cbpe_EDA_MNRAS.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os, json, sys\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import dask\n",
    "import dask.dataframe as dd\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\", category=DeprecationWarning)\n",
    "pd.options.mode.chained_assignment = None "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the data\n",
    "path = ''\n",
    "df = dd.read_csv(path+'data.csv')\n",
    "\n",
    "# drop rows with missing values\n",
    "df=df.replace(-9999.0,np.nan).dropna(how = 'any')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "351089\n",
      "209266\n",
      "81536\n",
      "60287\n"
     ]
    }
   ],
   "source": [
    "# undersample\n",
    "df = df.loc[::10]\n",
    "print(len(df))\n",
    "print(len(df[df['class'] == 'GALAXY']))\n",
    "print(len(df[df['class'] == 'STAR']))\n",
    "print(len(df[df['class'] == 'QSO']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['objid', 'ra', 'dec', 'Redshift', 'fiberMag_u', 'petroMag_u',\n",
       "       'modelMag_u', 'psfMag_u', 'fiberMag_g', 'petroMag_g', 'modelMag_g',\n",
       "       'psfMag_g', 'fiberMag_r', 'petroMag_r', 'modelMag_r', 'psfMag_r',\n",
       "       'fiberMag_i', 'petroMag_i', 'modelMag_i', 'psfMag_i', 'fiberMag_z',\n",
       "       'petroMag_z', 'modelMag_z', 'psfMag_z', 'class'], dtype=object)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['objid' 'ra' 'dec' 'Redshift' 'fiberMag_u' 'petroMag_u' 'modelMag_u'\n",
      " 'psfMag_u' 'fiberMag_g' 'petroMag_g' 'modelMag_g' 'psfMag_g' 'fiberMag_r'\n",
      " 'petroMag_r' 'modelMag_r' 'psfMag_r' 'fiberMag_i' 'petroMag_i'\n",
      " 'modelMag_i' 'psfMag_i' 'fiberMag_z' 'petroMag_z' 'modelMag_z' 'psfMag_z'\n",
      " 'class' 'modelMag_r-fiberMag_r' 'modelMag_u-modelMag_g'\n",
      " 'modelMag_u-modelMag_r' 'modelMag_u-modelMag_i' 'modelMag_u-modelMag_z'\n",
      " 'modelMag_g-modelMag_r' 'modelMag_g-modelMag_i' 'modelMag_g-modelMag_z'\n",
      " 'modelMag_r-modelMag_i' 'modelMag_r-modelMag_z' 'modelMag_i-modelMag_z']\n"
     ]
    }
   ],
   "source": [
    "# some rudimentary feature engineering\n",
    "# create a feature that is sensitive to the spatial size of a source:\n",
    "df['modelMag_r-fiberMag_r'] = df['modelMag_r']-df['fiberMag_r']\n",
    "\n",
    "# create colours:\n",
    "bands = ['u','g','r','i','z']\n",
    "for i,b1 in enumerate(bands[:-1]):\n",
    "    for b2 in bands[i+1:]:\n",
    "        df['modelMag_'+b1+'-modelMag_'+b2] = df['modelMag_'+b1] - df['modelMag_'+b2]\n",
    "\n",
    "print(df.columns.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         GALAXY\n",
       "10        GALAXY\n",
       "20        GALAXY\n",
       "30           QSO\n",
       "40           QSO\n",
       "           ...  \n",
       "161460    GALAXY\n",
       "161470       QSO\n",
       "161480       QSO\n",
       "161490      STAR\n",
       "161500    GALAXY\n",
       "Name: class, Length: 351089, dtype: object"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cols_ml = [x for x in df.columns.values.tolist() if 'Mag_' in x]\n",
    "df = df.compute()\n",
    "df['class']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>objid</th>\n",
       "      <th>ra</th>\n",
       "      <th>dec</th>\n",
       "      <th>Redshift</th>\n",
       "      <th>fiberMag_u</th>\n",
       "      <th>petroMag_u</th>\n",
       "      <th>modelMag_u</th>\n",
       "      <th>psfMag_u</th>\n",
       "      <th>fiberMag_g</th>\n",
       "      <th>petroMag_g</th>\n",
       "      <th>...</th>\n",
       "      <th>modelMag_u-modelMag_g</th>\n",
       "      <th>modelMag_u-modelMag_r</th>\n",
