tutorials/DSC_regression-tutorial.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5b59c50b",
   "metadata": {},
   "source": [
    "# \"Machine Learning for Climate Predictions\" Workshop\n",
    "#### Jupyter Notebook prepared by Cindy Gonzales"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b39c959",
   "metadata": {},
   "source": [
    "Artificial intelligence (AI) methods have begun to show great promise in climate science applications. This notebooke will introduce and describe a dataset and methods to make machine learning predictions of a multiresolution climate model ensemble."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c09f6244",
   "metadata": {},
   "source": [
    "For more information on the dataset and methods used, please reference the paper:\n",
    "\n",
    "[Machine Learning Predictions of Multi-Resolution Climate Model Ensembles– Anderson, G. J. and Lucas, D. D. (GRL: in review). IM release number: LLNL-JRNL-744178.](https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018GL077049)\n",
    "\n",
    "**Auspices Statement**: This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "07d66384",
   "metadata": {},
   "source": [
    "#### Downloading the data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd98b0f0",
   "metadata": {},
   "source": [
    "Let's start by downloading the data locally. Uncomment this cell to run the following commands to download the dataset or follow the instructions in the paper to access the data.\n",
    "\n",
    "**PRO TIP**: To uncomment a full cell, use `Ctrl` + `A` to select all lines then `Ctrl` + `/` to quick uncomment all selected lines."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "2a6c471a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ! wget ftp://gdo148.ucllnl.org/multires-random-forest.tgz\n",
    "# ! tar -xvzf multires-random-forest.tgz"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5785e4b2",
   "metadata": {},
   "source": [
    "#### Loading the needed packages and data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1764c619",
   "metadata": {},
   "source": [
    "Now that we've downloaded the data, let's first import some needed packages that we can use such as numpy, pandas, matplotlib and useful functions from sklearn. We'll also need to read the data into a useable format."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3f84be4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.ensemble import (\n",
    "    ExtraTreesRegressor, GradientBoostingRegressor, RandomForestRegressor\n",
    ")\n",
    "from sklearn.neural_network import MLPRegressor\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import explained_variance_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a6dd1a83",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = np.load('multi-resolution/multiresolution_params_and_metrics.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "42994592",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(\n",
    "    features, \n",
    "     columns=['cldfrc_rhminl', 'micromg_dcs', 'eddydiff_a2l', 'uwshcu_criqc', 'uwshcu_rkm',\n",
    "              'uwshcu_rpen', 'zmconv_alfa', 'zmconv_c0_ocn', 'zmconv_dmpdz', 'zmconv_ke', \n",
    "              'zmconv_tau', 'resolution', 'TOA_energy_flux', 'precipitation']\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4e4fceaf",
   "metadata": {
    "scrolled": false
   },
   "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>cldfrc_rhminl</th>\n",
       "      <th>micromg_dcs</th>\n",
       "      <th>eddydiff_a2l</th>\n",
       "      <th>uwshcu_criqc</th>\n",
       "      <th>uwshcu_rkm</th>\n",
       "      <th>uwshcu_rpen</th>\n",
       "      <th>zmconv_alfa</th>\n",
       "      <th>zmconv_c0_ocn</th>\n",
       "      <th>zmconv_dmpdz</th>\n",
       "      <th>zmconv_ke</th>\n",
       "      <th>zmconv_tau</th>\n",
       "      <th>resolution</th>\n",
       "      <th>TOA_energy_flux</th>\n",
       "      <th>precipitation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.513158</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>11.078644</td>\n",
       "      <td>3.287566e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.900214</td>\n",
       "      <td>3.378267e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>18.531844</td>\n",
       "      <td>3.236089e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.513158</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>8.339281</td>\n",
       "      <td>3.410277e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.513158</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>11.162554</td>\n",
       "      <td>3.275170e-08</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>901</th>\n",
       "      <td>0.841311</td>\n",
       "      <td>0.550732</td>\n",
       "      <td>0.862211</td>\n",
       "      <td>0.771593</td>\n",
       "      <td>0.626219</td>\n",
       "      <td>0.712388</td>\n",
       "      <td>0.473620</td>\n",
       "      <td>0.937692</td>\n",
       "      <td>0.952484</td>\n",
       "      <td>0.795456</td>\n",
       "      <td>0.953037</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>3.686973</td>\n",
       "      <td>3.562486e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>902</th>\n",
       "      <td>0.420755</td>\n",
       "      <td>0.150635</td>\n",
       "      <td>0.879170</td>\n",
       "      <td>0.933833</td>\n",
       "      <td>0.374344</td>\n",
       "      <td>0.412366</td>\n",
       "      <td>0.253196</td>\n",
       "      <td>0.537597</td>\n",
       "      <td>0.920912</td>\n",
       "      <td>0.719474</td>\n",
       "      <td>0.022809</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>-6.642911</td>\n",
       "      <td>3.608198e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>903</th>\n",
       "      <td>0.016315</td>\n",
       "      <td>0.200030</td>\n",
       "      <td>0.430349</td>\n",
       "      <td>0.025959</td>\n",
       "      <td>0.347203</td>\n",
       "      <td>0.921252</td>\n",
       "      <td>0.008454</td>\n",
       "      <td>0.543158</td>\n",
       "      <td>0.114437</td>\n",
       "      <td>0.566006</td>\n",
       "      <td>0.151339</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>-5.854439</td>\n",
       "      <td>3.636534e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>904</th>\n",
       "      <td>0.522297</td>\n",
       "      <td>0.052848</td>\n",
       "      <td>0.662738</td>\n",
       "      <td>0.006282</td>\n",
       "      <td>0.277326</td>\n",
       "      <td>0.519419</td>\n",
       "      <td>0.704073</td>\n",
       "      <td>0.868487</td>\n",
       "      <td>0.069545</td>\n",
       "      <td>0.254931</td>\n",
       "      <td>0.286714</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>4.026710</td>\n",
       "      <td>3.458463e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>905</th>\n",
       "      <td>0.876733</td>\n",
       "      <td>0.351345</td>\n",
       "      <td>0.615204</td>\n",
       "      <td>0.457433</td>\n",
       "      <td>0.027828</td>\n",
       "      <td>0.303024</td>\n",
       "      <td>0.687878</td>\n",
       "      <td>0.609783</td>\n",
       "      <td>0.016450</td>\n",
       "      <td>0.072018</td>\n",
       "      <td>0.347606</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>2.388479</td>\n",
       "      <td>3.377284e-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>906 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     cldfrc_rhminl  micromg_dcs  eddydiff_a2l  uwshcu_criqc  uwshcu_rkm  \\\n",
       "0         0.513158     0.750000      0.500000      0.200000    0.750000   \n",
       "1         0.000000     0.750000      0.500000      0.200000    0.750000   \n",
