prophet-demo/gdp2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "218af1ec",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "import pandas  as  pd\n",
    "import matplotlib.pyplot as plt\n",
    "import  json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "33e62663",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签SimHei\n",
    "plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "df=pd.read_excel('./input/gdphb.xlsx')\n",
    "\n",
    "display(df)\n",
    "# df=df[df[\"地区\"].str.strip()==\"安顺市\"]\n",
    "def minmax_norm(df_input):\n",
    "    return (df_input - df_input.min()) / (df_input.max() - df_input.min()) * 40 + 60\n",
    "global_df_arr=[]; #用来装需要预测的df数组\n",
    "global_corr_obj= {};#地区各指标相关数组\n",
    "for  s  in   df[\"地区\"].unique():\n",
    "    #print(s)\n",
    "    new_df=df[df[\"地区\"].str.strip()==s ]\n",
    "    # print(new_df.columns)\n",
    "    norm_df_input=new_df.drop(columns=['地区', 'ds']);\n",
    "    # print(norm_df_input.dtypes)\n",
    "    norm_df=  minmax_norm(norm_df_input);\n",
    "\n",
    "    corr= norm_df.corr()\n",
    "    # print(s+\"地区指标相关性\")\n",
    "    # display(corr)\n",
    "    result = corr.to_json(orient=\"index\")\n",
    "        # print(result)\n",
    "    parsed_corr = json.loads(result.replace(\":null\",\":\\\"NaN\\\"\"))\n",
    "    global_corr_obj[s]=parsed_corr;\n",
    "\n",
    "    # continue\n",
    "    for s_index  in  new_df.columns:\n",
    "        if s_index not in ['地区', 'ds', 'y'] :\n",
    "        #if sindex not in ['地区', 'ds', 'y'] and  \"增速\" not  in  sindex:\n",
    "            sdf=new_df.loc[:,['地区', 'ds']]\n",
    "            sdf['y']=df[s_index]\n",
    "            sdf['title-text']=s_index\n",
    "            sdf= sdf.reset_index(drop=True)\n",
    "            sdf=sdf.dropna()\n",
    "           #print(sdf)\n",
    "            global_df_arr.append(sdf)\n",
    "# print(df_arr)\n",
    "#global_df_arr\n",
    "\n",
    "# display(global_corr_obj)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "# global_df_arr"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "# print(\"各指标的相关性\")\n",
    "# df.corr()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "# m = Prophet(changepoints=['2020-12-01'])  #2020是个例外\n",
    "from prophet import Prophet\n",
    "from prophet.plot import plot_plotly, plot_components_plotly\n",
    "def show_prophet(dfs):\n",
    "    json_arr=[];\n",
    "    for mdf in  dfs:\n",
    "        # m = Prophet(changepoints=['2020-12-01'])\n",
    "        m = Prophet(changepoints=['2020-12-01'])\n",
    "        m.fit(mdf)\n",
    "        future = m.make_future_dataframe(periods=2,freq=\"Y\")\n",
    "        # print(future,m)\n",
    "        forecast = m.predict(future)\n",
    "        # forecast['y']=mdf.loc[:,['y']]\n",
    "        forecast= pd.concat([forecast,mdf.loc[:,'y']],axis=1)\n",
    "        # print(forecast.to_json(orient ='index'))\n",
    "        title= mdf[\"title-text\"][mdf.index[0]]\n",
    "        地区=mdf[\"地区\"][mdf.index[0]]\n",
    "        print(地区+title+\"回归分析\")\n",
    "        fig=plot_plotly(m, forecast,xlabel=\"时间\",ylabel=title,)\n",
    "        fig.show()\n",
    "        result = forecast.to_json(orient=\"index\")\n",
    "        # print(result)\n",
    "        parsed = json.loads(result.replace(\":null\",\":\\\"NaN\\\"\"))\n",
    "        #data_obj=json.load(forecast.to_json(orient ='index'))\n",
    "\n",
    "        json_obj={\"axis\":地区,\"info\":title,\"data\": parsed }\n",
    "        json_arr.append(json_obj)\n",
    "        # print(json_obj )\n",
    "        # print(title+\"趋势分析图\")\n",
    "        # fig.show()\n",
    "\n",
    "    return  json_arr\n",
    "res=show_prophet(global_df_arr[:])\n",
    "# print(res)\n"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "print(res)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ca17354e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fdf21f14-a2e5-4ee6-94fd-10dda9900f42",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "c09ee517baf2ef8cdccf4daa50b9eecbe95ad60e1eec620e80956ae2b91f8870"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}