       "      <th>modelMag_u-modelMag_i</th>\n",
       "      <th>modelMag_u-modelMag_z</th>\n",
       "      <th>modelMag_g-modelMag_r</th>\n",
       "      <th>modelMag_g-modelMag_i</th>\n",
       "      <th>modelMag_g-modelMag_z</th>\n",
       "      <th>modelMag_r-modelMag_i</th>\n",
       "      <th>modelMag_r-modelMag_z</th>\n",
       "      <th>modelMag_i-modelMag_z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>106310</th>\n",
       "      <td>1237679504318922907</td>\n",
       "      <td>0.001661</td>\n",
       "      <td>20.397911</td>\n",
       "      <td>0.169639</td>\n",
       "      <td>23.19317</td>\n",
       "      <td>24.06787</td>\n",
       "      <td>23.09053</td>\n",
       "      <td>22.82628</td>\n",
       "      <td>22.08305</td>\n",
       "      <td>22.35545</td>\n",
       "      <td>...</td>\n",
       "      <td>0.93142</td>\n",
       "      <td>1.18984</td>\n",
       "      <td>1.11208</td>\n",
       "      <td>1.57977</td>\n",
       "      <td>0.25842</td>\n",
       "      <td>0.18066</td>\n",
       "      <td>0.64835</td>\n",
       "      <td>-0.07776</td>\n",
       "      <td>0.38993</td>\n",
       "      <td>0.46769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85930</th>\n",
       "      <td>1237679477470265897</td>\n",
       "      <td>0.001699</td>\n",
       "      <td>22.313928</td>\n",
       "      <td>-0.000842</td>\n",
       "      <td>22.40629</td>\n",
       "      <td>21.80065</td>\n",
       "      <td>21.93820</td>\n",
       "      <td>21.91749</td>\n",
       "      <td>21.26552</td>\n",
       "      <td>21.00201</td>\n",
       "      <td>...</td>\n",
       "      <td>1.03176</td>\n",
       "      <td>1.42015</td>\n",
       "      <td>1.58159</td>\n",
       "      <td>1.58350</td>\n",
       "      <td>0.38839</td>\n",
       "      <td>0.54983</td>\n",
       "      <td>0.55174</td>\n",
       "      <td>0.16144</td>\n",
       "      <td>0.16335</td>\n",
       "      <td>0.00191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88680</th>\n",
       "      <td>1237679478007202593</td>\n",
       "      <td>0.003054</td>\n",
       "      <td>22.654332</td>\n",
       "      <td>0.687130</td>\n",
       "      <td>23.72759</td>\n",
       "      <td>20.56767</td>\n",
       "      <td>22.96340</td>\n",
       "      <td>23.26116</td>\n",
       "      <td>23.81548</td>\n",
       "      <td>22.47039</td>\n",
       "      <td>...</td>\n",
       "      <td>0.25837</td>\n",
       "      <td>2.29204</td>\n",
       "      <td>3.48837</td>\n",
       "      <td>4.19817</td>\n",
       "      <td>2.03367</td>\n",
       "      <td>3.23000</td>\n",
       "      <td>3.93980</td>\n",
       "      <td>1.19633</td>\n",
       "      <td>1.90613</td>\n",
       "      <td>0.70980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122220</th>\n",
       "      <td>1237656495113633949</td>\n",
       "      <td>0.004698</td>\n",
       "      <td>14.666738</td>\n",
       "      <td>0.494372</td>\n",
       "      <td>24.64123</td>\n",
       "      <td>27.72672</td>\n",
       "      <td>24.16196</td>\n",
       "      <td>24.22452</td>\n",
       "      <td>22.79229</td>\n",
       "      <td>21.48647</td>\n",
       "      <td>...</td>\n",
       "      <td>2.14622</td>\n",
       "      <td>3.96156</td>\n",
       "      <td>4.89292</td>\n",
       "      <td>5.30700</td>\n",
       "      <td>1.81534</td>\n",
       "      <td>2.74670</td>\n",
       "      <td>3.16078</td>\n",
       "      <td>0.93136</td>\n",
       "      <td>1.34544</td>\n",