       "2         1.000000     0.750000      0.500000      0.200000    0.750000   \n",
       "3         0.513158     0.000000      0.500000      0.200000    0.750000   \n",
       "4         0.513158     1.000000      0.500000      0.200000    0.750000   \n",
       "..             ...          ...           ...           ...         ...   \n",
       "901       0.841311     0.550732      0.862211      0.771593    0.626219   \n",
       "902       0.420755     0.150635      0.879170      0.933833    0.374344   \n",
       "903       0.016315     0.200030      0.430349      0.025959    0.347203   \n",
       "904       0.522297     0.052848      0.662738      0.006282    0.277326   \n",
       "905       0.876733     0.351345      0.615204      0.457433    0.027828   \n",
       "\n",
       "     uwshcu_rpen  zmconv_alfa  zmconv_c0_ocn  zmconv_dmpdz  zmconv_ke  \\\n",
       "0       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "1       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "2       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "3       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "4       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "..           ...          ...            ...           ...        ...   \n",
       "901     0.712388     0.473620       0.937692      0.952484   0.795456   \n",
       "902     0.412366     0.253196       0.537597      0.920912   0.719474   \n",
       "903     0.921252     0.008454       0.543158      0.114437   0.566006   \n",
       "904     0.519419     0.704073       0.868487      0.069545   0.254931   \n",
       "905     0.303024     0.687878       0.609783      0.016450   0.072018   \n",
       "\n",
       "     zmconv_tau  resolution  TOA_energy_flux  precipitation  \n",
       "0      0.544417      1.0000        11.078644   3.287566e-08  \n",
       "1      0.544417      1.0000         0.900214   3.378267e-08  \n",
       "2      0.544417      1.0000        18.531844   3.236089e-08  \n",
       "3      0.544417      1.0000         8.339281   3.410277e-08  \n",
       "4      0.544417      1.0000        11.162554   3.275170e-08  \n",
       "..          ...         ...              ...            ...  \n",
       "901    0.953037      0.0625         3.686973   3.562486e-08  \n",
       "902    0.022809      0.0625        -6.642911   3.608198e-08  \n",
       "903    0.151339      0.0625        -5.854439   3.636534e-08  \n",
       "904    0.286714      0.0625         4.026710   3.458463e-08  \n",
       "905    0.347606      0.0625         2.388479   3.377284e-08  \n",
       "\n",
       "[906 rows x 14 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4226e0d1",
   "metadata": {},
   "source": [
    "#### Looking at the data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6fd2ecd7",
   "metadata": {},
   "source": [
    "Overall, the data has 906 rows and 14 columns. The columns correspond with features and the rows correspond with data points collected. We will use a subset of the data points to train a predictive model. \n",
    "\n",
    "First, let's briefly discuss what features are included. We collected these features by running simulations in a climate model repeatedly. We have 11 features that have been identified as important that span five physical schemes related to clouds, cloud microphysics, turbulent mixing, and deep and shallow convection. Below is a table describing the features and their expert-provided ranges of variation:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76ab239c",
   "metadata": {},
   "source": [
    "Description of features:\n",
    "\n",
    "| Parameter | Scheme | Description | Range |\n",
    "| --------- | ------ | ----------- | ----- |\n",
    "| cldfrc_rhminl | Cloud processes | Humidity threshold for low clouds | \\[0.8, 0.99\\] |\n",
    "| micromg_dcs | Microphysics | Autoconversion size threshold (ice to snow) | \\[1.0E-4, 5.0E-4\\] |\n",
    "| eddydiff_a2l | Eddy diffusion | Moist entrainment enhancement | \\[10.0, 50.0\\] |\n",
    "| uwshcu_criqc| Shallow convection | Maximum updraft condensate |\\[0.0005, 0.0015\\] |\n",
    "| uwshcu_rkm | Shallow convection | Fractional updraft mixing efficiency | \\[8.0, 16.0\\] |\n",
    "| uwshcu_rpen | Shallow convection | Penetrative entrainment efficiency | \\[1.0, 10.0\\] |\n",
    "| zmconv_alfa | Deep convection | Initial cloud downdraft mass flux | \\[0.05, 0.6\\] |\n",
    "| zmconv_c0_ocn | Deep convection | Precipitation efficiency over ocean | \\[0.001, 0.1\\] |\n",
    "| zmconv_dmpdz | Deep convection | Mass entrainment rate | \\[0.0002, 0.002\\] |\n",
    "| zmconv_ke | Deep convection | Evaporation efficiency | \\[0.5E−6, 10.0E−6\\] |\n",
    "| zmconv_tau | Deep convection | Convective timescale | \\[300, 28800\\] |"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7f09a2cb",
   "metadata": {},
   "source": [
    "#### Quick EDA"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "304fe2d0",
   "metadata": {},
   "source": [
    "`.describe()` is a great function to start with as it shows us the relevant summary statistics. Really good for quantitative data, but can be used for categorical data also."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f44d6490",
   "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>cldfrc_rhminl</th>\n",
       "      <th>micromg_dcs</th>\n",
       "      <th>eddydiff_a2l</th>\n",
       "      <th>uwshcu_criqc</th>\n",
       "      <th>uwshcu_rkm</th>\n",
       "      <th>uwshcu_rpen</th>\n",
       "      <th>zmconv_alfa</th>\n",
       "      <th>zmconv_c0_ocn</th>\n",
       "      <th>zmconv_dmpdz</th>\n",
       "      <th>zmconv_ke</th>\n",
       "      <th>zmconv_tau</th>\n",
       "      <th>resolution</th>\n",
       "      <th>TOA_energy_flux</th>\n",
       "      <th>precipitation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>906.000000</td>\n",
       "      <td>9.060000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.498054</td>\n",
       "      <td>0.511377</td>\n",
       "      <td>0.501705</td>\n",
       "      <td>0.481428</td>\n",
       "      <td>0.510413</td>\n",
       "      <td>0.526425</td>\n",
       "      <td>0.483893</td>\n",
       "      <td>0.528215</td>\n",
       "      <td>0.511374</td>\n",
       "      <td>0.480589</td>\n",
       "      <td>0.504023</td>\n",
       "      <td>0.636175</td>\n",
       "      <td>-2.652540</td>\n",
       "      <td>3.345929e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.280627</td>\n",
       "      <td>0.288770</td>\n",
       "      <td>0.282951</td>\n",
       "      <td>0.293929</td>\n",
       "      <td>0.290655</td>\n",
       "      <td>0.307763</td>\n",
       "      <td>0.285501</td>\n",
       "      <td>0.290280</td>\n",
       "      <td>0.285480</td>\n",
       "      <td>0.285647</td>\n",
       "      <td>0.279392</td>\n",
       "      <td>0.410166</td>\n",
       "      <td>9.070179</td>\n",
       "      <td>1.309382e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.062500</td>\n",
       "      <td>-29.441491</td>\n",
       "      <td>2.997264e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.268623</td>\n",
       "      <td>0.261288</td>\n",
       "      <td>0.262730</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.256391</td>\n",
       "      <td>0.260556</td>\n",
       "      <td>0.264603</td>\n",
       "      <td>0.276082</td>\n",
       "      <td>0.266700</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.271315</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>-8.570273</td>\n",
       "      <td>3.255501e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.513158</td>\n",
       "      <td>0.531455</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.466926</td>\n",
       "      <td>0.519366</td>\n",
