(LSTM(units=50,return_sequences=True))\n",
    "#model.add(Dropout(0.25))\n",
    "\n",
    " \n",
    "model.add(Dense(units=64))\n",
    "#model.add(Dropout(0.25))\n",
    "model.add(Dense(units=16))\n",
    "#model.add(Dropout(0.25))\n",
    "model.add(Dense(units=8))\n",
    "#model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Dense(units=2,activation=\"sigmoid\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "baf75a45",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Implement Learning rate decay\n",
    "from keras.callbacks import EarlyStopping,ReduceLROnPlateau,ModelCheckpoint,LearningRateScheduler\n",
    "\n",
    "checkpoint = ModelCheckpoint(\"./模型保存/CNN+LSTM9为第一位20230520.h5\",\n",
    "                             monitor=\"val_loss\",\n",
    "                             mode=\"min\",\n",
    "                             save_best_only = True,\n",
    "                             verbose=1)\n",
    "\n",
    "earlystop = EarlyStopping(monitor = 'val_loss', \n",
    "                          min_delta =0, \n",
    "                          patience = 5,\n",
    "                          verbose = 1,\n",
    "                          restore_best_weights = True)\n",
    "\n",
    "reduce_lr = ReduceLROnPlateau(monitor = 'val_loss',\n",
    "                              factor = 0.2,\n",
    "                              patience = 5,\n",
    "                              verbose = 1)\n",
    "                              #min_delta = 0.00001)\n",
    "\n",
    "callbacks = [earlystop, checkpoint, reduce_lr]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "5b6e22e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer=\"adam\", loss='categorical_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "0c9f13bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "1557/1557 [==============================] - 8s 3ms/step - loss: 0.1077 - accuracy: 0.9590 - val_loss: 0.0623 - val_accuracy: 0.9792\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.06225, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 2/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0533 - accuracy: 0.9819 - val_loss: 0.0443 - val_accuracy: 0.9831\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.06225 to 0.04430, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 3/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0422 - accuracy: 0.9847 - val_loss: 0.0339 - val_accuracy: 0.9857\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.04430 to 0.03390, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 4/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0372 - accuracy: 0.9868 - val_loss: 0.0388 - val_accuracy: 0.9870\n",
      "\n",
      "Epoch 00004: val_loss did not improve from 0.03390\n",
      "Epoch 5/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0348 - accuracy: 0.9863 - val_loss: 0.0304 - val_accuracy: 0.9831\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.03390 to 0.03042, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 6/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0298 - accuracy: 0.9888 - val_loss: 0.0278 - val_accuracy: 0.9909\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.03042 to 0.02780, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 7/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0272 - accuracy: 0.9905 - val_loss: 0.0167 - val_accuracy: 0.9961\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.02780 to 0.01674, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 8/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0268 - accuracy: 0.9896 - val_loss: 0.0200 - val_accuracy: 0.9922\n",
      "\n",
      "Epoch 00008: val_loss did not improve from 0.01674\n",
      "Epoch 9/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0243 - accuracy: 0.9917 - val_loss: 0.0192 - val_accuracy: 0.9948\n",
      "\n",
      "Epoch 00009: val_loss did not improve from 0.01674\n",
      "Epoch 10/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0240 - accuracy: 0.9914 - val_loss: 0.0214 - val_accuracy: 0.9948\n",
      "\n",
      "Epoch 00010: val_loss did not improve from 0.01674\n",
      "Epoch 11/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0218 - accuracy: 0.9923 - val_loss: 0.0206 - val_accuracy: 0.9961\n",
      "\n",
      "Epoch 00011: val_loss did not improve from 0.01674\n",
      "Epoch 12/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0194 - accuracy: 0.9930 - val_loss: 0.0162 - val_accuracy: 0.9974\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.01674 to 0.01620, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 13/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0196 - accuracy: 0.9933 - val_loss: 0.0274 - val_accuracy: 0.9883\n",
      "\n",
      "Epoch 00013: val_loss did not improve from 0.01620\n",
      "Epoch 14/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0174 - accuracy: 0.9938 - val_loss: 0.0124 - val_accuracy: 0.9961\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.01620 to 0.01242, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 15/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0181 - accuracy: 0.9933 - val_loss: 0.0182 - val_accuracy: 0.9961\n",
      "\n",
      "Epoch 00015: val_loss did not improve from 0.01242\n",
      "Epoch 16/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0177 - accuracy: 0.9939 - val_loss: 0.0229 - val_accuracy: 0.9948\n",
      "\n",
      "Epoch 00016: val_loss did not improve from 0.01242\n",
      "Epoch 17/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0164 - accuracy: 0.9944 - val_loss: 0.0222 - val_accuracy: 0.9948\n",
      "\n",
      "Epoch 00017: val_loss did not improve from 0.01242\n",
      "Epoch 18/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0142 - accuracy: 0.9955 - val_loss: 0.0110 - val_accuracy: 0.9974\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.01242 to 0.01098, saving model to ./模型保存\\CNN+LSTM9为第一位.h5\n",
      "Epoch 19/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0144 - accuracy: 0.9948 - val_loss: 0.0123 - val_accuracy: 0.9974\n",
      "\n",
      "Epoch 00019: val_loss did not improve from 0.01098\n",
      "Epoch 20/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0137 - accuracy: 0.9948 - val_loss: 0.0251 - val_accuracy: 0.9935\n",
      "\n",
      "Epoch 00020: val_loss did not improve from 0.01098\n",
      "Epoch 21/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0160 - accuracy: 0.9954 - val_loss: 0.0113 - val_accuracy: 0.9974\n",
      "\n",
      "Epoch 00021: val_loss did not improve from 0.01098\n",
      "Epoch 22/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0112 - accuracy: 0.9959 - val_loss: 0.0223 - val_accuracy: 0.9948\n",
      "\n",
      "Epoch 00022: val_loss did not improve from 0.01098\n",
      "Epoch 23/30\n",
      "1557/1557 [==============================] - 4s 2ms/step - loss: 0.0136 - accuracy: 0.9949 - val_loss: 0.0178 - val_accuracy: 0.9961\n",
      "Restoring model weights from the end of the best epoch.\n",
      "\n",
      "Epoch 00023: val_loss did not improve from 0.01098\n",
      "\n",
      "Epoch 00023: ReduceLROnPlateau reducing learning rate to 0.00020000000949949026.\n",
      "Epoch 00023: early stopping\n"
     ]
    }
   ],
   "source": [
    "history_fit = model.fit(x=X_train, \n",
    "                        y=Y_train, \n",
    "                        batch_size=8, \n",
    "                        epochs=30, \n",
    "                        verbose=1, \n",
    "                        validation_data=(X_test, Y_test),\n",
    "                        callbacks=callbacks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "ddb84c17",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 画曲线\n",