       "      <td>0.41408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178342</th>\n",
       "      <td>1237657191978959047</td>\n",
       "      <td>0.004742</td>\n",
       "      <td>0.708960</td>\n",
       "      <td>0.652078</td>\n",
       "      <td>22.79361</td>\n",
       "      <td>22.60482</td>\n",
       "      <td>20.88348</td>\n",
       "      <td>22.85150</td>\n",
       "      <td>22.60140</td>\n",
       "      <td>20.66000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.56076</td>\n",
       "      <td>1.28619</td>\n",
       "      <td>2.00339</td>\n",
       "      <td>1.88960</td>\n",
       "      <td>0.72543</td>\n",
       "      <td>1.44263</td>\n",
       "      <td>1.32884</td>\n",
       "      <td>0.71720</td>\n",
       "      <td>0.60341</td>\n",
       "      <td>-0.11379</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>206160</th>\n",
       "      <td>1237680297281127272</td>\n",
       "      <td>359.996771</td>\n",
       "      <td>21.785129</td>\n",
       "      <td>0.403197</td>\n",
       "      <td>25.55326</td>\n",
       "      <td>23.06392</td>\n",
       "      <td>26.63804</td>\n",
       "      <td>26.12176</td>\n",
       "      <td>22.29978</td>\n",
       "      <td>21.04331</td>\n",
       "      <td>...</td>\n",
       "      <td>5.63014</td>\n",
       "      <td>7.61149</td>\n",
       "      <td>8.32248</td>\n",
       "      <td>8.74460</td>\n",
       "      <td>1.98135</td>\n",
       "      <td>2.69234</td>\n",
       "      <td>3.11446</td>\n",
       "      <td>0.71099</td>\n",
       "      <td>1.13311</td>\n",
       "      <td>0.42212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245641</th>\n",
       "      <td>1237663306917741344</td>\n",
       "      <td>359.997392</td>\n",
       "      <td>29.784564</td>\n",
       "      <td>2.181215</td>\n",
       "      <td>22.86479</td>\n",
       "      <td>22.65849</td>\n",
       "      <td>22.53520</td>\n",
       "      <td>22.41355</td>\n",
       "      <td>21.84419</td>\n",
       "      <td>21.55383</td>\n",
       "      <td>...</td>\n",
       "      <td>1.11071</td>\n",
       "      <td>1.41677</td>\n",
       "      <td>1.49492</td>\n",
       "      <td>2.14227</td>\n",
       "      <td>0.30606</td>\n",
       "      <td>0.38421</td>\n",
       "      <td>1.03156</td>\n",
       "      <td>0.07815</td>\n",
       "      <td>0.72550</td>\n",
       "      <td>0.64735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86230</th>\n",
       "      <td>1237663784197357819</td>\n",
       "      <td>359.997663</td>\n",
       "      <td>0.144871</td>\n",
       "      <td>1.355421</td>\n",
       "      <td>20.82982</td>\n",
       "      <td>20.60065</td>\n",
       "      <td>20.50475</td>\n",
       "      <td>20.45739</td>\n",
       "      <td>20.79091</td>\n",
       "      <td>20.53729</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.01320</td>\n",
       "      <td>0.21841</td>\n",
       "      <td>0.22166</td>\n",
       "      <td>0.16939</td>\n",
       "      <td>0.23161</td>\n",
       "      <td>0.23486</td>\n",
       "      <td>0.18259</td>\n",
       "      <td>0.00325</td>\n",
       "      <td>-0.04902</td>\n",
       "      <td>-0.05227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>238840</th>\n",
       "      <td>1237669768154054658</td>\n",
       "      <td>359.997837</td>\n",
       "      <td>8.470280</td>\n",
       "      <td>2.683349</td>\n",
       "      <td>18.47355</td>\n",
       "      <td>18.20370</td>\n",
       "      <td>18.16550</td>\n",
       "      <td>18.17316</td>\n",
       "      <td>18.54374</td>\n",