       "      <td>0.519419</td>\n",
       "      <td>0.463686</td>\n",
       "      <td>0.532881</td>\n",
       "      <td>0.528644</td>\n",
       "      <td>0.459701</td>\n",
       "      <td>0.535002</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-2.436454</td>\n",
       "      <td>3.337256e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.730864</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.740264</td>\n",
       "      <td>0.740867</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.797725</td>\n",
       "      <td>0.724125</td>\n",
       "      <td>0.815526</td>\n",
       "      <td>0.729065</td>\n",
       "      <td>0.726704</td>\n",
       "      <td>0.732510</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.109631</td>\n",
       "      <td>3.433868e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>19.044781</td>\n",
       "      <td>3.704008e-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cldfrc_rhminl  micromg_dcs  eddydiff_a2l  uwshcu_criqc  uwshcu_rkm  \\\n",
       "count     906.000000   906.000000    906.000000    906.000000  906.000000   \n",
       "mean        0.498054     0.511377      0.501705      0.481428    0.510413   \n",
       "std         0.280627     0.288770      0.282951      0.293929    0.290655   \n",
       "min         0.000000     0.000000      0.000000      0.000000    0.000000   \n",
       "25%         0.268623     0.261288      0.262730      0.200000    0.256391   \n",
       "50%         0.513158     0.531455      0.500000      0.466926    0.519366   \n",
       "75%         0.730864     0.750000      0.740264      0.740867    0.750000   \n",
       "max         1.000000     1.000000      1.000000      1.000000    1.000000   \n",
       "\n",
       "       uwshcu_rpen  zmconv_alfa  zmconv_c0_ocn  zmconv_dmpdz   zmconv_ke  \\\n",
       "count   906.000000   906.000000     906.000000    906.000000  906.000000   \n",
       "mean      0.526425     0.483893       0.528215      0.511374    0.480589   \n",
       "std       0.307763     0.285501       0.290280      0.285480    0.285647   \n",
       "min       0.000000     0.000000       0.000000      0.000000    0.000000   \n",
       "25%       0.260556     0.264603       0.276082      0.266700    0.231378   \n",
       "50%       0.519419     0.463686       0.532881      0.528644    0.459701   \n",
       "75%       0.797725     0.724125       0.815526      0.729065    0.726704   \n",
       "max       1.000000     1.000000       1.000000      1.000000    1.000000   \n",
       "\n",
       "       zmconv_tau  resolution  TOA_energy_flux  precipitation  \n",
       "count  906.000000  906.000000       906.000000   9.060000e+02  \n",
       "mean     0.504023    0.636175        -2.652540   3.345929e-08  \n",
       "std      0.279392    0.410166         9.070179   1.309382e-09  \n",
       "min      0.000000    0.062500       -29.441491   2.997264e-08  \n",
       "25%      0.271315    0.250000        -8.570273   3.255501e-08  \n",
       "50%      0.535002    1.000000        -2.436454   3.337256e-08  \n",
       "75%      0.732510    1.000000         4.109631   3.433868e-08  \n",
       "max      1.000000    1.000000        19.044781   3.704008e-08  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5f63e14",
   "metadata": {},
   "source": [
    "We can also refine by certain columns by specifying those as an array. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "af11547c",
   "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>TOA_energy_flux</th>\n",
       "      <th>precipitation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>906.000000</td>\n",
       "      <td>9.060000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-2.652540</td>\n",
       "      <td>3.345929e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.070179</td>\n",
       "      <td>1.309382e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-29.441491</td>\n",
       "      <td>2.997264e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-8.570273</td>\n",
       "      <td>3.255501e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-2.436454</td>\n",
       "      <td>3.337256e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4.109631</td>\n",
       "      <td>3.433868e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>19.044781</td>\n",
       "      <td>3.704008e-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       TOA_energy_flux  precipitation\n",
       "count       906.000000   9.060000e+02\n",
       "mean         -2.652540   3.345929e-08\n",
       "std           9.070179   1.309382e-09\n",
       "min         -29.441491   2.997264e-08\n",
       "25%          -8.570273   3.255501e-08\n",
       "50%          -2.436454   3.337256e-08\n",
       "75%           4.109631   3.433868e-08\n",
       "max          19.044781   3.704008e-08"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['TOA_energy_flux', 'precipitation']].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6279402",
   "metadata": {},
   "source": [
    "We can also group the dataframe by a specified variable and run some .agg to get a dataframe of specified calculations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f7620618",
   "metadata": {
    "scrolled": true
   },
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">cldfrc_rhminl</th>\n",
       "      <th colspan=\"2\" halign=\"left\">micromg_dcs</th>\n",
       "      <th colspan=\"2\" halign=\"left\">eddydiff_a2l</th>\n",
       "      <th colspan=\"2\" halign=\"left\">uwshcu_criqc</th>\n",
       "      <th colspan=\"2\" halign=\"left\">uwshcu_rkm</th>\n",
       "      <th>...</th>\n",
       "      <th colspan=\"2\" halign=\"left\">zmconv_dmpdz</th>\n",
       "      <th colspan=\"2\" halign=\"left\">zmconv_ke</th>\n",
       "      <th colspan=\"2\" halign=\"left\">zmconv_tau</th>\n",
       "      <th colspan=\"2\" halign=\"left\">TOA_energy_flux</th>\n",
       "      <th colspan=\"2\" halign=\"left\">precipitation</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>...</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resolution</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0625</th>\n",
       "      <td>0.501189</td>\n",
       "      <td>0.270830</td>\n",
       "      <td>0.519213</td>\n",
       "      <td>0.289553</td>\n",
       "      <td>0.512733</td>\n",
       "      <td>0.278059</td>\n",
       "      <td>0.453806</td>\n",
       "      <td>0.301031</td>\n",
       "      <td>0.509305</td>\n",
       "      <td>0.292473</td>\n",
       "      <td>...</td>\n",
       "      <td>0.531118</td>\n",
       "      <td>0.286086</td>\n",
       "      <td>0.471942</td>\n",
       "      <td>0.280746</td>\n",
       "      <td>0.508458</td>\n",
       "      <td>0.271686</td>\n",
       "      <td>-4.478959</td>\n",
       "      <td>7.609836</td>\n",
       "      <td>3.497668e-08</td>\n",
       "      <td>1.003896e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.2500</th>\n",
       "      <td>0.491592</td>\n",
       "      <td>0.279774</td>\n",
       "      <td>0.509769</td>\n",
       "      <td>0.289946</td>\n",
       "      <td>0.498927</td>\n",
       "      <td>0.283970</td>\n",
       "      <td>0.483722</td>\n",
       "      <td>0.297147</td>\n",
       "      <td>0.511696</td>\n",
       "      <td>0.294843</td>\n",
       "      <td>...</td>\n",
       "      <td>0.507315</td>\n",
       "      <td>0.282661</td>\n",
       "      <td>0.468582</td>\n",
       "      <td>0.282024</td>\n",
       "      <td>0.505974</td>\n",
       "      <td>0.274488</td>\n",
       "      <td>-4.266962</td>\n",
       "      <td>8.632905</td>\n",
       "      <td>3.406218e-08</td>\n",
       "      <td>9.982327e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0000</th>\n",