    "def plot_performance(history=None,figure_directory=None,ylim_pad=[0,0]):\n",
    "    xlabel=\"Epoch\"\n",
    "    legends=[\"Training\",\"Validation\"]\n",
    "    \n",
    "    plt.figure(figsize=(20,5))\n",
    "    \n",
    "    y1=history.history[\"accuracy\"]\n",
    "    y2=history.history[\"val_accuracy\"]\n",
    "    \n",
    "    min_y=min(min(y1),min(y2))-ylim_pad[0]\n",
    "    max_y=max(max(y1),max(y2))+ylim_pad[0]\n",
    "    \n",
    "    plt.subplot(121)\n",
    "    \n",
    "    plt.plot(y1)\n",
    "    plt.plot(y2)\n",
    "    \n",
    "    plt.title(\"Model Accuracy\\n\",fontsize=17)\n",
    "    plt.xlabel(xlabel,fontsize=15)\n",
    "    plt.ylabel(\"Accuracy\",fontsize=15)\n",
    "    plt.ylim(min_y,max_y)\n",
    "    plt.legend(legends,loc=\"upper left\")\n",
    "    plt.grid()\n",
    "    \n",
    "    y1=history.history[\"loss\"]\n",
    "    y2=history.history[\"val_loss\"]\n",
    "    \n",
    "    min_y=min(min(y1),min(y2))-ylim_pad[1]\n",
    "    max_y=max(max(y1),max(y2))+ylim_pad[1]\n",
    "    \n",
    "    plt.subplot(122)\n",
    "    \n",
    "    plt.plot(y1)\n",
    "    plt.plot(y2)\n",
    "    \n",
    "    plt.title(\"Model Loss:\\n\",fontsize=17)\n",
    "    plt.xlabel(xlabel,fontsize=15)\n",
    "    plt.ylabel(\"Loss\",fontsize=15)\n",
    "    plt.ylim(min_y,max_y)\n",
    "    plt.legend(legends,loc=\"upper left\")\n",
    "    plt.grid()\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "9ebd3b17",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 可视化\n",
    "plot_performance(history=history_fit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "0ffef768",
   "metadata": {},
   "outputs": [],
   "source": [
    "##导入保存模型\n",
    "#from keras import models \n",
    "#from keras.models import load_model\n",
    "#\n",
    "#\n",
    "#model = load_model('model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "e15e6533",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\SUN\\anaconda3\\envs\\LSTM\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\sequential.py:455: UserWarning: `model.predict_classes()` is deprecated and will be removed after 2021-01-01. Please use instead:* `np.argmax(model.predict(x), axis=-1)`,   if your model does multi-class classification   (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype(\"int32\")`,   if your model does binary classification   (e.g. if it uses a `sigmoid` last-layer activation).\n",
      "  warnings.warn('`model.predict_classes()` is deprecated and '\n"
     ]
    }
   ],
   "source": [
    "# 预测\n",
    "# predict_y = (model.predict(X_test) > 0.5)*1 \n",
    "# predict_y = np.around(predict_y) # np.around\n",
    "predict_y = model.predict_classes(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "2095b024",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0],\n",
       "       [0],\n",
       "       [1],\n",
       "       [0]], dtype=int64)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "4d3653c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0      0\n",
      "1      0\n",
      "2      0\n",
      "3      0\n",
      "4      0\n",
      "      ..\n",
      "764    1\n",
      "765    0\n",
      "766    0\n",
      "767    1\n",
      "768    0\n",
      "Name: Jammer, Length: 769, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "fffb1a15",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0     1.0000    0.9971    0.9985       687\n",
      "           1     0.9762    1.0000    0.9880        82\n",
      "\n",
      "    accuracy                         0.9974       769\n",
      "   macro avg     0.9881    0.9985    0.9932       769\n",
      "weighted avg     0.9975    0.9974    0.9974       769\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score,f1_score,confusion_matrix,classification_report\n",
    "\n",
    "print(classification_report(y_test,predict_y,digits=4))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "c719fb3b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[685   0]\n",
      " [  2  82]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "a = confusion_matrix(predict_y, y_test)\n",
    "print(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "03b6bdab",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PCA测试集准确率: 0.9974\n",
      "\n",
      "\n",
      "PCA测试集精确率: 1.0000\n",
      "\n",
      "\n",
      "PCA测试集召回率: 0.9762\n",
      "\n",
      "\n",
      "PCA测试集F1值: 0.9880\n"
     ]
    }
   ],
   "source": [
    "#模型预测\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "\n",
    "accuracy_pca = accuracy_score(predict_y, y_test)\n",
    "print(\"PCA测试集准确率: {0:.4f}\".format(accuracy_pca))\n",
    "\n",
    "precision_pca =precision_score(predict_y, y_test)\n",
    "print(\"\\n\\nPCA测试集精确率: {0:.4f}\".format(precision_pca))\n",
    "\n",
    "recall_pca =recall_score(predict_y, y_test)\n",
    "print(\"\\n\\nPCA测试集召回率: {0:.4f}\".format(recall_pca))\n",
    "\n",
    "f1_pca =f1_score(predict_y, y_test)\n",
    "print(\"\\n\\nPCA测试集F1值: {0:.4f}\".format(f1_pca))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "452ccc83",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "classes = list(set(y))\n",
    "classes.sort()\n",
    "confusion = confusion_matrix(predict_y, y_test)\n",
    "plt.imshow(confusion, cmap=plt.cm.Blues)\n",
    "indices = range(len(confusion))\n",
    "plt.xticks(indices, classes, fontsize=20)\n",
    "plt.yticks(indices, classes, fontsize=20)\n",
    "plt.colorbar().ax.tick_params(labelsize=20)\n",
    "plt.xlabel('Predicted label', fontsize=20)\n",
    "plt.ylabel('True label', fontsize=20)\n",
    "for first_index in range(len(confusion)):\n",
    "    for second_index in range(len(confusion[first_index])):\n",
    "        plt.text(first_index, second_index, confusion[first_index][second_index],\n",
    "                 va='center', ha='center',fontsize=20, )\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "4f8f636d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1]\n"
     ]
    },
    {
     "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": [
    "from sklearn.metrics import precision_recall_curve\n",
    "precision, recall, t =precision_recall_curve(y_test, predict_y) #y_score是预测的概率值，y_test是真实值标签值\n",
    "print(t) #t是阈值\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.ylim([0.0, 1.0])\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.plot(recall, precision)\n",
    "plt.title(\"Precision-Recall\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "80fbce7a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gpsPdop</th>\n",
       "      <th>gpsNumSv</th>\n",
       "      <th>gpsSec</th>\n",
       "      <th>P_Sol_Status</th>\n",
       "      <th>AvionicsTemp</th>\n",
       "      <th>Lat</th>\n",
       "      <th>Lon</th>\n",
       "      <th>H</th>\n",
       "      <th>tas</th>\n",
       "      <th>Eu[0]</th>\n",
       "      <th>...</th>\n",
       "      <th>elevatorCmd</th>\n",
       "      <th>rudderCmd</th>\n",
       "      <th>lonCmd</th>\n",
       "      <th>XaccBias</th>\n",
       "      <th>YaccBias</th>\n",
       "      <th>ZaccBias</th>\n",
       "      <th>AirHeight</th>\n",
       "      <th>WWest</th>\n",
       "      <th>WSouth</th>\n",
       "      <th>Jammer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>gpsPdop</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.234634</td>\n",
       "      <td>-0.045842</td>\n",
       "      <td>0.063523</td>\n",
       "      <td>0.178332</td>\n",
       "      <td>0.150694</td>\n",
       "      <td>0.161161</td>\n",
       "      <td>-0.121958</td>\n",
       "      <td>-0.137200</td>\n",