       "      <td>18.27373</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.06948</td>\n",
       "      <td>-0.36008</td>\n",
       "      <td>-0.60637</td>\n",
       "      <td>-0.78399</td>\n",
       "      <td>-0.29060</td>\n",
       "      <td>-0.53689</td>\n",
       "      <td>-0.71451</td>\n",
       "      <td>-0.24629</td>\n",
       "      <td>-0.42391</td>\n",
       "      <td>-0.17762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>226751</th>\n",
       "      <td>1237663235523477803</td>\n",
       "      <td>359.999954</td>\n",
       "      <td>34.747133</td>\n",
       "      <td>2.397724</td>\n",
       "      <td>21.85833</td>\n",
       "      <td>21.53611</td>\n",
       "      <td>21.66895</td>\n",
       "      <td>21.70696</td>\n",
       "      <td>21.22537</td>\n",
       "      <td>20.99523</td>\n",
       "      <td>...</td>\n",
       "      <td>0.77329</td>\n",
       "      <td>1.09625</td>\n",
       "      <td>1.44160</td>\n",
       "      <td>1.77464</td>\n",
       "      <td>0.32296</td>\n",
       "      <td>0.66831</td>\n",
       "      <td>1.00135</td>\n",
       "      <td>0.34535</td>\n",
       "      <td>0.67839</td>\n",
       "      <td>0.33304</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>351089 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      objid          ra        dec  Redshift  fiberMag_u  \\\n",
       "106310  1237679504318922907    0.001661  20.397911  0.169639    23.19317   \n",
       "85930   1237679477470265897    0.001699  22.313928 -0.000842    22.40629   \n",
       "88680   1237679478007202593    0.003054  22.654332  0.687130    23.72759   \n",
       "122220  1237656495113633949    0.004698  14.666738  0.494372    24.64123   \n",
       "178342  1237657191978959047    0.004742   0.708960  0.652078    22.79361   \n",
       "...                     ...         ...        ...       ...         ...   \n",
       "206160  1237680297281127272  359.996771  21.785129  0.403197    25.55326   \n",
       "245641  1237663306917741344  359.997392  29.784564  2.181215    22.86479   \n",
       "86230   1237663784197357819  359.997663   0.144871  1.355421    20.82982   \n",
       "238840  1237669768154054658  359.997837   8.470280  2.683349    18.47355   \n",
       "226751  1237663235523477803  359.999954  34.747133  2.397724    21.85833   \n",
       "\n",
       "        petroMag_u  modelMag_u  psfMag_u  fiberMag_g  petroMag_g  ...  \\\n",
       "106310    24.06787    23.09053  22.82628    22.08305    22.35545  ...   \n",
       "85930     21.80065    21.93820  21.91749    21.26552    21.00201  ...   \n",
       "88680     20.56767    22.96340  23.26116    23.81548    22.47039  ...   \n",
       "122220    27.72672    24.16196  24.22452    22.79229    21.48647  ...   \n",
       "178342    22.60482    20.88348  22.85150    22.60140    20.66000  ...   \n",
       "...            ...         ...       ...         ...         ...  ...   \n",
       "206160    23.06392    26.63804  26.12176    22.29978    21.04331  ...   \n",
       "245641    22.65849    22.53520  22.41355    21.84419    21.55383  ...   \n",
       "86230     20.60065    20.50475  20.45739    20.79091    20.53729  ...   \n",
       "238840    18.20370    18.16550  18.17316    18.54374    18.27373  ...   \n",
       "226751    21.53611    21.66895  21.70696    21.22537    20.99523  ...   \n",
       "\n",
       "        modelMag_u-modelMag_g  modelMag_u-modelMag_r  modelMag_u-modelMag_i  \\\n",
       "106310                0.93142                1.18984                1.11208   \n",
       "85930                 1.03176                1.42015                1.58159   \n",