       "      <td>0.500540</td>\n",
       "      <td>0.284262</td>\n",
       "      <td>0.509999</td>\n",
       "      <td>0.288477</td>\n",
       "      <td>0.500035</td>\n",
       "      <td>0.284272</td>\n",
       "      <td>0.488044</td>\n",
       "      <td>0.290338</td>\n",
       "      <td>0.510058</td>\n",
       "      <td>0.288514</td>\n",
       "      <td>...</td>\n",
       "      <td>0.507907</td>\n",
       "      <td>0.287106</td>\n",
       "      <td>0.489301</td>\n",
       "      <td>0.289137</td>\n",
       "      <td>0.501750</td>\n",
       "      <td>0.284551</td>\n",
       "      <td>-1.293317</td>\n",
       "      <td>9.451195</td>\n",
       "      <td>3.271540e-08</td>\n",
       "      <td>9.471624e-10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           cldfrc_rhminl           micromg_dcs           eddydiff_a2l  \\\n",
       "                    mean       std        mean       std         mean   \n",
       "resolution                                                              \n",
       "0.0625          0.501189  0.270830    0.519213  0.289553     0.512733   \n",
       "0.2500          0.491592  0.279774    0.509769  0.289946     0.498927   \n",
       "1.0000          0.500540  0.284262    0.509999  0.288477     0.500035   \n",
       "\n",
       "                     uwshcu_criqc           uwshcu_rkm            ...  \\\n",
       "                 std         mean       std       mean       std  ...   \n",
       "resolution                                                        ...   \n",
       "0.0625      0.278059     0.453806  0.301031   0.509305  0.292473  ...   \n",
       "0.2500      0.283970     0.483722  0.297147   0.511696  0.294843  ...   \n",
       "1.0000      0.284272     0.488044  0.290338   0.510058  0.288514  ...   \n",
       "\n",
       "           zmconv_dmpdz           zmconv_ke           zmconv_tau            \\\n",
       "                   mean       std      mean       std       mean       std   \n",
       "resolution                                                                   \n",
       "0.0625         0.531118  0.286086  0.471942  0.280746   0.508458  0.271686   \n",
       "0.2500         0.507315  0.282661  0.468582  0.282024   0.505974  0.274488   \n",
       "1.0000         0.507907  0.287106  0.489301  0.289137   0.501750  0.284551   \n",
       "\n",
       "           TOA_energy_flux           precipitation                \n",
       "                      mean       std          mean           std  \n",
       "resolution                                                        \n",
       "0.0625           -4.478959  7.609836  3.497668e-08  1.003896e-09  \n",
       "0.2500           -4.266962  8.632905  3.406218e-08  9.982327e-10  \n",
       "1.0000           -1.293317  9.451195  3.271540e-08  9.471624e-10  \n",
       "\n",
       "[3 rows x 26 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_gb = df.groupby('resolution')\n",
    "df_gb.agg(['mean', 'std'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "618353ca",
   "metadata": {},
   "source": [
    "We can also do this individually if we wanted to by calling a column directly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b2339409",
   "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>eddydiff_a2l</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   eddydiff_a2l\n",
       "0           0.5"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['eddydiff_a2l']].mode()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0d571c7b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "eddydiff_a2l    0.501705\n",
       "dtype: float64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['eddydiff_a2l']].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c6e2042b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "resolution\n",
       "1.0000        502\n",
       "0.2500        262\n",
       "0.0625        142\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['resolution']].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c7382f5a",
   "metadata": {},
   "source": [
    "Another important function we may want to use is to check if any columns have NaN (not a number) and filter those if needed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "cd97f252",
   "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>resolution</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>901</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>902</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>903</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>904</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>905</th>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>906 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     resolution\n",
       "0         False\n",
       "1         False\n",
       "2         False\n",
       "3         False\n",
       "4         False\n",
       "..          ...\n",
       "901       False\n",
       "902       False\n",
       "903       False\n",
       "904       False\n",
       "905       False\n",
       "\n",
       "[906 rows x 1 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['resolution']].isnull()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c82e428f",
   "metadata": {},
   "outputs": [],
   "source": [
    "null_only = df[df['resolution'].isna()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "9821e640",
   "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>cldfrc_rhminl</th>\n",
       "      <th>micromg_dcs</th>\n",
       "      <th>eddydiff_a2l</th>\n",
       "      <th>uwshcu_criqc</th>\n",
       "      <th>uwshcu_rkm</th>\n",
       "      <th>uwshcu_rpen</th>\n",
       "      <th>zmconv_alfa</th>\n",
       "      <th>zmconv_c0_ocn</th>\n",
       "      <th>zmconv_dmpdz</th>\n",
       "      <th>zmconv_ke</th>\n",
       "      <th>zmconv_tau</th>\n",
       "      <th>resolution</th>\n",
       "      <th>TOA_energy_flux</th>\n",
       "      <th>precipitation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [cldfrc_rhminl, micromg_dcs, eddydiff_a2l, uwshcu_criqc, uwshcu_rkm, uwshcu_rpen, zmconv_alfa, zmconv_c0_ocn, zmconv_dmpdz, zmconv_ke, zmconv_tau, resolution, TOA_energy_flux, precipitation]\n",
       "Index: []"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "null_only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "2b25928d",
   "metadata": {},
   "outputs": [],
   "source": [
    "without_null = df[df['resolution'].notna()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "c16ae3d5",
   "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>cldfrc_rhminl</th>\n",
       "      <th>micromg_dcs</th>\n",
       "      <th>eddydiff_a2l</th>\n",
       "      <th>uwshcu_criqc</th>\n",
       "      <th>uwshcu_rkm</th>\n",
       "      <th>uwshcu_rpen</th>\n",
       "      <th>zmconv_alfa</th>\n",
       "      <th>zmconv_c0_ocn</th>\n",
       "      <th>zmconv_dmpdz</th>\n",
       "      <th>zmconv_ke</th>\n",
       "      <th>zmconv_tau</th>\n",
       "      <th>resolution</th>\n",
       "      <th>TOA_energy_flux</th>\n",
       "      <th>precipitation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.513158</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>11.078644</td>\n",
       "      <td>3.287566e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.900214</td>\n",
       "      <td>3.378267e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>18.531844</td>\n",
       "      <td>3.236089e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.513158</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>8.339281</td>\n",
       "      <td>3.410277e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.513158</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.278943</td>\n",