       "      <td>0.166629</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.191682</td>\n",
       "      <td>0.071393</td>\n",
       "      <td>0.018013</td>\n",
       "      <td>0.006322</td>\n",
       "      <td>-0.050946</td>\n",
       "      <td>0.042359</td>\n",
       "      <td>-0.143207</td>\n",
       "      <td>-0.066625</td>\n",
       "      <td>0.044012</td>\n",
       "      <td>0.162625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gpsNumSv</th>\n",
       "      <td>-0.234634</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.468722</td>\n",
       "      <td>-0.258311</td>\n",
       "      <td>-0.762855</td>\n",
       "      <td>-0.946884</td>\n",
       "      <td>-0.951386</td>\n",
       "      <td>0.719566</td>\n",
       "      <td>0.926694</td>\n",
       "      <td>-0.726410</td>\n",
       "      <td>...</td>\n",
       "      <td>0.862620</td>\n",
       "      <td>-0.499020</td>\n",
       "      <td>-0.116267</td>\n",
       "      <td>0.025268</td>\n",
       "      <td>0.257591</td>\n",
       "      <td>-0.339790</td>\n",
       "      <td>0.943205</td>\n",
       "      <td>0.563636</td>\n",
       "      <td>-0.480048</td>\n",
       "      <td>-0.846080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gpsSec</th>\n",
       "      <td>-0.045842</td>\n",
       "      <td>-0.468722</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.255043</td>\n",
       "      <td>0.096325</td>\n",
       "      <td>0.367168</td>\n",
       "      <td>0.320803</td>\n",
       "      <td>-0.213461</td>\n",
       "      <td>-0.394395</td>\n",
       "      <td>0.014606</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.090324</td>\n",
       "      <td>0.155759</td>\n",
       "      <td>0.048774</td>\n",
       "      <td>-0.140194</td>\n",
       "      <td>-0.167229</td>\n",
       "      <td>0.421922</td>\n",
       "      <td>-0.384407</td>\n",
       "      <td>-0.430665</td>\n",
       "      <td>0.076728</td>\n",
       "      <td>0.312062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>P_Sol_Status</th>\n",
       "      <td>0.063523</td>\n",
       "      <td>-0.258311</td>\n",
       "      <td>-0.255043</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.506207</td>\n",
       "      <td>0.403854</td>\n",
       "      <td>0.395404</td>\n",
       "      <td>-0.224357</td>\n",
       "      <td>-0.320000</td>\n",
       "      <td>0.474704</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.506260</td>\n",
       "      <td>0.212534</td>\n",
       "      <td>0.044415</td>\n",
       "      <td>0.126167</td>\n",
       "      <td>-0.130686</td>\n",
       "      <td>-0.171715</td>\n",
       "      <td>-0.319387</td>\n",
       "      <td>-0.175525</td>\n",
       "      <td>0.340965</td>\n",
       "      <td>0.458052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AvionicsTemp</th>\n",
       "      <td>0.178332</td>\n",
       "      <td>-0.762855</td>\n",
       "      <td>0.096325</td>\n",
       "      <td>0.506207</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.794126</td>\n",
       "      <td>0.849744</td>\n",
       "      <td>-0.695607</td>\n",
       "      <td>-0.742844</td>\n",
       "      <td>0.808681</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.902229</td>\n",
       "      <td>0.456009</td>\n",
       "      <td>0.101640</td>\n",
       "      <td>0.041917</td>\n",
       "      <td>-0.298337</td>\n",
       "      <td>0.183141</td>\n",
       "      <td>-0.762237</td>\n",
       "      <td>-0.504123</td>\n",
       "      <td>0.483972</td>\n",
       "      <td>0.726737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lat</th>\n",
       "      <td>0.150694</td>\n",
       "      <td>-0.946884</td>\n",
       "      <td>0.367168</td>\n",
       "      <td>0.403854</td>\n",
       "      <td>0.794126</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.979929</td>\n",
       "      <td>-0.655615</td>\n",
       "      <td>-0.948755</td>\n",
       "      <td>0.791329</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.921223</td>\n",
       "      <td>0.531221</td>\n",
       "      <td>0.114726</td>\n",
       "      <td>-0.010135</td>\n",
       "      <td>-0.256051</td>\n",
       "      <td>0.333126</td>\n",
       "      <td>-0.963945</td>\n",
       "      <td>-0.559601</td>\n",
       "      <td>0.508046</td>\n",
       "      <td>0.853407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lon</th>\n",
       "      <td>0.161161</td>\n",
       "      <td>-0.951386</td>\n",
       "      <td>0.320803</td>\n",
       "      <td>0.395404</td>\n",
       "      <td>0.849744</td>\n",
       "      <td>0.979929</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.774480</td>\n",
       "      <td>-0.951897</td>\n",
       "      <td>0.825410</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.954003</td>\n",
       "      <td>0.543721</td>\n",
       "      <td>0.110886</td>\n",
       "      <td>-0.008930</td>\n",
       "      <td>-0.260223</td>\n",
       "      <td>0.336020</td>\n",
       "      <td>-0.971546</td>\n",
       "      <td>-0.555436</td>\n",
       "      <td>0.527449</td>\n",
       "      <td>0.856196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>H</th>\n",
       "      <td>-0.121958</td>\n",
       "      <td>0.719566</td>\n",
       "      <td>-0.213461</td>\n",
       "      <td>-0.224357</td>\n",
       "      <td>-0.695607</td>\n",
       "      <td>-0.655615</td>\n",
       "      <td>-0.774480</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.737802</td>\n",
       "      <td>-0.653549</td>\n",
       "      <td>...</td>\n",
       "      <td>0.727669</td>\n",
       "      <td>-0.460461</td>\n",
       "      <td>-0.010743</td>\n",
       "      <td>0.009332</td>\n",
       "      <td>0.211839</td>\n",
       "      <td>-0.275919</td>\n",
       "      <td>0.757811</td>\n",
       "      <td>0.399980</td>\n",
       "      <td>-0.435937</td>\n",
       "      <td>-0.658316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tas</th>\n",
       "      <td>-0.137200</td>\n",
       "      <td>0.926694</td>\n",
       "      <td>-0.394395</td>\n",
       "      <td>-0.320000</td>\n",
       "      <td>-0.742844</td>\n",
       "      <td>-0.948755</td>\n",
       "      <td>-0.951897</td>\n",
       "      <td>0.737802</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.769987</td>\n",
       "      <td>...</td>\n",
       "      <td>0.865022</td>\n",
       "      <td>-0.572597</td>\n",
       "      <td>0.028993</td>\n",
       "      <td>0.033232</td>\n",
       "      <td>0.271552</td>\n",
       "      <td>-0.342690</td>\n",
       "      <td>0.982895</td>\n",
       "      <td>0.550253</td>\n",
       "      <td>-0.510074</td>\n",
       "      <td>-0.817636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Eu[0]</th>\n",
       "      <td>0.166629</td>\n",
       "      <td>-0.726410</td>\n",
       "      <td>0.014606</td>\n",
       "      <td>0.474704</td>\n",
       "      <td>0.808681</td>\n",
       "      <td>0.791329</td>\n",
       "      <td>0.825410</td>\n",
       "      <td>-0.653549</td>\n",
       "      <td>-0.769987</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.838015</td>\n",
       "      <td>0.747374</td>\n",
       "      <td>0.034341</td>\n",
       "      <td>0.019134</td>\n",
       "      <td>-0.271707</td>\n",
       "      <td>0.155659</td>\n",
       "      <td>-0.786663</td>\n",
       "      <td>-0.380390</td>\n",
       "      <td>0.463749</td>\n",
       "      <td>0.696942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Eu[1]</th>\n",
       "      <td>-0.186365</td>\n",
       "      <td>0.883724</td>\n",
