{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# 训练像素模型\n",
    "用这个文件可以训练出需要使用的光谱像素点模型"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pickle\n",
    "from utils import read_envi_ascii\n",
    "from config import Config\n",
    "from models import ManualTree"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "# 一些变量"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "outputs": [],
   "source": [
    "data_path = r'data/envi20220802.txt'\n",
    "name_dict = {'tobacco': 1, 'yantou':2, 'kazhi':3, 'bomo':4, 'jiaodai':5, 'background':0}"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "# 构建数据集"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "outputs": [],
   "source": [
    "data = read_envi_ascii(data_path)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "zibian (569, 448)\n",
      "tobacco (1457, 448)\n",
      "yantou (354, 448)\n",
      "kazhi (449, 448)\n",
      "bomo (1154, 448)\n",
      "jiaodai (566, 448)\n",
      "background (1235, 448)\n"
     ]
    }
   ],
   "source": [
    "_ = [print(class_name, d.shape) for class_name, d in data.items()]"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "outputs": [],
   "source": [
    "data_x = [d for class_name, d in data.items() if class_name in name_dict.keys()]\n",
    "data_y = [np.ones((d.shape[0], ))*name_dict[class_name] for class_name, d in data.items() if class_name in name_dict.keys()]\n",
    "data_x, data_y = np.concatenate(data_x), np.concatenate(data_y)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 取出需要的22个特征谱段"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "这些是现在的数据:  (5215, 448) (5215,)\n",
      "截取其中需要的部分后:  (5215, 22) (5215,)\n"
     ]
    }
   ],
   "source": [
    "print(\"这些是现在的数据: \", data_x.shape, data_y.shape)\n",
    "data_x_cut = data_x[..., Config.bands]\n",
    "print(\"截取其中需要的部分后: \", data_x_cut.shape, data_y.shape)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 进行样本平衡"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "这是重采样后的数据:  (8742, 22) (8742,)\n"
     ]
    }
   ],
   "source": [
    "from imblearn.over_sampling import RandomOverSampler\n",
    "ros = RandomOverSampler(random_state=0)\n",
    "x_resampled, y_resampled = ros.fit_resample(data_x_cut, data_y)\n",
    "print('这是重采样后的数据: ', x_resampled.shape, y_resampled.shape)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "#"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}