       "      <td>0.826606</td>\n",
       "      <td>0.698970</td>\n",
       "      <td>0.231378</td>\n",
       "      <td>0.544417</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>11.162554</td>\n",
       "      <td>3.275170e-08</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>901</th>\n",
       "      <td>0.841311</td>\n",
       "      <td>0.550732</td>\n",
       "      <td>0.862211</td>\n",
       "      <td>0.771593</td>\n",
       "      <td>0.626219</td>\n",
       "      <td>0.712388</td>\n",
       "      <td>0.473620</td>\n",
       "      <td>0.937692</td>\n",
       "      <td>0.952484</td>\n",
       "      <td>0.795456</td>\n",
       "      <td>0.953037</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>3.686973</td>\n",
       "      <td>3.562486e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>902</th>\n",
       "      <td>0.420755</td>\n",
       "      <td>0.150635</td>\n",
       "      <td>0.879170</td>\n",
       "      <td>0.933833</td>\n",
       "      <td>0.374344</td>\n",
       "      <td>0.412366</td>\n",
       "      <td>0.253196</td>\n",
       "      <td>0.537597</td>\n",
       "      <td>0.920912</td>\n",
       "      <td>0.719474</td>\n",
       "      <td>0.022809</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>-6.642911</td>\n",
       "      <td>3.608198e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>903</th>\n",
       "      <td>0.016315</td>\n",
       "      <td>0.200030</td>\n",
       "      <td>0.430349</td>\n",
       "      <td>0.025959</td>\n",
       "      <td>0.347203</td>\n",
       "      <td>0.921252</td>\n",
       "      <td>0.008454</td>\n",
       "      <td>0.543158</td>\n",
       "      <td>0.114437</td>\n",
       "      <td>0.566006</td>\n",
       "      <td>0.151339</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>-5.854439</td>\n",
       "      <td>3.636534e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>904</th>\n",
       "      <td>0.522297</td>\n",
       "      <td>0.052848</td>\n",
       "      <td>0.662738</td>\n",
       "      <td>0.006282</td>\n",
       "      <td>0.277326</td>\n",
       "      <td>0.519419</td>\n",
       "      <td>0.704073</td>\n",
       "      <td>0.868487</td>\n",
       "      <td>0.069545</td>\n",
       "      <td>0.254931</td>\n",
       "      <td>0.286714</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>4.026710</td>\n",
       "      <td>3.458463e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>905</th>\n",
       "      <td>0.876733</td>\n",
       "      <td>0.351345</td>\n",
       "      <td>0.615204</td>\n",
       "      <td>0.457433</td>\n",
       "      <td>0.027828</td>\n",
       "      <td>0.303024</td>\n",
       "      <td>0.687878</td>\n",
       "      <td>0.609783</td>\n",
       "      <td>0.016450</td>\n",
       "      <td>0.072018</td>\n",
       "      <td>0.347606</td>\n",
       "      <td>0.0625</td>\n",
       "      <td>2.388479</td>\n",
       "      <td>3.377284e-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>906 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     cldfrc_rhminl  micromg_dcs  eddydiff_a2l  uwshcu_criqc  uwshcu_rkm  \\\n",
       "0         0.513158     0.750000      0.500000      0.200000    0.750000   \n",
       "1         0.000000     0.750000      0.500000      0.200000    0.750000   \n",
       "2         1.000000     0.750000      0.500000      0.200000    0.750000   \n",
       "3         0.513158     0.000000      0.500000      0.200000    0.750000   \n",
       "4         0.513158     1.000000      0.500000      0.200000    0.750000   \n",
       "..             ...          ...           ...           ...         ...   \n",
       "901       0.841311     0.550732      0.862211      0.771593    0.626219   \n",
       "902       0.420755     0.150635      0.879170      0.933833    0.374344   \n",
       "903       0.016315     0.200030      0.430349      0.025959    0.347203   \n",
       "904       0.522297     0.052848      0.662738      0.006282    0.277326   \n",
       "905       0.876733     0.351345      0.615204      0.457433    0.027828   \n",
       "\n",
       "     uwshcu_rpen  zmconv_alfa  zmconv_c0_ocn  zmconv_dmpdz  zmconv_ke  \\\n",
       "0       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "1       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "2       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "3       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "4       1.000000     0.278943       0.826606      0.698970   0.231378   \n",
       "..           ...          ...            ...           ...        ...   \n",
       "901     0.712388     0.473620       0.937692      0.952484   0.795456   \n",
       "902     0.412366     0.253196       0.537597      0.920912   0.719474   \n",
       "903     0.921252     0.008454       0.543158      0.114437   0.566006   \n",
       "904     0.519419     0.704073       0.868487      0.069545   0.254931   \n",
       "905     0.303024     0.687878       0.609783      0.016450   0.072018   \n",
       "\n",
       "     zmconv_tau  resolution  TOA_energy_flux  precipitation  \n",
       "0      0.544417      1.0000        11.078644   3.287566e-08  \n",
       "1      0.544417      1.0000         0.900214   3.378267e-08  \n",
       "2      0.544417      1.0000        18.531844   3.236089e-08  \n",
       "3      0.544417      1.0000         8.339281   3.410277e-08  \n",
       "4      0.544417      1.0000        11.162554   3.275170e-08  \n",
       "..          ...         ...              ...            ...  \n",
       "901    0.953037      0.0625         3.686973   3.562486e-08  \n",
       "902    0.022809      0.0625        -6.642911   3.608198e-08  \n",
       "903    0.151339      0.0625        -5.854439   3.636534e-08  \n",
       "904    0.286714      0.0625         4.026710   3.458463e-08  \n",
       "905    0.347606      0.0625         2.388479   3.377284e-08  \n",
       "\n",
       "[906 rows x 14 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "without_null"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f58c7afd",
   "metadata": {},
   "source": [
    "#### Splitting the data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2bea25f",
   "metadata": {},
   "source": [
    "For this workshop, let's use the first 12 features (the 11 features listed in the table above plus an added feature for horizontal resolution of the climate models.) Let's also select a value to predict. We can start by predicting the global annual mean net TOA energy flux, which describes the extent to which the Earth's energy budget is in balance. This is an important climate quantity for model developers. We can label these X and Y, respectively."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1f9bb0d6",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = features[:, :12]\n",
    "Y = features[:, 12]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03fb08ec",
   "metadata": {},
   "source": [
    "This will be a regression problem, since we want to predict the value of TOA energy flux."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0800a460",
   "metadata": {},
   "source": [
    "Next, we'll want to split these arrays into a training set and a testing set. The training set will be used to train the model and the testing set will be used to evaluate the model's performance. `sklearn` has a handy function called `train_test_split` which will split the arrays provided into a \"train\" and a \"test\" set based on the percentage of data indicated for the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "52cc60cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, x_test, y_train, y_test = train_test_split(X, Y, test_size=0.2, random_state=12345)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8519f0f8",