       "      <td>-0.137515</td>\n",
       "      <td>-0.487213</td>\n",
       "      <td>-0.892004</td>\n",
       "      <td>-0.938013</td>\n",
       "      <td>-0.966955</td>\n",
       "      <td>0.739809</td>\n",
       "      <td>0.895173</td>\n",
       "      <td>-0.849015</td>\n",
       "      <td>...</td>\n",
       "      <td>0.983930</td>\n",
       "      <td>-0.507337</td>\n",
       "      <td>-0.100711</td>\n",
       "      <td>-0.036434</td>\n",
       "      <td>0.256058</td>\n",
       "      <td>-0.257257</td>\n",
       "      <td>0.912889</td>\n",
       "      <td>0.534357</td>\n",
       "      <td>-0.544507</td>\n",
       "      <td>-0.822157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Eu[2]</th>\n",
       "      <td>-0.006212</td>\n",
       "      <td>0.059529</td>\n",
       "      <td>-0.035913</td>\n",
       "      <td>-0.059183</td>\n",
       "      <td>0.029262</td>\n",
       "      <td>-0.076414</td>\n",
       "      <td>-0.033865</td>\n",
       "      <td>-0.040091</td>\n",
       "      <td>0.054414</td>\n",
       "      <td>-0.032697</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.001914</td>\n",
       "      <td>-0.014049</td>\n",
       "      <td>0.022056</td>\n",
       "      <td>0.084570</td>\n",
       "      <td>-0.020556</td>\n",
       "      <td>0.133792</td>\n",
       "      <td>0.042956</td>\n",
       "      <td>-0.229469</td>\n",
       "      <td>0.089768</td>\n",
       "      <td>-0.029657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>acc[0]</th>\n",
       "      <td>-0.181658</td>\n",
       "      <td>0.863275</td>\n",
       "      <td>-0.115037</td>\n",
       "      <td>-0.491100</td>\n",
       "      <td>-0.883101</td>\n",
       "      <td>-0.920491</td>\n",
       "      <td>-0.950126</td>\n",
       "      <td>0.728101</td>\n",
       "      <td>0.878002</td>\n",
       "      <td>-0.846259</td>\n",
       "      <td>...</td>\n",
       "      <td>0.968641</td>\n",
       "      <td>-0.505431</td>\n",
       "      <td>-0.100408</td>\n",
       "      <td>-0.040527</td>\n",
       "      <td>0.254352</td>\n",
       "      <td>-0.248915</td>\n",
       "      <td>0.897502</td>\n",
       "      <td>0.519453</td>\n",
       "      <td>-0.544480</td>\n",
       "      <td>-0.806853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>acc[1]</th>\n",
       "      <td>-0.133515</td>\n",
       "      <td>0.472060</td>\n",
       "      <td>0.124859</td>\n",
       "      <td>-0.419335</td>\n",
       "      <td>-0.649911</td>\n",
       "      <td>-0.523242</td>\n",
       "      <td>-0.565995</td>\n",
       "      <td>0.453089</td>\n",
       "      <td>0.474938</td>\n",
       "      <td>-0.596491</td>\n",
       "      <td>...</td>\n",
       "      <td>0.650110</td>\n",
       "      <td>-0.320327</td>\n",
       "      <td>-0.071030</td>\n",
       "      <td>-0.042423</td>\n",
       "      <td>0.160317</td>\n",
       "      <td>-0.060052</td>\n",
       "      <td>0.483672</td>\n",
       "      <td>0.300722</td>\n",
       "      <td>-0.378389</td>\n",
       "      <td>-0.458647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>acc[2]</th>\n",
       "      <td>0.116198</td>\n",
       "      <td>-0.428461</td>\n",
       "      <td>-0.066448</td>\n",
       "      <td>0.349478</td>\n",
       "      <td>0.542615</td>\n",
       "      <td>0.482773</td>\n",
       "      <td>0.511992</td>\n",
       "      <td>-0.398267</td>\n",
       "      <td>-0.456695</td>\n",
       "      <td>0.548967</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.553422</td>\n",
       "      <td>0.320730</td>\n",
       "      <td>0.040082</td>\n",
       "      <td>0.047136</td>\n",
       "      <td>-0.158328</td>\n",
       "      <td>0.071159</td>\n",
       "      <td>-0.459004</td>\n",
       "      <td>-0.249704</td>\n",
       "      <td>0.322840</td>\n",
       "      <td>0.428079</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pqr[0]</th>\n",
       "      <td>-0.091475</td>\n",
       "      <td>0.176002</td>\n",
       "      <td>0.352176</td>\n",
       "      <td>-0.392650</td>\n",
       "      <td>-0.433812</td>\n",
       "      <td>-0.247013</td>\n",
       "      <td>-0.288959</td>\n",
       "      <td>0.228628</td>\n",
       "      <td>0.208407</td>\n",
       "      <td>-0.390577</td>\n",
       "      <td>...</td>\n",
       "      <td>0.416900</td>\n",
       "      <td>-0.134908</td>\n",
       "      <td>-0.033596</td>\n",
       "      <td>-0.062480</td>\n",
       "      <td>0.092983</td>\n",
       "      <td>0.067977</td>\n",
       "      <td>0.210589</td>\n",
       "      <td>0.123588</td>\n",
       "      <td>-0.287966</td>\n",
       "      <td>-0.247045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pqr[1]</th>\n",
       "      <td>-0.081389</td>\n",
       "      <td>0.318671</td>\n",
       "      <td>0.114452</td>\n",
       "      <td>-0.322354</td>\n",
       "      <td>-0.454104</td>\n",
       "      <td>-0.378392</td>\n",
       "      <td>-0.404666</td>\n",
       "      <td>0.317522</td>\n",
       "      <td>0.370475</td>\n",
       "      <td>-0.556371</td>\n",
       "      <td>...</td>\n",
       "      <td>0.429816</td>\n",
       "      <td>-0.384863</td>\n",
       "      <td>-0.012337</td>\n",
       "      <td>-0.034795</td>\n",
       "      <td>0.153046</td>\n",
       "      <td>-0.029754</td>\n",
       "      <td>0.363021</td>\n",
       "      <td>0.201089</td>\n",
       "      <td>-0.281464</td>\n",
       "      <td>-0.346322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pqr[2]</th>\n",
       "      <td>-0.038703</td>\n",
       "      <td>-0.217393</td>\n",
       "      <td>0.424826</td>\n",
       "      <td>-0.199205</td>\n",
       "      <td>-0.060487</td>\n",
       "      <td>0.181097</td>\n",
       "      <td>0.150396</td>\n",
       "      <td>-0.130628</td>\n",
       "      <td>-0.226383</td>\n",
       "      <td>0.243355</td>\n",
       "      <td>...</td>\n",
       "      <td>0.022819</td>\n",
       "      <td>0.275730</td>\n",
       "      <td>-0.042655</td>\n",
       "      <td>-0.069243</td>\n",
       "      <td>-0.062777</td>\n",
       "      <td>0.186169</td>\n",
       "      <td>-0.240843</td>\n",
       "      <td>-0.048606</td>\n",
       "      <td>-0.047962</td>\n",
       "      <td>0.129898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InertVNED[0]</th>\n",
       "      <td>0.012898</td>\n",
       "      <td>-0.067304</td>\n",
       "      <td>-0.021144</td>\n",
       "      <td>0.105986</td>\n",
       "      <td>0.094051</td>\n",
       "      <td>0.091529</td>\n",
       "      <td>0.107814</td>\n",
       "      <td>-0.088667</td>\n",
       "      <td>-0.092385</td>\n",
       "      <td>0.149811</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.101043</td>\n",
       "      <td>0.144381</td>\n",
       "      <td>0.001304</td>\n",
       "      <td>0.047118</td>\n",
       "      <td>-0.072044</td>\n",
       "      <td>-0.042087</td>\n",
       "      <td>-0.091015</td>\n",
       "      <td>0.042908</td>\n",
       "      <td>0.135644</td>\n",
       "      <td>0.117072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InertVNED[1]</th>\n",
       "      <td>-0.013160</td>\n",
       "      <td>0.039457</td>\n",
       "      <td>-0.079274</td>\n",
       "      <td>0.000269</td>\n",
       "      <td>-0.038155</td>\n",
       "      <td>-0.075998</td>\n",
       "      <td>-0.056936</td>\n",
       "      <td>0.056664</td>\n",
       "      <td>0.058302</td>\n",
       "      <td>-0.116978</td>\n",
       "      <td>...</td>\n",
       "      <td>0.024111</td>\n",