       "88680                 0.25837                2.29204                3.48837   \n",
       "122220                2.14622                3.96156                4.89292   \n",
       "178342                0.56076                1.28619                2.00339   \n",
       "...                       ...                    ...                    ...   \n",
       "206160                5.63014                7.61149                8.32248   \n",
       "245641                1.11071                1.41677                1.49492   \n",
       "86230                -0.01320                0.21841                0.22166   \n",
       "238840               -0.06948               -0.36008               -0.60637   \n",
       "226751                0.77329                1.09625                1.44160   \n",
       "\n",
       "        modelMag_u-modelMag_z  modelMag_g-modelMag_r  modelMag_g-modelMag_i  \\\n",
       "106310                1.57977                0.25842                0.18066   \n",
       "85930                 1.58350                0.38839                0.54983   \n",
       "88680                 4.19817                2.03367                3.23000   \n",
       "122220                5.30700                1.81534                2.74670   \n",
       "178342                1.88960                0.72543                1.44263   \n",
       "...                       ...                    ...                    ...   \n",
       "206160                8.74460                1.98135                2.69234   \n",
       "245641                2.14227                0.30606                0.38421   \n",
       "86230                 0.16939                0.23161                0.23486   \n",
       "238840               -0.78399               -0.29060               -0.53689   \n",
       "226751                1.77464                0.32296                0.66831   \n",
       "\n",
       "        modelMag_g-modelMag_z  modelMag_r-modelMag_i  modelMag_r-modelMag_z  \\\n",
       "106310                0.64835               -0.07776                0.38993   \n",
       "85930                 0.55174                0.16144                0.16335   \n",
       "88680                 3.93980                1.19633                1.90613   \n",
       "122220                3.16078                0.93136                1.34544   \n",
       "178342                1.32884                0.71720                0.60341   \n",
       "...                       ...                    ...                    ...   \n",
       "206160                3.11446                0.71099                1.13311   \n",
       "245641                1.03156                0.07815                0.72550   \n",
       "86230                 0.18259                0.00325               -0.04902   \n",
       "238840               -0.71451               -0.24629               -0.42391   \n",
       "226751                1.00135                0.34535                0.67839   \n",
       "\n",
       "        modelMag_i-modelMag_z  \n",
       "106310                0.46769  \n",
       "85930                 0.00191  \n",
       "88680                 0.70980  \n",
       "122220                0.41408  \n",
       "178342               -0.11379  \n",
       "...                       ...  \n",
       "206160                0.42212  \n",
       "245641                0.64735  \n",
       "86230                -0.05227  \n",
       "238840               -0.17762  \n",
       "226751                0.33304  \n",
       "\n",