   "metadata": {},
   "source": [
    "Great, now that we've split the dataset..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3cb089d7",
   "metadata": {},
   "source": [
    "### Train your own predictive model!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23a8579a",
   "metadata": {},
   "source": [
    "In the paper, the model presented is an Extra Trees Regressor and achieved a goodness-of-fit metric ($R^{2}$) of **0.86**. Can you beat this?\n",
    "\n",
    "You can use the helper function below to also help you visualize how well your model is fitting the data. It will generate a plot of the predictions vs. ground truth."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "b2260311",
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_plot(train_pred, train_actual, test_pred, test_actual):\n",
    "    plt.scatter(train_actual, train_pred, color='darkcyan')\n",
    "    plt.scatter(test_actual, test_pred, color='lightcoral')\n",
    "    plt.show()\n",
    "    plt.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e482cb1d",
   "metadata": {},
   "source": [
    "We've included some codes below to help you get started. The first cell contains the code used to train the model presented in the paper. The next few are some thoughts from Gemma and Cindy for methods that might achieve similar (or better) performance. Can you think of any other methods that can be used on this data?\n",
    "\n",
    "**PRO TIP**: To uncomment a full cell, use `Ctrl` + `A` to select all lines then `Ctrl` + `/` to quick uncomment all selected lines."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "676aace9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The goodness-of-fit is 0.8668940555785043.\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": [
    "# clf = ExtraTreesRegressor(n_estimators=50, max_depth=8, n_jobs=4, random_state=12345, bootstrap=True, oob_score=True)\n",
    "# clf.fit(x_train, y_train)\n",
    "# print(f'The goodness-of-fit is {clf.oob_score_}.')\n",
    "# test_preds = clf.predict(x_test)\n",
    "# train_preds = clf.predict(x_train)\n",
    "# show_plot(train_preds, y_train, test_preds, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "1d02e02b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gonzales72/.local/lib/python3.6/site-packages/sklearn/neural_network/_multilayer_perceptron.py:617: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The goodness-of-fit is 0.9215141735980409.\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAD4CAYAAAAJmJb0AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAA1h0lEQVR4nO2df3RcZ3nnP+9IMx6NkhFYziEQW3ZOCTkr49AfTkq3Pmc3Nji2DiGFKi1db0mBVqUkJOnWsMl6j73xNmy6cWkSkhBEIfxSoMSEEKcOdrA5h2PWNE4KjoggxeD4RzAFy61kazSe0cy7f8y94zt33vfOHc2MNJp5PufoWLr3ztx7beu5zzzv9/k+SmuNIAiC0JpE5vsCBEEQhMYhQV4QBKGFkSAvCILQwkiQFwRBaGEkyAuCILQwnfN9AV6WLFmiV6xYMd+XIQiCsKB44YUXTmutLzHta6ogv2LFCp5//vn5vgxBEIQFhVLqmG2flGsEQRBaGAnygiAILYwEeUEQhBZGgrwgCEILI0FeEAShhWkqdY0gCEI7MDI2xpYDBzg+OUlfMsnda9awqb+/IeeSIC8IglBngoL4yNgYQ3v3kpqZAeDY5CRDe/cCNCTQS7lGEAShjrhB/NjkJJoLQXxkbAyALQcOFAO8S2pmhi0HDjTkeiTIC4Ig1JFKQfz45KTxdbbttSJBXhAEoY5UCuJ9yaRxv217rUiQFwRBqCOVgvjda9aQ6CxdDk10dnL3mjUNuR4J8oIgCHWkUhDf1N/P8Pr1LE8mUcDyZJLh9esbpq5RzTTjdfXq1VoMygRBaFbCSh+rkUjWQ06plHpBa73auE+CvCAIQimmwAuUSB+hkKHXkoX75ZSzfU8J8oIgCCGxBd6uzk7G0+my45cnk7wyNDSrc60YHuaYYaG22vcMCvLSDCUIguDBJoH0b3O5enycyfvuQ09MoHp6iK9bR2zVqlDnmgs5pSy8CoIgeKgmwA5ms3wik0FPTACgJyaY3rWLzOhoqNfPhZxSgrwgCIKHagLstmyWLv/GbJafPvlkscM1iLmQU0q5RhCEtse70Lo4HqcDyAUcryg8DJZOTRn3X5rPh/Kjcfc10qxMFl4FQWhrTAutQXgXRd1avJ/jSnFVIlHTomw1yMKrIAhtj5utXz0+zvaZGV6fzxPp6uJ3p6c5CZxUiu3RKDujUet7+Esp8XXrmN61C7LZ4rYUsN15j0b50VSDBHlBEFoGf9kF4Ew6zeJ4nLOZDO88f54HMhkSzvF6eprXOt/3ac0DmQyANdD79euuiuanTz7Jpfl82YOiUX401SBBXhCElsBfdvFq2t3vt2azxQBvIuEcYwryy5NJY608tmoV/9zRYdTWN8qPphokyAuC0BKY9O1+loZYg7QdExSw52IBdbbIwqsgCC1BZMcOKkWzF1Mp+irEPHfR1Ity/mym4O0laOFVdPKCICwoRsbGWDE8TGTHDlYMDxf16GHq39ujUVIB+72Lpl608+Wf8rQQkCAvCMKCIWi0Xpj69xPRKLfGYvw8EiEPnHa+8hQy+FtjsUB1DTR2VF8jkJq8IAgLBpuvzE3PPEMuROlZA4/feSdQeGC875lnyM6iZN0M0siwSCYvCMKCwRZcwwR4KC3pbOrv59GNG+l1pJYA3Z3h8t5mkEaGRTJ5QRCaHlf/7g3lg9ksW7NZlmodqpEJYHx6msiOHSULqN5F1BXDw0xVyNKbRRoZFlHXCILQ1JhsBwaz2ZKmJigsmoapqbtElSLW0cGU874RCrX5IJZ7Bog0k1xSbA0EQViwmOrwpqamoEYmE1mtyXreNyjA98bjnL7lFqD8oeMu/kKwGdl8ITV5QRCaGlMd3tawFKbZqVoSnZ3cv3Zt8Wfb4m+zKm4kyAuC0JS4enh/2B7MZq1Z90mlSo57MZXizNQUL6ZSDHpMxMKyPJks86uZi2lO9UTKNYIgNB3ekoh3gfUMcDHmwKWBPZFC3uqv2YcxHwtLXzJpnMvarIobyeQFQWg63JKIG6z7tCYCLAEWWV6jgOvyhRw/qGZfDaYO17mY5lRPJJMXBGFe8doDu0oVt/RRyTXSj1uTn03N3qaucevtbsmmmc3ITNQlyCulPgu8A/il1vrNzrbFwD8AK4BXgD/QWv9bPc4nCEJzkRkdJb1vH3piAtXTQ3zduqLXuh+/5/vZTIaMk4G7mfPieJzxdLrqhVQ3SJ9UymhE5q3Ze1FAbvNmq8mZv97u19c3M/XK5D8HPAh8wbPtDmCf1voepdQdzs//vU7nEwShSciMjpZMR9ITE0zv2sV3T57kfSdOlGS7gNXz3SU1M0Pa2W8L1jY6nD+3R6NGHb3JfAwu1NMXWr09DHWpyWutvwOc8W2+Afi88/3ngd+rx7kEQWgu0vv2lYy/AyCb5ZJDh8qMxG7bvz/ULFU3I98ejVZsUPJywsnUdzpGZMeVqmg+5q2nL7R6exgaWZN/ndb6lPP9L4DXmQ5SSg0BQwB9fX0NvBxBEBqBaZA1wGW+DDw1M1OmlrHZEXiPMRdYCmoa7z5/pr4zhM3Bcl89faHV28MwJwuvWmutlDJ+5tJaDwPDULA1mIvrEQShfqieHmOgN9W/g6SNUFhoXaY1msplBu+754GRjo7Q8shEZ2eZ/t1lIdXbw9BICeW/KqVeD+D8+csGnksQhHkivm4d+IKrrf5tkzbek8kUpZKK6gNThAvyyUqYGpxamUZm8k8BNwH3OH9+o4HnEgShgZhkjm6QdFU06X37yE9McFIp7rKUSmxqmV6wlmVc3OlMynJsJSVOUPbeytQlk1dKfRk4CFyplDqplPoAheD+dqXUT4C3OT8LgrDACJrG5BJbtYpd69dzWU8PqxIJa9nEJmEMwwmlWNzdXVxcDfPeHUqhaL/s3UtdMnmt9R9Zdq2rx/sLgjB/BBlyeYOm6Tg/NmljikI3qw1v+SesPLJdM3c/0vEqCEIgNuOtY5OTxWz+tv37jZp3P26G71fXAGWB29vY5FXg2N7D++mhQ6kSZ8h2DvQyNEQQhCKmztU3HTxobBACiEUizOTzVWnZbQxms9yTydDr/DwObOvqYiRSXVVZQUnXajtk9EFDQ8SgTBAE4ELnqiuHdDtXH122rKxBqPiaEAG+GsvfBBcWVpcA905PV2UR7A/w0Nxe73OBBHlBEAB75+qSQ4foCjng2o/fRdLVxZsCdzXOkaYHR3dnp9F3BprX630ukCAvCAIAeUvn6lKtK9bbbdl6UOD2v2ZZSOdI04Pj4ZkZzqxbx3KLx8xC9p6pFQnygiAwMjZWlTTRS1C2btOuL9Oah3yvseE/v+nBEcvnSe/b15LeM7UiQV4QBLYcOMD2aJSUb/t5IKF1YD09KFsPsvb1D/8wHWmSRtoeHHpigk39/QyvX8/yZLLt9fEuIqEUBIHjk5Mc80kT3VF7rn7dNkIvaEDHUCxWJo0Mi4ayhw7Y7YdVTw/Qet4ztSKZvCC0Ae5Q7MiOHawYHi7pVgVYHI8DheB9VSLB4u5uUkqVZdumhVBbtn7GObaLcsVLGFyFjVv66XDOY/rEMQ28cOWVszhL6yM6eUFocbxDsV1ikQgXx2KcSafpSyYZn57mnC94n5maMmaBeWBxd3fxZ7+zJECWQplg9iYGpZxQiod/53f4/EsvWe2Kd3d1tW1pRnTygtDGmOwGMvk84+l00YvGH+DBnqF7t7vBNg7MUMjYT1PfAA+F0s/Db397sd7u/cRxleOV0+56eBsS5AWhxZmtRtxUFslTUMa8mEpxbzpdopDpBFxn+GoCvOsumcNe1ol46u2vDA1Z37+d9fA2ZOFVEJqUIHtfd9+xyUk6lCKnddmUIxfb3FIvQdOatnqkkG5W2Kc1H8jlyrLERUCsyvt0A/Z5CoM/NuVypQu10WjBsz7EPbWzHt6GBHlBaBK8QX1xPM7k+fNkneDq2vu6eGvsOcMx3kB/95o1xeNNwRwInNYEZg/3epcBEsCmRYvo3bChzD/H9aw33VPx9W2uh7chC6+C0ASYFkdNuB2dQZn58mSSV4aGyt7/wN69bJ2YCG3ze5pC4J2N/NGPfx6r/2cvPdu2hXrPoE867UbQwqtk8oLQQEyujv6sFMJ5sUO4mvOxyUmWPPggAOPpdLGcMzY9bWxa6rK8T5hpTZVw6+3erD8PnANMhRVX6x4G0cOHQ4K8IDQI19XRNf1yXR2BskAfdsGwL0QmD5R4zbjlnEstM1BnG8jPUwjelUZn+8s6EQoLtCko+1TRu07mDNUbUdcIQoOwuTqm9+0rOzbMgqFbczb5s4ShmtF7GjgbsO+4Utwci/G5jo5Aq2HbGRcDt8ZiHFeKvPN+23t6jJ9yhNqQIC8IDUJbXB1N202BOxaJ0BuPl3mweP1ZgGInaCVMkkgbCuimXNLoD+jX5fOzCiKvKlWidX9rMsma9etn8U5CJaRcIwgNQvX0GAO6qe7s1pbDLiS6wd4rpayEXxKZJzgAdHi+99fWXQWOrZ4fSDTKr37jN1h+4oQsms4Boq4RhAbhr8kDEI3Sdf31sy5L+GWWbzt3ji2ZjHXWaRAmO4JqmSFkpqgUaB24+CzMHlHXCMI84AaySuoavxRw4PLL2X30aFmW65dZXnv2LH9n0bfbAr1fJz/S0cG7crlZK2kilC+g+uWRKeDHV1/NtRs3zuIMQq1IJi8I80gYfbw7iNpflnkxlTJa7h5XiqsSpfm5d0i2PwDfGiv0qIYt4/jPtT0aLXlw7IlEuC6fL/l0cai3t0y7L9QPyeQFoUkJo493jbf8MssgH3cvQWUZ1zrYNfmqdLyXPBTLQ/5PDh/xHavEU2bekCAvCA0mqDPTpI/3llTyFBZAT6RSPJxM8ojngWAdngEcn5oiQ0GqWCkz9z8U/Au0JksD9zxh6//iKTN/SJAXhAbiL8f4/WX8Rlv+LNqrZtl+9ixv6ujgbTMzxclN5zGP0fOG1EoSR1c/b/O1eSSTsQaKF1Opiou94ikzv4hOXhAaiKkc4/U99+vjTfNSXWL5PO/LZovWvq7fTK6G60sBeyIRjkxN8WnfYO2HMhkeCgjwitKh3f59MmO1OZBMXhAaiE2/7pZp3OB32/79jKfT1jq7i79ssojZjdbTwDjwdZO1r+e9w+DW9b3ZvAb05s2zuDKh3kiQF9qeRrkZfujZZ637/DXqaccGuNFaNze43xGLsTMa5cVUqi4uk/6HU9guXKHxSJAX2ppKNfMgghwmR8bGeOTwYetrz2UyjIyNsam/ny0HDjAwPc0DmUxJl6kfmz3vONBDZaMwnNenHEsBsCt0qsXvi5