       "      <td>-0.161352</td>\n",
       "      <td>0.007612</td>\n",
       "      <td>-0.012743</td>\n",
       "      <td>0.063161</td>\n",
       "      <td>-0.059165</td>\n",
       "      <td>0.063507</td>\n",
       "      <td>0.067592</td>\n",
       "      <td>-0.077510</td>\n",
       "      <td>-0.067917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InertVNED[2]</th>\n",
       "      <td>0.044930</td>\n",
       "      <td>-0.237525</td>\n",
       "      <td>0.049916</td>\n",
       "      <td>0.153865</td>\n",
       "      <td>0.237377</td>\n",
       "      <td>0.259884</td>\n",
       "      <td>0.268470</td>\n",
       "      <td>-0.209451</td>\n",
       "      <td>-0.248107</td>\n",
       "      <td>0.185940</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.304770</td>\n",
       "      <td>0.087234</td>\n",
       "      <td>0.033822</td>\n",
       "      <td>0.118643</td>\n",
       "      <td>-0.023252</td>\n",
       "      <td>0.073167</td>\n",
       "      <td>-0.254739</td>\n",
       "      <td>-0.138344</td>\n",
       "      <td>0.142929</td>\n",
       "      <td>0.241911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DynPressure</th>\n",
       "      <td>-0.148412</td>\n",
       "      <td>0.912439</td>\n",
       "      <td>-0.335461</td>\n",
       "      <td>-0.364083</td>\n",
       "      <td>-0.785045</td>\n",
       "      <td>-0.941745</td>\n",
       "      <td>-0.952959</td>\n",
       "      <td>0.750785</td>\n",
       "      <td>0.988745</td>\n",
       "      <td>-0.794300</td>\n",
       "      <td>...</td>\n",
       "      <td>0.891579</td>\n",
       "      <td>-0.576338</td>\n",
       "      <td>0.050645</td>\n",
       "      <td>0.019561</td>\n",
       "      <td>0.276626</td>\n",
       "      <td>-0.331147</td>\n",
       "      <td>0.971854</td>\n",
       "      <td>0.547073</td>\n",
       "      <td>-0.513894</td>\n",
       "      <td>-0.818833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XMagFiled</th>\n",
       "      <td>0.029594</td>\n",
       "      <td>-0.120539</td>\n",
       "      <td>-0.074899</td>\n",
       "      <td>0.127495</td>\n",
       "      <td>0.197762</td>\n",
       "      <td>0.178365</td>\n",
       "      <td>0.206027</td>\n",
       "      <td>-0.164489</td>\n",
       "      <td>-0.172284</td>\n",
       "      <td>0.232712</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.231348</td>\n",
       "      <td>0.192570</td>\n",
       "      <td>-0.007971</td>\n",
       "      <td>0.072167</td>\n",
       "      <td>-0.103479</td>\n",
       "      <td>0.102863</td>\n",
       "      <td>-0.179030</td>\n",
       "      <td>-0.118307</td>\n",
       "      <td>0.104257</td>\n",
       "      <td>0.161257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YMagFiled</th>\n",
       "      <td>0.134108</td>\n",
       "      <td>-0.576498</td>\n",
       "      <td>0.173119</td>\n",
       "      <td>0.307716</td>\n",
       "      <td>0.495371</td>\n",
       "      <td>0.619893</td>\n",
       "      <td>0.603347</td>\n",
       "      <td>-0.439258</td>\n",
       "      <td>-0.597344</td>\n",
       "      <td>0.679881</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.544057</td>\n",
       "      <td>0.545680</td>\n",
       "      <td>-0.015397</td>\n",
       "      <td>-0.069716</td>\n",
       "      <td>-0.169362</td>\n",
       "      <td>0.065594</td>\n",
       "      <td>-0.607403</td>\n",
       "      <td>-0.116913</td>\n",
       "      <td>0.156086</td>\n",
       "      <td>0.511110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZMagFiled</th>\n",
       "      <td>0.014747</td>\n",
       "      <td>-0.520065</td>\n",
       "      <td>0.453739</td>\n",
       "      <td>-0.027390</td>\n",
       "      <td>0.266027</td>\n",
       "      <td>0.500066</td>\n",
       "      <td>0.470131</td>\n",
       "      <td>-0.353320</td>\n",
       "      <td>-0.509516</td>\n",
       "      <td>0.372830</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.322083</td>\n",
       "      <td>0.369452</td>\n",
       "      <td>0.042962</td>\n",
       "      <td>-0.034674</td>\n",
       "      <td>-0.149854</td>\n",
       "      <td>0.253668</td>\n",
       "      <td>-0.506638</td>\n",
       "      <td>-0.103305</td>\n",
       "      <td>0.274983</td>\n",
       "      <td>0.429471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Compass</th>\n",
       "      <td>-0.067922</td>\n",
       "      <td>0.131683</td>\n",
       "      <td>-0.123788</td>\n",
       "      <td>-0.103775</td>\n",
       "      <td>-0.040248</td>\n",
       "      <td>-0.144104</td>\n",
       "      <td>-0.104868</td>\n",
       "      <td>0.020593</td>\n",
       "      <td>0.113908</td>\n",
       "      <td>-0.074208</td>\n",
       "      <td>...</td>\n",
       "      <td>0.061503</td>\n",
       "      <td>-0.043060</td>\n",
       "      <td>0.010892</td>\n",
       "      <td>0.096169</td>\n",
       "      <td>-0.004775</td>\n",
       "      <td>0.105515</td>\n",
       "      <td>0.112166</td>\n",
       "      <td>-0.109873</td>\n",
       "      <td>0.213558</td>\n",
       "      <td>-0.083610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>aileronLeftCmd</th>\n",
       "      <td>-0.183675</td>\n",
       "      <td>0.745256</td>\n",
       "      <td>0.062447</td>\n",
       "      <td>-0.533541</td>\n",
       "      <td>-0.855085</td>\n",
       "      <td>-0.812223</td>\n",
       "      <td>-0.851095</td>\n",
       "      <td>0.666255</td>\n",
       "      <td>0.770325</td>\n",
       "      <td>-0.859083</td>\n",
       "      <td>...</td>\n",
       "      <td>0.908012</td>\n",
       "      <td>-0.449528</td>\n",
       "      <td>-0.057363</td>\n",
       "      <td>-0.062840</td>\n",
       "      <td>0.233826</td>\n",
       "      <td>-0.123520</td>\n",
       "      <td>0.788632</td>\n",
       "      <td>0.393943</td>\n",
       "      <td>-0.476577</td>\n",
       "      <td>-0.716213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>elevatorCmd</th>\n",
       "      <td>-0.191682</td>\n",
       "      <td>0.862620</td>\n",
       "      <td>-0.090324</td>\n",
       "      <td>-0.506260</td>\n",
       "      <td>-0.902229</td>\n",
       "      <td>-0.921223</td>\n",
       "      <td>-0.954003</td>\n",
       "      <td>0.727669</td>\n",
       "      <td>0.865022</td>\n",
       "      <td>-0.838015</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.483568</td>\n",
       "      <td>-0.135234</td>\n",
       "      <td>-0.041515</td>\n",
       "      <td>0.250049</td>\n",
       "      <td>-0.235980</td>\n",
       "      <td>0.889678</td>\n",
       "      <td>0.510373</td>\n",
       "      <td>-0.538591</td>\n",
       "      <td>-0.808793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rudderCmd</th>\n",
       "      <td>0.071393</td>\n",
       "      <td>-0.499020</td>\n",
       "      <td>0.155759</td>\n",
       "      <td>0.212534</td>\n",
       "      <td>0.456009</td>\n",
       "      <td>0.531221</td>\n",
       "      <td>0.543721</td>\n",
       "      <td>-0.460461</td>\n",
       "      <td>-0.572597</td>\n",
       "      <td>0.747374</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.483568</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.199211</td>\n",
       "      <td>-0.055551</td>\n",
       "      <td>-0.231455</td>\n",
       "      <td>0.200277</td>\n",
       "      <td>-0.580402</td>\n",
       "      <td>-0.267558</td>\n",
       "      <td>0.290256</td>\n",
       "      <td>0.463255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lonCmd</th>\n",
       "      <td>0.018013</td>\n",