       "[351089 rows x 36 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=df.sort_values(by=['ra'])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['isStar'] = np.where(df['class']=='STAR',1,0)\n",
    "df['frac_star'] = df['isStar'].rolling(1000).mean()\n",
    "\n",
    "df['isQSO'] = np.where(df['class']=='QSO',1,0)\n",
    "df['frac_qso'] = df['isQSO'].rolling(1000).mean()\n",
    "\n",
    "df['isGalaxy'] = np.where(df['class']=='GALAXY',1,0)\n",
    "df['frac_galaxy'] = df['isGalaxy'].rolling(1000).mean()\n",
    "\n",
    "plt.plot(df.ra,df.frac_star,label='Stars')\n",
    "plt.plot(df.ra,df.frac_galaxy,label='Galaxies')\n",
    "plt.plot(df.ra,df.frac_qso,label='QSOs')\n",
    "plt.xlabel('RA [degree]')\n",
    "plt.ylabel('Class fraction')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "#plt.savefig(path+'class_fraction.pdf')\n",
    "#plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cycle = plt.rcParams['axes.prop_cycle'].by_key()['color']\n",
    "\n",
    "df['g-i_rolling'] = df['modelMag_g-modelMag_i'].rolling(1000).median()\n",
    "\n",
    "df_stars = df[df.isStar==1]\n",
    "df_stars['g-i_rolling'] = df_stars['modelMag_g-modelMag_i'].rolling(1000).median()\n",
    "df_stars['g-i_rolling_mean'] = df_stars['modelMag_g-modelMag_i'].rolling(1000).mean()\n",
    "df_stars['g-i_rolling_plus_1sigma'] = df_stars['g-i_rolling_mean']+df_stars['modelMag_g-modelMag_i'].rolling(1000).std()\n",
    "df_stars['g-i_rolling_minus_1sigma'] = df_stars['g-i_rolling_mean']-df_stars['modelMag_g-modelMag_i'].rolling(1000).std()\n",
    "\n",
    "\n",
    "df_gal = df[df.isGalaxy==1]\n",
    "df_gal['g-i_rolling'] = df_gal['modelMag_g-modelMag_i'].rolling(1000).median()\n",
    "df_gal['g-i_rolling_mean'] = df_gal['modelMag_g-modelMag_i'].rolling(1000).mean()\n",
    "df_gal['g-i_rolling_plus_1sigma'] = df_gal['g-i_rolling_mean']+df_gal['modelMag_g-modelMag_i'].rolling(1000).std()\n",
    "df_gal['g-i_rolling_minus_1sigma'] = df_gal['g-i_rolling_mean']-df_gal['modelMag_g-modelMag_i'].rolling(1000).std()\n",
    "\n",
    "\n",
    "df_qso = df[df.isQSO==1]\n",
    "df_qso['g-i_rolling'] = df_qso['modelMag_g-modelMag_i'].rolling(1000).median()\n",
    "df_qso['g-i_rolling_mean'] = df_qso['modelMag_g-modelMag_i'].rolling(1000).mean()\n",
    "df_qso['g-i_rolling_plus_1sigma'] = df_qso['g-i_rolling_mean']+df_qso['modelMag_g-modelMag_i'].rolling(1000).std()\n",
    "df_qso['g-i_rolling_minus_1sigma'] = df_qso['g-i_rolling_mean']-df_qso['modelMag_g-modelMag_i'].rolling(1000).std()\n",
    "\n",
    "\n",
    "plt.plot(df_stars.ra,df_stars['g-i_rolling'],label='Stars')\n",
    "plt.fill_between(df_stars.ra,\n",
    "                 df_stars['g-i_rolling_plus_1sigma'],\n",
    "                df_stars['g-i_rolling_minus_1sigma'],color=cycle[0],alpha=0.1)\n",
    "\n",
    "plt.plot(df_gal.ra,df_gal['g-i_rolling'],label='Galaxies')\n",
    "plt.fill_between(df_gal.ra,\n",
    "                 df_gal['g-i_rolling_plus_1sigma'],\n",
    "                df_gal['g-i_rolling_minus_1sigma'],color=cycle[1],alpha=0.1)\n",
    "\n",
    "plt.plot(df_qso.ra,df_qso['g-i_rolling'],label='QSOs')\n",
    "plt.fill_between(df_qso.ra,\n",
    "                 df_qso['g-i_rolling_plus_1sigma'],\n",
    "                df_qso['g-i_rolling_minus_1sigma'],color=cycle[2],alpha=0.1)\n",
    "\n",
    "\n",
    "plt.xlabel('RA [degree]')\n",
    "plt.ylabel('Median $g-i$ [mag]')\n",
    "\n",
    "plt.show()\n",
    "\n",
    "#plt.savefig(path2+'median_colours.pdf')\n",
    "#plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:intel]",
   "language": "python",
   "name": "conda-env-intel-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}