NrIml2uyM1eaGtqVQzt5H6x39k+oknirYFrsNkZnS0+L5BYW48nWZo715GxsY4PjkZWIuHQoA/+LrXlXnPpChk5WaXHDPewB5kWnbe+fJvy/i2paC4SOuyXNQ0TYMEeaEtGRkbY8XwcMWauYnM6ChZU9Oex2EyjHWw+zDpSyZDZdQD584VnRs1BUuBOIV6+OKKr77AGc/3JtMydxj3zbEYN/ucIv+yq4sfXnNN0X8nlUiwuaurpB4vaprmQso1QtsRpsvUXzP31u1fmp7mDZbXuZl9mLmqUCgP9cbjVs27i2v7u9Myrs/2yjzlmdzFFOSSOw1zXE3+N97vFfDYSy8V5sk6qpmBsTG+KxOamhaxNRDajqAMHi7YCHiHZnsfCmempqwfgX8eibCyq6tsRqsNRSFzHsxmeSiTsSpazlPIrN3F0qAHgvc1GQpB3Y/J+qBa/H9PwvwRZGvQ8HKNUmqDUuplpdQRpdQdjT6fIFQiqJRi0nW7C6MvplKcmZqyDsnIA1s7OwsKlnQapVTRD743HicWKf11cwM8FLJl25AOKMgZP57JcHpqimUhE7OzFDzhTVS74DqYzRbv/8VUisFsNtTahTD/NDTIK6U6gIeAjUA/8EdKKXnsC/OKrcXeHYDtz0xv+eUvGfY0CnViHqbxmY6OktJGJp/n38+f54sDA5y+5RY+u2EDy5NJBrNZxqanGfcETKBiXf1i59xhxYmLsS+sVjMlyu3C9TZKuQ1QYccWCvNHozP5a4AjWuufaa0zwFeAGxp8TkEIxDSFybZYmBkd5QO5XNkviqKw8AmFISBDsRgficfLXp/TuqiiAdiYSvFAJsMbnIlK3oBZKfBWqzw/qRT/Nx43KnL8apggTMoftwFKPGman0YvvF4GnPD8fBL4be8BSqkhYAigr6+vwZcjCNVNYUrv22fNhDqAnm3bAPj63/4tWEogqZkZ3rt7N52RCM+fO2cNmNujUav7o00jr4FzFMoy3ut0A/m3EwlyWnPn+fNVDRZxS0m98ThLp6aMxyzVWlQ0C4B5V9dorYeBYSgsvM7z5Qhtgjs+rxK2Oa0AEc8Yv0rNP3kK5Zsge2Cv0sWtu7uB3ZbFuy6V3gCvgS7gnkyGO86e5cvRKF+ucpFVc6F8deree0mkyqfDphMJ499hozqIhdnR6HLNq8Ayz89LnW2C0NS4OvrjlhKKBk5PTPCpxx4Dwjf/VKqRu8OtTyhVsTyjgSkoy/xdc7AlwEMG87CwuPX23g0bwJ/5R6OF7T5cJdKxyUk0FzqI3XKVMPc0OsgfAq5QSl2ulIoB7wGeavA5BaEmvIHK1CwEhSDaC7zzJz/hfw0Pcy7j7wM1Y3o/U43clvG7A7VngL/v6OCiCudbROGTgYneeDzQY8att8dWraLr+uuLDVCqp8c6p3a2HcRC42houUZrPaOUugXYQ+FT5We11i818pyCUCveQOUvofhDYgL4/V/8grsCyiH3ptO8L5ejg4It8HeU4o0QWCO3NUed8OjbB7NZPpCrbDRsemAkOju5f+1a/nj3buvrvPX22KpVoYZv29Q2osKZPxpek9da7wbs/5MEoU4E1YKDzMT8+AOS2xl6ZmrKWEIJ0pzfm07zp7lc8XWdwLVa8/cdHUY1jotpEdaf8W/NZkN9FPeXiBQUewH8c2NdeuPxWdXRbZ2+osKZP8S7RmgJgmrBmdFRpnftKjETG3/iCf7ivvuMtWJbQLLV0/NQUvf2Ng55A7yLAt5fIQPfGY0y0tHBDBfKMyM+HX6YhqbzlD4YokrxxYGBYgC3yUnvX7u24nubqEaeKswNEuSFliCoFpzetw98dekE8CHLoqApUIG5ng6F7NzVuvsbh2wV70q/eIPZLJtyuWLzUyewKZcreZjYHjpu3f6MUtwci/H1WAwoLA4/unFjSYa+qb+f4fXrWZ5M1mWSU73fT6gd8a4RWoLIjh1Gky4F/JtF550HFnd3F6WCXkbGxrjpmWfKpJGD2ax15ulpCl7tYXxlNPDabpvpAFZ/muO+mvzDmQwx3zEZ4IfXXMO1GzdWvA6hNZhX7xpBmAtsJZa+ZLKoCvHjZsKmRcFN/f0MXXVV2fad0aj1l6YXQvvKnKuwP0hP770W03JmDFhy6BArhodFuijMfzOUINSDu9esKXGKHMxm2ZbNFro1u7ogEoH8BWsx7yJmRCkiO3bQl0zy6LJl/NboKPnpaT4G/BUXRuW52JQvrtVBpV+qHPCXMX/+XYrtHP4Sjc3vZqnWVU25EloXyeSFpsNtRIrs2BE6G/XWggezWT6RyRQlj3p6GpTifDRaHH5xqydw5xw/9qvHx1n53HPo6enAhqLt0ajVvz0CxiEcXuwu9hcIq6ev1FwlGnVBgrzQVNTSMbmpv59Xhob4TFcXXf6duRzxRILdN97I9Zdeytei0bJF0a3ZrNHP3dtQNJjNWpuLoBBcb/VMU5qhfPHV/35+C18olGJu9U1lutX3iQLCPQxEo97eSLlGaCqCVDJhSw42vxk9MVH0rBkZG+O/+hqBgiSJy7Tm36am0NgzIze4mjxo/CzVuqjE8U54esDpnHW1+ZWMxHZGo0SU4n9mMtbmqkZp1MWjZmEgQV5oKurRMal6eoyBXvX0FAOTqWEnaARfkFGYplBnjwOPZDJ8OpMJfBi45wqy8K0U3L3XNX755bz11CnjOMNGadT907Kk/t+8SLlGaCqCVDJhia9bZzTUeuHKK4ulIBPbo1HOhz5LKZ1QHCiiCP7FcjP+MAqaSmjg4KlT3LRyZdEkzfWjaaRGXTxqFg6SyQtNhV8lA9Vno65dQdHGoKsLDfz6c8/xvQA/dXfbPZkMvZ7tldwgww7z0BS8Z9zzb81mQyloKpGamWH30aNlWv96YSrLiEfNwkGCvNBUuFnnbfv3M55OA9Bl6D6thGuo9e1nnuE/PPccXVAyiQmwBnrv9tNTU4G/JLZhHib85mIJR9XjfX21U5tcGhVcbWWZxfF48d/Hi3jUNB9SrhGakmlPJj+eThsVNpWkliNjY1xy6FCZ0sate4ehw7JdU1C8jId6l8Lxe5xB3u6C6xIuBHhNoWPWpKAJQ6OCq60sA4hHzQJBMnmh6ahU73UXTt0RdXAhw/zuq6+y++hRjk9OElGKX9VY9x6noJX342blg9ksn85kQpV03pvL8U+OBNM06COlVDHAu743lY2EGxtcbZ8QzqTTfHFgQNQ1CwAJ8kJTMTI2Zl0YdQO5+wDwh+nUzAyfPHy4+HPOkRTOtu49mM1ysWG719lxZzTK32UyxuP8LAL+JpMJ7FKFQoD/wsAAcOGB1qEUOa1ZnkwycPnlxQdZo4NrkHVw2BGKwvwiQV5oGtz6r40OpYwywSD2RCJ8IJczDrmuhK056iwX6vmD2WyZQVgQiwHV1QXT02X7TipForOzRBEz30G0HgvhwvwiNXmhaTCVaVwSnZ0Vh2X7Gcxmea8vwOcp92W3YSvpeDNx24PAhgKmslljl+rHE4mms+UV6+CFj2TyQtMQpBAZXr/e2sRk455MpiwAR4B35XJ8JMTrK5V6BgM6WoPomplhKBZjazZb0qX67XicR5sweEpZZmEjQV5oGmz13+VO/RcoKx0E0Vvldj9BI/hchUx1ivYCJ50F1rJPEwZJoiDUipRrhKah0ug4t3RQD0ymYF6zsNNTU3w6kyFFQdroNwkzKWRc8lyQRPo7aGergxeE2SJBXpgTwtgH++u/vfE4XZ2d/PHu3cXXVFM2yAfsc8fzuc1R96bTJWP7XHuCJRQy/zOUmo/Z6vUaGIrFeG13N2/s7uZmg5Pk1yxBvjdgsLcgzBYp1wgNpxozK69LpO01YbGVUvzbE8D7nHmqtuOXQEmnrK1efyoSKSnD2Jwko0qR9bw+qtSsh2cLQhCSyQsNZzZmVkGvCVsHr8YDxtbZ6sXbKWvzcd8awoLBHajtVaz4B2wLQr2QTF5oOLMxswp6TVg9i2nhNI85swn7OHDLNF7PeFch8//e+EYOnTsHAfflrjEEKVbEp12oJxLkhYYT1DXpxRvcIk6Hp+01YaSUpkC8JxJhUy5nXTSthPfTgbcUs27ZMr71h3/IRWNj7H76ae5Ip8uGeCwPEbDFp12oNxLkhYYzcPnlPHL4cEkG7u+a9Ac3U4D3viaslNJUE3f9Y9wZsH40FzL+sA6RB0+dYmRsjBtzOTZkMnQ619+nNY/kcoy84x1FC+Qg6jEZSxC8SE1eaCgfevbZsgCvgJtWriwJWkHdri7ua1wVTlg1in+OKgQP4x4Heru7+bMQM1Zd3ECc3rePzlyprVhnLsdPn3wy1Jxa8WkX6o1k8kLDGBkbKwvwUMiUv/ryyyUmW2HKL48cPswjhw/T55h0TfseCu6QbW+ZBDDOUU1hz3B6nPcKM2PVy/HJSfTUlHHfpfl8qLJL2NKWIIRFMnmhYWw5cMCeLafTHHMWUV3b4Epo5+vY5CSPHD5ckvm7Hah+/fvHfQuvUFDJBHW9RinMah0M6Tnv0pdMonp6jPtOOuZqlcbjVWoIE4RqkSAvNIxqSgxBihlvueWI8zXu61a1DcUOYwFsopPCJ4Cwgb6DQoA2zZf11vIr/Z2IIZhQb6RcIzSMsGWYIAazWR7yGI15B3h4R/nZjMJsnxDGKbhJBmU5ri4+TMnGrcK7i6s/ffJJLs3nS9Q1EK7sIoZgQj2RTF5oGPUoMZicJL0knGOq8YLUwB2xGJ/p6Kj4urATpIBiKSa2ahX//Pu/z9KeHq5KJIoBXsouwnwgQV5oGJv6+0MrYGwZdxjHyF6q+498hoK08iPxOH8Wi3Eae7koqGvWr9q5evzCxFcpuwjNgpRrhLqTGR0lvW8femKC0USCzVrzWCQ4DK9dtoyDp06VLKbGKrxmNsx0dLAtdmGWk6ugcRduTbbCJvzH92nNJzIZMqOjxZKNlF2EZqCm3yKl1I1KqZeUUnml1GrfvjuVUkeUUi8rpa6r7TKFhUJmdJTpXbvQExMAJFIpdkxPV1zA/MGvflXMfAezWV6anuYXZ8+GOmclZY6rykklElx8ww18yfDw2BmNcmssVlTHqJ6eQF28aaG3C0jv2weEc90UhLmg1kz+h8C7gU95Nyql+oH3ACuBNwDfUkq9SWsdZvi80MR4s3TV00N83bqSTs70vn3gC+hBC5hFbfvUFB179/KuK64ge/o05IOMgqtDUcjKn7rsMv7PwYPW43ZGozx+++3Fnw8NDxt9aJYnkyyz6OHzExNiTSA0FTUFea31jwBUed3yBuArWuvzwFGl1BHgGsD+GyY0PW6W7gZxPTFR+Bl4vKODLQcO8P2JCePHQ9MCpr/koScmOP/88zUvFGnMdsL/8cgRjiXsrjW98XjJQ+ygodTkLp5G9u4tflrxclIpbtu/X6wJhKahUTX5y4DveX4+6WwrQyk1BAwB9PX1NehyhHpgytLJZhn/5jcZikZJzcxYXR5Nebmp5GEL8KbAXS1BSpmoUjz+a79W8hBLpFI80NFB8qKL+NTMTIkjZCaX48wTT9DleY8UcFc0yrhljJ9YEwjzQcUgr5T6FnCpYdcWrfU3ar0ArfUwMAywevXq6qciC3UlyObWlLkCxFMpUt3dgN2X3bS9GnliWFLO1xLDPpNSZjCbZfvMDK/P51GHDoHvmjpzOf4G+OTmzSXbY6tW8eGnny6zUQjS1Is1gTAfVAzyWuu3zeJ9XwWWeX5e6mwTmhA3sLv2Am6Y89eSVU+PtUThcsIyMemEIcDapiv5yVOQPZoCtwbOAhc572fyqwGzUqZMUWMb6Wd5uB3q7eUqQ3beG48zPTNTUrIRjbwwXzRKJ/8U8B6l1CKl1OXAFcBzDTqXUAPuIqHbmeoPc16/FVPLPtEof+2RJNomJpmkiKZjbdwRixlLPgr4d6VY3N1dbDxylTKVHCS3BQzjLjmHxY/G5jNz/9q1opEXmoaaavJKqXcBnwAuAf5RKfUDrfV1WuuXlFJfBcaAGeBmUdY0J2Esft1asqui8atrvrpnT/FY06AOWxnDf6zGXNZxm5c+7VgY+DGVfSo5SC5PJrnMopApIRotPNwMuEHbVt6SyU9CM6B0A+qis2X16tX6+eefn+/LaDmCgkpkx46Krf3Lk0leGRqy7l/y4IPWxcZqODI1ZSzJnAbe2N3Ni6mUsbxzXCmuClDN+FFAfvNmRrdvDywXmSSiteKXV0Ih+5dMX6gFpdQLWuvVpn1ia9BCmBpwvOUY16Z3aO/eYnNOpcXASrXkkbExJs+fr8v1L66wvZpSUBDuPT+cTFofcBq4KpEgvmdPXZuZZjPUXBBqQYJ8i2AL5kGabTDXld0l0jC15C0HDpCt06fBM5bt7sLuM11d/LdFi0JPazLhfWitWb+eMxZvmpNKWR+MtSCTn4S5RoJ8i2DLECtptk1GWl8cGEBv3swrQ0MVSwje4OQ37Kpm6MZgNmv0fj8P3BOP86WBAc7dfjvveOc7uf7SS0sWWr30xuP8xVveQocTvCNAd2encQF0U38/x66+mmnfOacp6N291Cvbtn1yEnml0CjEoKxFqDYT9AaVao20vDX+iFLktDYadrle70GZtmtrYBuqfRZ4LBJhz/79xevc1N/Ph559lk8ePlx2/L+l0/zuZZfx8NvfHupert24kczSpSWLyR+enjZecz2y7bvXrDHW5EVeKTQKCfItgm1AR7012/7B3DmnVGObzPRIQKA3OT/6cevx4+k0I2NjxYfR7qNHjcfnoWr7gNiqVSWLqzbPmnpk25UUOYJQbyTItwi2DPH+tWuB6oKKzYTMNpgb7N2r7hg9KA/0pgeDH2+jlTd4B2XVtWbcjc62xYJYmEskyLcIQRliZnSU61Mp9NQUpFLw+ONMGuSBI2NjHNi7l60TEyWmYa4J2ZaDBwOHa9jkiDYXykq2Bn7ljDd4B40WrDXjlmxbaCVEJ9/i+J0jvaSAW2Mxvh6L8Z+XLuXgqVO8ODFh1Kr/PBKhv6vLsKdApdKLBl7r+Nu42HTvmoINgr+JyqvXHxkb4/3f/CYZnyVxVCke3bhRArLQVgTp5CWTb3GMzpEO3gx734kTDGaz1nF7l1bwd3eD8aczGeMCqqndeXs0avSYcWWRrlpnqda8qhS/WrmyeJwbxG/bv7+oIOqNx7l/7VpuzOWYvO8+q+e9ILQTEuRbHJu5lou3ZLI1m7Xa+YYZ4bEzGmXYEuRNWt0gC4T3zMzwcc8DYJnWLPv+98ksXRo4Xi/I814CvdCOSJBvMfwWBgcTCRIpuw2Yd2EzqEZusxA2vZ+pBGMbiG3zmPkfptJPNkt6377AYG3zvK/0OkFoVaQZqoUwdb1u1trq9Ohf2LQFYjBbBZuol/WA7YFT6ZOJbX+l1wlCqyJBvoUwdb0+Folwu2MFoClYgtosAbZHo5iq9+cJH6TD2vxC4dNBbzxufB/bA8dm+1tpv3+7DNoW2gUJ8i2ETR/+1c5O3ppM8trubpZ0d1stAaC89q6BL3R0VOUPszMa5apEonie3V1drFu2rOy4iFL8wZVXlnnngMVrPsD218Xmee99XSXTNkFoJSTItxA2fbjr2XKRIVDHPEOqt2azLPLtV8B1FZQ1QfTG4wyvX88RQ7kkqzW7jx4teud48X8iUD09dF1/fWBd3W3iIpsF55OA6XXiBCm0ExLkWwjbpCK3U3PKIKX06sxtdfBaZrFOO8E0yH1xU38/rwwN8aWBgZLr3xmN8tZkkt033kjy9tsrBvjpXbsu1N61Lmbw/teJE6TQTkiQbyFMjpKu6+KWAwcCh4MMZrNWmWTQgqwX01FuhhzGfTHo+isRpKoJOmeY7YKwkBEJZYvglU7+eWcnW7UmceoUau9eMrlcYJb6J8DHMhnjf4ZKypjeeJwz6XSgzcDxyUm+ODAQyg9mtr4u1ahqxAlSaCckyDcZbrA+NjlJh2Pju9zgneIN6ovjcSbPnyfrWP5unZoq857580SCRwyzXBXwMa2NdgQzUHEox+lbbilei42I80lgeP36hvnBqJ4eY0A3qW3Em0ZoJ8S7pokwzf+EQillm9MVGunp4We9vXQePcplhkHZNj+YVCLBG6PRsvfu7uzkxMSEsW6XBxb7/Gb8/MVb3mJ1pvTSiDmmXrdMurogk4Gcx0AhGq24WCsIrYDMeF0gmFQfrvGXO1RDT0yw4mc/Y5nWRLgwnMOdwmRbJE2kUiX17t54nKhSTM3MWGvuJ5VieTJJt0HiCHBRNBoqwEP91StlC63T06A1yjFRC6PGEYR2QMo1TYSpbm7yXPeHZK/RmM1WQPX0lNS7VwwPF429TEZhRKNcef31vOL4yPsdH2ORCIs6OjhXxYi/eqpXjAut+TzEYvR89KN1O48gLHQkk59H/F2Xiw3dn2Hli+5xYZuIvAHXr0n/eSTCnyrFmw4eLE5j+uyGDSWql89u2MCZdLqqua71VK9UWmiVjlZBKCCZ/Dzhr78fm5wkFokQVYqsJ7AHDePw4pZc3Nr8XTMzvCGfJ2Kx2vWrYYxGYU4nKJhVLwf27i1Z5A2a61pv9UrQQqvp79Z7H4LQTkgmX2fCZpCm+nsmnye5aFGx+7PDWVT1Z+b+kG+SOea0RgMnz55l09NPl13L3WvWhHKWDKql2+a6bvVl8wq4aeXKugbYIPsC6WgVhAtIJl9HqskgbfXpM+k0p2+5pfhzZMcOoNRzfU8kwnX5fJkHO5RPaLosn+eBTIZbx8fLrsU0yMOE7VptFsb+EpOmfPC23xK5Wgmj+8nENIv2+J49Vd2HILQyEuTrSFAG6Q9gtuYhf926L5lk5+RkWfnjI5ZrCMqud0aj3LZ/f7ED1s9gNmsc4GGrpdtKJia1jjfA1qucElu1yqieCft3KwjtgJRr6kg1niiPLlvGqG/B0lS3NvnRmHD/ISv5z4yn03zo2WfLrsn9BNDnk2b+l3zeWks3lUymMXfIegNso8splTx8BKGdkCBfR0zqGCjPIDOjo/zm979fonX/RCbD01deWZbJev1cglAUZI1BmneXTx4+XLbf9glgh1LW7Dq2ahVd119f7CpVPT386Jpr2O0b+O0PsI02CKvFA0cQWg0p19SJkbExzjrKEi9RpcoySJPGuwv4rZdfho0by97Dq2xZ8uCDRX27lxyQz+f5eCLBxzyKFzAvzPrz/aAmqiD8JZNrgeHly0vq7Y8uW8Zv7d3LxOOPo3p6+PPOTqPFQj3LKbP1wBGEVkOCfJ3YcuBASbOQS3LRorJgU+2IOu8iZZCYUgOfA7JdXdyRThttD2wENVFVizfAFjtTPYO1/7qjg8lYjMc8XvZSThGExiBBvk4EqWX82BYsUYqJu+4qUYrY/GyCGIlE+Gp3N4s6O0N3pJq6Xmc6Ori4wiSmSpg+tXTmcuxQiu8mk2IQJggNRoJ8nahG0RFft47pXbvK2/KdTNp1jgTYcvBgVQHeJas1b+jqCh3k3Uzfq655+KKL+KRTipmt5NH26SSRSvHKR2waIUEQ6oUsvPqYbTt8NYoO/4IlpsVSZ+DF1ePjoW0D/ByfnKy4YOvFP5v1U87DpZaZqGEHawuC0BhqCvJKqXuVUj9WSr2olPq6Uuo1nn13KqWOKKVeVkpdV/OVzgG1BLNN/f3ctHIlHU7A7lCKm1au5MZcjsn77mPirruYvO8+MqOjQCHQJ2+/nZ5t24oZvB89McEnDLLGwWyW5ckkXxoYME5jcnEz7jASTNvroTbJY5jB2oIgNI5aM/lngTdrra8C/gW4E0Ap1Q+8B1gJbAAeVkqF6aKfV2oJZiNjY3z+pZfIOQE7pzUT3/8+Z7/xjWLJwi3DuIHexZrVKkWXb1MC2JbNFsslQQux7jFeOWGHRWJZ5mzp+RRSi+TRJLMUC2BBmDtqqslrrfd6fvweMOh8fwPwFa31eeCoUuoIcA1wsJbzNZpagpn/AeE2F5X9BTtlGG+QM9Xop4G44yHvZ6nWvOPAAf549+7i9Cg/vfF4sWbuVbuYFnITnZ3ctHIlu48eNdbca+0gtXWmCoLQeOq58Pp+4B+c7y+jEPRdTjrbylBKDQFDAH19fXW8nOqpJZh5HwTWAO+Qm5goWvhCIQh+9+RJLjl0qET2uC2bZZkhgJ9UqnidpgCf6Ozk/rVrjeeezeg7mYkqCAuXikFeKfUt4FLDri1a6284x2yhMBJ0pNoL0FoPA8NQGP9X7evrSS3BbHE8XmxSMnWPejmpFEN79/LdV18tZs8Rpcglyl/1iUympGQzDdxl0Lx3KEVe61BBu9pGIZmJKggLl4pBXmv9tqD9Sqk/Ad4BrNMXBsa+CizzHLbU2dbU1CuYBQ36cLtPUzMzJaPz3pXJGM3BAD7T1VV0Wvzw9LSxsSmvNfnNm6u6zmqQDlJBWJjUVK5RSm0APgr8J621t//9KeAxpdTHgTcAVwDP1XKuuWK2wczb9GTrHp0Bbo3FikHaPcJvD+wdvnGot5fk0FDxPQ4ND4M4LAqCEJJa1TUPAhcDzyqlfqCUegRAa/0S8FVgDPgmcLPWOqx9eVU0y5g3b5A1DfpIAR/0BHgvNnMwV0XjRRwWBUGoBqUDSgtzzerVq/Xzzz8f+nibUmQ+HAf91zKYzbLNKb9MJxJs1rrEq0VxIZM/MzVlfNpq4Ndf//qy0lGtAzcEQWgtlFIvaK1Xm/YtaFuDaoZ0NBp/Pf9Qby+H16zhzf39vAYYGBvju57APHD55Xz+pZdIzcxYyzsnPCoa72CNG3M5rk+l0FNTqM5O4rmGfEgSBKEFWNCZfGTHDmMzkALrIuR8Z8He87v+89eePVumoklRWr93+WBnJ/ekUqW+N9GoNBgJQhsTlMkvaO8a22KjbXsttgX1wH/+8XSa6ZkZfu+GG1j87nejenrIA8eVMgZ4gA9NTpYbmzkNVoIgCH4WdJCvdhGy0WPnKhF0ftfLptcxB7P5v9vkmTa3R0EQ2psFHeSrHfPW6LFzlQhz/iApZKKzk7ShYQrE1VEQBDMLeuEVqtO11+rBUithzm/quoWCF839a9fSm8uVe9GLq6MgCBYWdCZfLfOtMQ9zftOnky8NDHD6llvY1N8vro6CIFTFglbXzIZmUteIxl0QhHoQpK5puyA/V0gwFwRhrmjZZqhmxd/96m1kkkAvCMJc0lY1eT+N8r2Zb6mmIAiCS9tm8o3MtudbqikIguDStpl8I7PtajtxBUEQGkXbBvlGZtvzLdUUBEFwadsg38hsu9pOXEEQhEbRtjX5sPNcZyuFlHF5giA0A20b5MPMcxUppCAICx1phgpgxfCw0WtmeTLJK565q4IgCPNJy/rJNxqRQgqCsNCRIB+ASCEFQVjoSJAPQKSQgiAsdNp24TUMYRZnBUEQmhkJ8hUQKaQgCAsZKdcIgiC0MBLkBUEQWpi2LdfIUA9BENqBtgzy0skqCEK70JblGhnqIQhCu9CWmXyzdrJKCUkQhHrTlpl8M3ayuiWkY5OTaC6UkOo1klAQhPakLYN8M3aySglJEIRG0JblmmbsZG3WEpIgCAubtgzy0HydrH3JpNHWWMzQBEGohbYs1zQjzVhCEgRh4VNTkFdK/W+l1ItKqR8opfYqpd7gbFdKqQeUUkec/b9Zn8ttXWQurCAIjaCmyVBKqaTWetL5/lagX2v9QaXUAPBhYAD4beB+rfVvV3q/ZpsMJQiCsBBo2GQoN8A7dAPuE+MG4Au6wPeA1yilXl/LuQRBEITqqXnhVSl1N/BeYAK41tl8GXDCc9hJZ9spw+uHgCGAvr6+Wi9HEARB8FAxk1dKfUsp9UPD1w0AWustWutlwAhwS7UXoLUe1lqv1lqvvuSSS6q/A0EQBMFKxUxea/22kO81AuwGtgGvAss8+5Y62wRBEIQ5pFZ1zRWeH28Afux8/xTwXkdl81ZgQmtdVqoRBEEQGkut6pqvAVcCeeAY8EGt9atKKQU8CGwAUsD7tNYVZTNKqV857xOWJcDpqi+8NWjXe2/X+wa5d7l3O8u11sZ6d01Bfr5RSj1vkw21Ou167+163yD3Lvc+O6TjVRAEoYWRIC8IgtDCLPQgPzzfFzCPtOu9t+t9g9x7u1LTvS/omrwgCIIQzELP5AVBEIQAJMgLgiC0MAsyyLerxbFS6l6l1I+de/u6Uuo1nn13Ovf9slLqunm8zIaglLpRKfWSUiqvlFrt29fS9w6glNrg3N8RpdQd8309jUQp9Vml1C+VUj/0bFuslHpWKfUT58/Xzuc1NgKl1DKl1LeVUmPO//XbnO013fuCDPLAvVrrq7TWvw48DWx1tm8ErnC+hoBPzs/lNYxngTdrra8C/gW4E0Ap1Q+8B1hJoQHtYaVUx7xdZWP4IfBu4Dveje1w7879PETh/3c/8EfOfbcqn6Pwb+nlDmCf1voKYJ/zc6sxA/yV1rofeCtws/PvXNO9L8gg364Wx1rrvVprd9r39yh4AkHhvr+itT6vtT4KHAGumY9rbBRa6x9prV827Gr5e6dwP0e01j/TWmeAr1C475ZEa/0d4Ixv8w3A553vPw/83lxe01ygtT6ltf5n5/uzwI8ouPfWdO8LMshDweJYKXUC2MSFTN5mcdyKvB94xvm+ne7bTzvcezvcYyVe5/G/+gXwuvm8mEajlFoB/AbwT9R47007yFsp9S3gUsOuLVrrb2ittwBblFJ3UrA43janF9ggKt23c8wWCh/tRuby2hpNmHsXBK21Vkq1rPZbKXUR8DXgdq31ZMEKrMBs7r1pg3y7WhxXum+l1J8A7wDW6QtNDgv+vqGqf3MvLXHvFWiHe6zEvyqlXq+1PuWUYH853xfUCJRSUQoBfkRr/YSzuaZ7X5Dlmna1OFZKbQA+CrxTa53y7HoKeI9SapFS6nIKC8/Pzcc1zgPtcO+HgCuUUpcrpWIUFpqfmudrmmueAm5yvr8JaLlPdo5772eAH2mtP+7ZVdu9a60X3BeFJ90PgReBXcBlznZFQYXwU2AUWD3f11rn+z5CoTb7A+frEc++Lc59vwxsnO9rbcC9v4tCLfo88K/Anna5d+ceBygoqn5KoXw179fUwHv9MoVRoVnn3/wDQC8FZclPgG8Bi+f7Ohtw32soiEhe9PyOD9R672JrIAiC0MIsyHKNIAiCEA4J8oIgCC2MBHlBEIQWRoK8IAhCCyNBXhAEoYWRIC8IgtDCSJAXBEFoYf4/Ofm2BXeLcqAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# clf = MLPRegressor()\n",
    "# clf.fit(x_train, y_train)\n",
    "# print(f'The goodness-of-fit is {clf.score(x_test, y_test)}.')\n",
    "# test_preds = clf.predict(x_test)\n",
    "# train_preds = clf.predict(x_train)\n",
    "# show_plot(train_preds, y_train, test_preds, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "0a7d1a2b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The goodness-of-fit is 0.8341147640130518.\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": [
    "# clf = RandomForestRegressor(n_estimators=50, max_depth=8, random_state=12345)\n",
    "# clf.fit(x_train, y_train)\n",
    "# print(f'The goodness-of-fit is {clf.score(x_test, y_test)}.')\n",
    "# test_preds = clf.predict(x_test)\n",
    "# train_preds = clf.predict(x_train)\n",
    "# show_plot(train_preds, y_train, test_preds, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "77b33d54",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The goodness-of-fit is 0.8158775609361482.\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": [
    "# clf = GradientBoostingRegressor(n_estimators=50, max_depth=8, random_state=12345)\n",
    "# clf.fit(x_train, y_train)\n",
    "# print(f'The goodness-of-fit is {clf.score(x_test, y_test)}.')\n",
    "# test_preds = clf.predict(x_test)\n",
    "# train_preds = clf.predict(x_train)\n",
    "# show_plot(train_preds, y_train, test_preds, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56c717e3",
   "metadata": {},
   "source": [
    "Want to make these models better? Consider hyperparameter tuning! Each of the methods above have keyword arguments for their hyperparameters that you can change the value of to see how the model's performance improves or declines. Consider looking over the source documents to learn more about the hyperparameters you can tune:\n",
    "* [Extra Trees Regressor](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.ExtraTreesRegressor.html)\n",
    "* [MLP Regressor](https://scikit-learn.org/stable/modules/generated/sklearn.neural_network.MLPRegressor.html)\n",
    "* [Random Forest Regressor](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html)\n",
    "* [Gradient Boosting Regressor](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingRegressor.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "28f58f8e",
   "metadata": {},
   "source": [
    "Reviewed and Released: LLNL-PRES-832485"
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}