       "      <td>-0.116267</td>\n",
       "      <td>0.048774</td>\n",
       "      <td>0.044415</td>\n",
       "      <td>0.101640</td>\n",
       "      <td>0.114726</td>\n",
       "      <td>0.110886</td>\n",
       "      <td>-0.010743</td>\n",
       "      <td>0.028993</td>\n",
       "      <td>0.034341</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.135234</td>\n",
       "      <td>-0.199211</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.030313</td>\n",
       "      <td>0.049329</td>\n",
       "      <td>0.037793</td>\n",
       "      <td>-0.022211</td>\n",
       "      <td>-0.052881</td>\n",
       "      <td>0.028370</td>\n",
       "      <td>0.095710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>XaccBias</th>\n",
       "      <td>0.006322</td>\n",
       "      <td>0.025268</td>\n",
       "      <td>-0.140194</td>\n",
       "      <td>0.126167</td>\n",
       "      <td>0.041917</td>\n",
       "      <td>-0.010135</td>\n",
       "      <td>-0.008930</td>\n",
       "      <td>0.009332</td>\n",
       "      <td>0.033232</td>\n",
       "      <td>0.019134</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.041515</td>\n",
       "      <td>-0.055551</td>\n",
       "      <td>-0.030313</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.408730</td>\n",
       "      <td>-0.552741</td>\n",
       "      <td>0.006759</td>\n",
       "      <td>0.000402</td>\n",
       "      <td>0.146351</td>\n",
       "      <td>0.099195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YaccBias</th>\n",
       "      <td>-0.050946</td>\n",
       "      <td>0.257591</td>\n",
       "      <td>-0.167229</td>\n",
       "      <td>-0.130686</td>\n",
       "      <td>-0.298337</td>\n",
       "      <td>-0.256051</td>\n",
       "      <td>-0.260223</td>\n",
       "      <td>0.211839</td>\n",
       "      <td>0.271552</td>\n",
       "      <td>-0.271707</td>\n",
       "      <td>...</td>\n",
       "      <td>0.250049</td>\n",
       "      <td>-0.231455</td>\n",
       "      <td>0.049329</td>\n",
       "      <td>-0.408730</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.054403</td>\n",
       "      <td>0.278911</td>\n",
       "      <td>0.292403</td>\n",
       "      <td>-0.199916</td>\n",
       "      <td>-0.283392</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZaccBias</th>\n",
       "      <td>0.042359</td>\n",
       "      <td>-0.339790</td>\n",
       "      <td>0.421922</td>\n",
       "      <td>-0.171715</td>\n",
       "      <td>0.183141</td>\n",
       "      <td>0.333126</td>\n",
       "      <td>0.336020</td>\n",
       "      <td>-0.275919</td>\n",
       "      <td>-0.342690</td>\n",
       "      <td>0.155659</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.235980</td>\n",
       "      <td>0.200277</td>\n",
       "      <td>0.037793</td>\n",
       "      <td>-0.552741</td>\n",
       "      <td>0.054403</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.340498</td>\n",
       "      <td>-0.365933</td>\n",
       "      <td>0.044133</td>\n",
       "      <td>0.218115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AirHeight</th>\n",
       "      <td>-0.143207</td>\n",
       "      <td>0.943205</td>\n",
       "      <td>-0.384407</td>\n",
       "      <td>-0.319387</td>\n",
       "      <td>-0.762237</td>\n",
       "      <td>-0.963945</td>\n",
       "      <td>-0.971546</td>\n",
       "      <td>0.757811</td>\n",
       "      <td>0.982895</td>\n",
       "      <td>-0.786663</td>\n",
       "      <td>...</td>\n",
       "      <td>0.889678</td>\n",
       "      <td>-0.580402</td>\n",
       "      <td>-0.022211</td>\n",
       "      <td>0.006759</td>\n",
       "      <td>0.278911</td>\n",
       "      <td>-0.340498</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.537606</td>\n",
       "      <td>-0.503144</td>\n",
       "      <td>-0.843884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WWest</th>\n",
       "      <td>-0.066625</td>\n",
       "      <td>0.563636</td>\n",
       "      <td>-0.430665</td>\n",
       "      <td>-0.175525</td>\n",
       "      <td>-0.504123</td>\n",
       "      <td>-0.559601</td>\n",
       "      <td>-0.555436</td>\n",
       "      <td>0.399980</td>\n",
       "      <td>0.550253</td>\n",
       "      <td>-0.380390</td>\n",
       "      <td>...</td>\n",
       "      <td>0.510373</td>\n",
       "      <td>-0.267558</td>\n",
       "      <td>-0.052881</td>\n",
       "      <td>0.000402</td>\n",
       "      <td>0.292403</td>\n",
       "      <td>-0.365933</td>\n",
       "      <td>0.537606</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.510738</td>\n",
       "      <td>-0.539402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WSouth</th>\n",
       "      <td>0.044012</td>\n",
       "      <td>-0.480048</td>\n",
       "      <td>0.076728</td>\n",
       "      <td>0.340965</td>\n",
       "      <td>0.483972</td>\n",
       "      <td>0.508046</td>\n",
       "      <td>0.527449</td>\n",
       "      <td>-0.435937</td>\n",
       "      <td>-0.510074</td>\n",
       "      <td>0.463749</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.538591</td>\n",
       "      <td>0.290256</td>\n",
       "      <td>0.028370</td>\n",
       "      <td>0.146351</td>\n",
       "      <td>-0.199916</td>\n",
       "      <td>0.044133</td>\n",
       "      <td>-0.503144</td>\n",
       "      <td>-0.510738</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.559624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Jammer</th>\n",
       "      <td>0.162625</td>\n",
       "      <td>-0.846080</td>\n",
       "      <td>0.312062</td>\n",
       "      <td>0.458052</td>\n",
       "      <td>0.726737</td>\n",
       "      <td>0.853407</td>\n",
       "      <td>0.856196</td>\n",
       "      <td>-0.658316</td>\n",
       "      <td>-0.817636</td>\n",
       "      <td>0.696942</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.808793</td>\n",
       "      <td>0.463255</td>\n",
       "      <td>0.095710</td>\n",
       "      <td>0.099195</td>\n",
       "      <td>-0.283392</td>\n",
       "      <td>0.218115</td>\n",
       "      <td>-0.843884</td>\n",
       "      <td>-0.539402</td>\n",
       "      <td>0.559624</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>37 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 gpsPdop  gpsNumSv    gpsSec  P_Sol_Status  AvionicsTemp  \\\n",
       "gpsPdop         1.000000 -0.234634 -0.045842      0.063523      0.178332   \n",
       "gpsNumSv       -0.234634  1.000000 -0.468722     -0.258311     -0.762855   \n",
       "gpsSec         -0.045842 -0.468722  1.000000     -0.255043      0.096325   \n",
       "P_Sol_Status    0.063523 -0.258311 -0.255043      1.000000      0.506207   \n",
       "AvionicsTemp    0.178332 -0.762855  0.096325      0.506207      1.000000   \n",
       "Lat             0.150694 -0.946884  0.367168      0.403854      0.794126   \n",
       "Lon             0.161161 -0.951386  0.320803      0.395404      0.849744   \n",
       "H              -0.121958  0.719566 -0.213461     -0.224357     -0.695607   \n",
       "tas            -0.137200  0.926694 -0.394395     -0.320000     -0.742844   \n",
       "Eu[0]           0.166629 -0.726410  0.014606      0.474704      0.808681   \n",
       "Eu[1]          -0.186365  0.883724 -0.137515     -0.487213     -0.892004   \n",
       "Eu[2]          -0.006212  0.059529 -0.035913     -0.059183      0.029262   \n",
       "acc[0]         -0.181658  0.863275 -0.115037     -0.491100     -0.883101   \n",
       "acc[1]         -0.133515  0.472060  0.124859     -0.419335     -0.649911   \n",
       "acc[2]          0.116198 -0.428461 -0.066448      0.349478      0.542615   \n",
       "pqr[0]         -0.091475  0.176002  0.352176     -0.392650     -0.433812   \n",
       "pqr[1]         -0.081389  0.318671  0.114452     -0.322354     -0.454104   \n",
       "pqr[2]         -0.038703 -0.217393  0.424826     -0.199205     -0.060487   \n",
       "InertVNED[0]    0.012898 -0.067304 -0.021144      0.105986      0.094051   \n",
       "InertVNED[1]   -0.013160  0.039457 -0.079274      0.000269     -0.038155   \n",
       "InertVNED[2]    0.044930 -0.237525  0.049916      0.153865      0.237377   \n",
       "DynPressure    -0.148412  0.912439 -0.335461     -0.364083     -0.785045   \n",
       "XMagFiled       0.029594 -0.120539 -0.074899      0.127495      0.197762   \n",
       "YMagFiled       0.134108 -0.576498  0.173119      0.307716      0.495371   \n",
       "ZMagFiled       0.014747 -0.520065  0.453739     -0.027390      0.266027   \n",
       "Compass        -0.067922  0.131683 -0.123788     -0.103775     -0.040248   \n",
       "aileronLeftCmd -0.183675  0.745256  0.062447     -0.533541     -0.855085   \n",
       "elevatorCmd    -0.191682  0.862620 -0.090324     -0.506260     -0.902229   \n",
       "rudderCmd       0.071393 -0.499020  0.155759      0.212534      0.456009   \n",
       "lonCmd          0.018013 -0.116267  0.048774      0.044415      0.101640   \n",
       "XaccBias        0.006322  0.025268 -0.140194      0.126167      0.041917   \n",
       "YaccBias       -0.050946  0.257591 -0.167229     -0.130686     -0.298337   \n",
       "ZaccBias        0.042359 -0.339790  0.421922     -0.171715      0.183141   \n",
       "AirHeight      -0.143207  0.943205 -0.384407     -0.319387     -0.762237   \n",
       "WWest          -0.066625  0.563636 -0.430665     -0.175525     -0.504123   \n",
       "WSouth          0.044012 -0.480048  0.076728      0.340965      0.483972   \n",
       "Jammer          0.162625 -0.846080  0.312062      0.458052      0.726737   \n",
       "\n",
       "                     Lat       Lon         H       tas     Eu[0]  ...  \\\n",
       "gpsPdop         0.150694  0.161161 -0.121958 -0.137200  0.166629  ...   \n",
       "gpsNumSv       -0.946884 -0.951386  0.719566  0.926694 -0.726410  ...   \n",
       "gpsSec          0.367168  0.320803 -0.213461 -0.394395  0.014606  ...   \n",
       "P_Sol_Status    0.403854  0.395404 -0.224357 -0.320000  0.474704  ...   \n",
       "AvionicsTemp    0.794126  0.849744 -0.695607 -0.742844  0.808681  ...   \n",
       "Lat             1.000000  0.979929 -0.655615 -0.948755  0.791329  ...   \n",
       "Lon             0.979929  1.000000 -0.774480 -0.951897  0.825410  ...   \n",
       "H              -0.655615 -0.774480  1.000000  0.737802 -0.653549  ...   \n",
       "tas            -0.948755 -0.951897  0.737802  1.000000 -0.769987  ...   \n",
       "Eu[0]           0.791329  0.825410 -0.653549 -0.769987  1.000000  ...   \n",
       "Eu[1]          -0.938013 -0.966955  0.739809  0.895173 -0.849015  ...   \n",
       "Eu[2]          -0.076414 -0.033865 -0.040091  0.054414 -0.032697  ...   \n",
       "acc[0]         -0.920491 -0.950126  0.728101  0.878002 -0.846259  ...   \n",
       "acc[1]         -0.523242 -0.565995  0.453089  0.474938 -0.596491  ...   \n",
       "acc[2]          0.482773  0.511992 -0.398267 -0.456695  0.548967  ...   \n",
       "pqr[0]         -0.247013 -0.288959  0.228628  0.208407 -0.390577  ...   \n",
       "pqr[1]         -0.378392 -0.404666  0.317522  0.370475 -0.556371  ...   \n",
       "pqr[2]          0.181097  0.150396 -0.130628 -0.226383  0.243355  ...   \n",
       "InertVNED[0]    0.091529  0.107814 -0.088667 -0.092385  0.149811  ...   \n",
       "InertVNED[1]   -0.075998 -0.056936  0.056664  0.058302 -0.116978  ...   \n",
       "InertVNED[2]    0.259884  0.268470 -0.209451 -0.248107  0.185940  ...   \n",
       "DynPressure    -0.941745 -0.952959  0.750785  0.988745 -0.794300  ...   \n",
       "XMagFiled       0.178365  0.206027 -0.164489 -0.172284  0.232712  ...   \n",
       "YMagFiled       0.619893  0.603347 -0.439258 -0.597344  0.679881  ...   \n",
       "ZMagFiled       0.500066  0.470131 -0.353320 -0.509516  0.372830  ...   \n",
       "Compass        -0.144104 -0.104868  0.020593  0.113908 -0.074208  ...   \n",
       "aileronLeftCmd -0.812223 -0.851095  0.666255  0.770325 -0.859083  ...   \n",
       "elevatorCmd    -0.921223 -0.954003  0.727669  0.865022 -0.838015  ...   \n",
       "rudderCmd       0.531221  0.543721 -0.460461 -0.572597  0.747374  ...   \n",
       "lonCmd          0.114726  0.110886 -0.010743  0.028993  0.034341  ...   \n",
       "XaccBias       -0.010135 -0.008930  0.009332  0.033232  0.019134  ...   \n",
       "YaccBias       -0.256051 -0.260223  0.211839  0.271552 -0.271707  ...   \n",
       "ZaccBias        0.333126  0.336020 -0.275919 -0.342690  0.155659  ...   \n",
       "AirHeight      -0.963945 -0.971546  0.757811  0.982895 -0.786663  ...   \n",
       "WWest          -0.559601 -0.555436  0.399980  0.550253 -0.380390  ...   \n",
       "WSouth          0.508046  0.527449 -0.435937 -0.510074  0.463749  ...   \n",
       "Jammer          0.853407  0.856196 -0.658316 -0.817636  0.696942  ...   \n",
       "\n",
       "                elevatorCmd  rudderCmd    lonCmd  XaccBias  YaccBias  \\\n",
       "gpsPdop           -0.191682   0.071393  0.018013  0.006322 -0.050946   \n",
       "gpsNumSv           0.862620  -0.499020 -0.116267  0.025268  0.257591   \n",
       "gpsSec            -0.090324   0.155759  0.048774 -0.140194 -0.167229   \n",
       "P_Sol_Status      -0.506260   0.212534  0.044415  0.126167 -0.130686   \n",
       "AvionicsTemp      -0.902229   0.456009  0.101640  0.041917 -0.298337   \n",
       "Lat               -0.921223   0.531221  0.114726 -0.010135 -0.256051   \n",
       "Lon               -0.954003   0.543721  0.110886 -0.008930 -0.260223   \n",
       "H                  0.727669  -0.460461 -0.010743  0.009332  0.211839   \n",
       "tas                0.865022  -0.572597  0.028993  0.033232  0.271552   \n",
       "Eu[0]             -0.838015   0.747374  0.034341  0.019134 -0.271707   \n",
       "Eu[1]              0.983930  -0.507337 -0.100711 -0.036434  0.256058   \n",
       "Eu[2]             -0.001914  -0.014049  0.022056  0.084570 -0.020556   \n",
       "acc[0]             0.968641  -0.505431 -0.100408 -0.040527  0.254352   \n",
       "acc[1]             0.650110  -0.320327 -0.071030 -0.042423  0.160317   \n",
       "acc[2]            -0.553422   0.320730  0.040082  0.047136 -0.158328   \n",
       "pqr[0]             0.416900  -0.134908 -0.033596 -0.062480  0.092983   \n",
       "pqr[1]             0.429816  -0.384863 -0.012337 -0.034795  0.153046   \n",
       "pqr[2]             0.022819   0.275730 -0.042655 -0.069243 -0.062777   \n",
       "InertVNED[0]      -0.101043   0.144381  0.001304  0.047118 -0.072044   \n",
       "InertVNED[1]       0.024111  -0.161352  0.007612 -0.012743  0.063161   \n",
       "InertVNED[2]      -0.304770   0.087234  0.033822  0.118643 -0.023252   \n",
       "DynPressure        0.891579  -0.576338  0.050645  0.019561  0.276626   \n",
       "XMagFiled         -0.231348   0.192570 -0.007971  0.072167 -0.103479   \n",
       "YMagFiled         -0.544057   0.545680 -0.015397 -0.069716 -0.169362   \n",
       "ZMagFiled         -0.322083   0.369452  0.042962 -0.034674 -0.149854   \n",
       "Compass            0.061503  -0.043060  0.010892  0.096169 -0.004775   \n",
       "aileronLeftCmd     0.908012  -0.449528 -0.057363 -0.062840  0.233826   \n",
       "elevatorCmd        1.000000  -0.483568 -0.135234 -0.041515  0.250049   \n",
       "rudderCmd         -0.483568   1.000000 -0.199211 -0.055551 -0.231455   \n",
       "lonCmd            -0.135234  -0.199211  1.000000 -0.030313  0.049329   \n",
       "XaccBias          -0.041515  -0.055551 -0.030313  1.000000 -0.408730   \n",
       "YaccBias           0.250049  -0.231455  0.049329 -0.408730  1.000000   \n",
       "ZaccBias          -0.235980   0.200277  0.037793 -0.552741  0.054403   \n",
       "AirHeight          0.889678  -0.580402 -0.022211  0.006759  0.278911   \n",
       "WWest              0.510373  -0.267558 -0.052881  0.000402  0.292403   \n",
       "WSouth            -0.538591   0.290256  0.028370  0.146351 -0.199916   \n",
       "Jammer            -0.808793   0.463255  0.095710  0.099195 -0.283392   \n",
       "\n",
       "                ZaccBias  AirHeight     WWest    WSouth    Jammer  \n",
       "gpsPdop         0.042359  -0.143207 -0.066625  0.044012  0.162625  \n",
       "gpsNumSv       -0.339790   0.943205  0.563636 -0.480048 -0.846080  \n",
       "gpsSec          0.421922  -0.384407 -0.430665  0.076728  0.312062  \n",
       "P_Sol_Status   -0.171715  -0.319387 -0.175525  0.340965  0.458052  \n",
       "AvionicsTemp    0.183141  -0.762237 -0.504123  0.483972  0.726737  \n",
       "Lat             0.333126  -0.963945 -0.559601  0.508046  0.853407  \n",
       "Lon             0.336020  -0.971546 -0.555436  0.527449  0.856196  \n",
       "H              -0.275919   0.757811  0.399980 -0.435937 -0.658316  \n",
       "tas            -0.342690   0.982895  0.550253 -0.510074 -0.817636  \n",
       "Eu[0]           0.155659  -0.786663 -0.380390  0.463749  0.696942  \n",
       "Eu[1]          -0.257257   0.912889  0.534357 -0.544507 -0.822157  \n",
       "Eu[2]           0.133792   0.042956 -0.229469  0.089768 -0.029657  \n",
       "acc[0]         -0.248915   0.897502  0.519453 -0.544480 -0.806853  \n",
       "acc[1]         -0.060052   0.483672  0.300722 -0.378389 -0.458647  \n",
       "acc[2]          0.071159  -0.459004 -0.249704  0.322840  0.428079  \n",
       "pqr[0]          0.067977   0.210589  0.123588 -0.287966 -0.247045  \n",
       "pqr[1]         -0.029754   0.363021  0.201089 -0.281464 -0.346322  \n",
       "pqr[2]          0.186169  -0.240843 -0.048606 -0.047962  0.129898  \n",
       "InertVNED[0]   -0.042087  -0.091015  0.042908  0.135644  0.117072  \n",
       "InertVNED[1]   -0.059165   0.063507  0.067592 -0.077510 -0.067917  \n",
       "InertVNED[2]    0.073167  -0.254739 -0.138344  0.142929  0.241911  \n",
       "DynPressure    -0.331147   0.971854  0.547073 -0.513894 -0.818833  \n",
       "XMagFiled       0.102863  -0.179030 -0.118307  0.104257  0.161257  \n",
       "YMagFiled       0.065594  -0.607403 -0.116913  0.156086  0.511110  \n",
       "ZMagFiled       0.253668  -0.506638 -0.103305  0.274983  0.429471  \n",
       "Compass         0.105515   0.112166 -0.109873  0.213558 -0.083610  \n",
       "aileronLeftCmd -0.123520   0.788632  0.393943 -0.476577 -0.716213  \n",
       "elevatorCmd    -0.235980   0.889678  0.510373 -0.538591 -0.808793  \n",
       "rudderCmd       0.200277  -0.580402 -0.267558  0.290256  0.463255  \n",
       "lonCmd          0.037793  -0.022211 -0.052881  0.028370  0.095710  \n",
       "XaccBias       -0.552741   0.006759  0.000402  0.146351  0.099195  \n",
       "YaccBias        0.054403   0.278911  0.292403 -0.199916 -0.283392  \n",
       "ZaccBias        1.000000  -0.340498 -0.365933  0.044133  0.218115  \n",
       "AirHeight      -0.340498   1.000000  0.537606 -0.503144 -0.843884  \n",
       "WWest          -0.365933   0.537606  1.000000 -0.510738 -0.539402  \n",
       "WSouth          0.044133  -0.503144 -0.510738  1.000000  0.559624  \n",
       "Jammer          0.218115  -0.843884 -0.539402  0.559624  1.000000  \n",
       "\n",
       "[37 rows x 37 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corrdf=data.corr()\n",
    "corrdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "78ecc49b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Jammer            1.000000\n",
       "Lon               0.856196\n",
       "Lat               0.853407\n",
       "AvionicsTemp      0.726737\n",
       "Eu[0]             0.696942\n",
       "WSouth            0.559624\n",
       "YMagFiled         0.511110\n",
       "rudderCmd         0.463255\n",
       "P_Sol_Status      0.458052\n",
       "ZMagFiled         0.429471\n",
       "acc[2]            0.428079\n",
       "gpsSec            0.312062\n",
       "InertVNED[2]      0.241911\n",
       "ZaccBias          0.218115\n",
       "gpsPdop           0.162625\n",
       "XMagFiled         0.161257\n",
       "pqr[2]            0.129898\n",
       "InertVNED[0]      0.117072\n",
       "XaccBias          0.099195\n",
       "lonCmd            0.095710\n",
       "Eu[2]            -0.029657\n",
       "InertVNED[1]     -0.067917\n",
       "Compass          -0.083610\n",
       "pqr[0]           -0.247045\n",
       "YaccBias         -0.283392\n",
       "pqr[1]           -0.346322\n",
       "acc[1]           -0.458647\n",
       "WWest            -0.539402\n",
       "H                -0.658316\n",
       "aileronLeftCmd   -0.716213\n",
       "acc[0]           -0.806853\n",
       "elevatorCmd      -0.808793\n",
       "tas              -0.817636\n",
       "DynPressure      -0.818833\n",
       "Eu[1]            -0.822157\n",
       "AirHeight        -0.843884\n",
       "gpsNumSv         -0.846080\n",
       "Name: Jammer, dtype: float64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corrdf['Jammer'].sort_values(ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "5413d290",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "time_distributed (TimeDistri (None, 1, 1, 64)          1024      \n",
      "_________________________________________________________________\n",
      "time_distributed_1 (TimeDist (None, 1, 1, 64)          0         \n",
      "_________________________________________________________________\n",
      "time_distributed_2 (TimeDist (None, 1, 64)             0         \n",
      "_________________________________________________________________\n",
      "lstm (LSTM)                  (None, 1, 50)             23000     \n",
      "_________________________________________________________________\n",
      "lstm_1 (LSTM)                (None, 1, 50)             20200     \n",
      "_________________________________________________________________\n",
      "lstm_2 (LSTM)                (None, 1, 50)             20200     \n",
      "_________________________________________________________________\n",
      "lstm_3 (LSTM)                (None, 1, 50)             20200     \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1, 64)             3264      \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1, 16)             1040      \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1, 8)              136       \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 1, 2)              18        \n",
      "=================================================================\n",
      "Total params: 89,082\n",
      "Trainable params: 89,082\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "63def1c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from keras.utils.vis_utils import plot_model\n",
    "os.environ[\"PATH\"] += os.pathsep +'C:\\Program Files\\Graphviz/bin'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "4754b83b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_model(model, to_file='lstm.png',show_shapes='True')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a705074a",
   "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": "3.6.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}