Crop_models/Seminar_Sensitivity_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "7M3sjbWCff8u"
   },
   "source": [
    "# Introduction to Digital agro \n",
    "\n",
    "## Sensitivity Analysis of crop models\n",
    "\n",
    "____\n",
    "\n",
    "### Mikhail Gasanov\n",
    "\n",
    "E-mail: Mikhail.Gasanov[a]skoltech.ru\n",
    "\n",
    "tg:@misha_grol"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Wq03bRoDff8y"
   },
   "source": [
    "\n",
    "## Clone utils and files from GitHub"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "id": "Jp5PPrbBAHzf",
    "outputId": "77cdc600-09f2-4bb7-8eae-3ea815bb8189"
   },
   "outputs": [],
   "source": [
    "# !git clone https://github.com/EDSEL-skoltech/Intro_to_Digital_Agriculture.git\n",
    "# %cd ./Intro_to_Digital_Agriculture/Crop_models \n",
    "# !pip install SALib\n",
    "# !pip install git+https://github.com/mishagrol/pcse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This notebook was built with:\n",
      "python version: 3.6.11 | packaged by conda-forge | (default, Aug  5 2020, 20:19:23) \n",
      "[GCC Clang 10.0.1 ] \n",
      "PCSE version: 5.4.2\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import sys, os\n",
    "import pcse\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import yaml\n",
    "matplotlib.style.use(\"ggplot\")\n",
    "import matplotlib.pyplot as plt\n",
    "from tqdm.notebook import tqdm\n",
    "\n",
    "from pcse.models import Wofost71_PP\n",
    "from pcse.base import ParameterProvider\n",
    "from pcse.db import NASAPowerWeatherDataProvider\n",
    "from pcse.fileinput import YAMLCropDataProvider\n",
    "from pcse.util import WOFOST71SiteDataProvider, DummySoilDataProvider\n",
    "\n",
    "\n",
    "print(\"This notebook was built with:\")\n",
    "print(\"python version: %s \" % sys.version)\n",
    "print(\"PCSE version: %s\" %  pcse.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "UOsFYjiOff-o"
   },
   "source": [
    "# Sensitivity analysis of models\n",
    "___\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"http://drive.google.com/uc?export=view&id=1uijz-Bm9Mxwv8eI6CL8QsumiP7ISqnFo\">\n",
    "<!-- <img src=\"http://drive.google.com/uc?export=view&id=1dPvni3r4B3FqMFKugxtT8GfH1_W1jNNR\"> -->\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## WOFOST parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>Name</th>\n",
       "      <th>Typical Value</th>\n",
       "      <th>Category</th>\n",
       "      <th>Description</th>\n",
       "      <th>Unit</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CFET</td>\n",
       "      <td>1.000</td>\n",
       "      <td>Water use</td>\n",
       "      <td>correction factor for evapotranspiration</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CVL</td>\n",
       "      <td>0.720</td>\n",
       "      <td>Conversion of assimilates into biomass</td>\n",
       "      <td>efficiency conversion of assimilates into leaf...</td>\n",
       "      <td>kg / kg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CVO</td>\n",
       "      <td>0.820</td>\n",
       "      <td>Conversion of assimilates into biomass</td>\n",
       "      <td>efficiency conversion of assimilates into stor...</td>\n",
       "      <td>kg / kg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CVR</td>\n",
       "      <td>0.720</td>\n",
       "      <td>Conversion of assimilates into biomass</td>\n",
       "      <td>efficiency conversion of assimilates into root...</td>\n",
       "      <td>kg / kg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CVS</td>\n",
       "      <td>0.690</td>\n",
       "      <td>Conversion of assimilates into biomass</td>\n",
       "      <td>efficiency conversion of assimilates into stem...</td>\n",
       "      <td>kg / kg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>DLC</td>\n",
       "      <td>16.000</td>\n",
       "      <td>Phenology</td>\n",
       "      <td>critical day length for development (lower thr...</td>\n",
       "      <td>hours</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>DLO</td>\n",
       "      <td>8.000</td>\n",
       "      <td>Phenology</td>\n",
       "      <td>optimum day length for development</td>\n",
       "      <td>hours</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>PERDL</td>\n",
       "      <td>0.030</td>\n",
       "      <td>Death rates</td>\n",
       "      <td>max. relative death rate of leaves due to wate...</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Q10</td>\n",
       "      <td>2.000</td>\n",
       "      <td>Maintenance respiration</td>\n",
       "      <td>increase of the respiration rate per 10 °C tem...</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>RDI</td>\n",
       "      <td>10.000</td>\n",
       "      <td>Water use</td>\n",
       "      <td>initial rooting depth</td>\n",
       "      <td>cm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>RDMCR</td>\n",
       "      <td>120.000</td>\n",
       "      <td>Water use</td>\n",
       "      <td>crop-dependent maximum rooting depth</td>\n",
       "      <td>cm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>RGRLAI</td>\n",
       "      <td>0.016</td>\n",
       "      <td>Initial state</td>\n",
       "      <td>max. relative increase in leaf area index</td>\n",
       "      <td>day-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>RML</td>\n",
       "      <td>0.030</td>\n",
       "      <td>Maintenance respiration</td>\n",
       "      <td>maintenance respiration rate coefficient of le...</td>\n",
       "      <td>day-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>RMO</td>\n",
       "      <td>0.003</td>\n",
       "      <td>Maintenance respiration</td>\n",
       "      <td>maintenance respiration rate coefficient of st...</td>\n",
       "      <td>day-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>RMR</td>\n",
       "      <td>0.015</td>\n",
       "      <td>Maintenance respiration</td>\n",
       "      <td>maintenance respiration rate coefficient of ro...</td>\n",
       "      <td>day-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>RMS</td>\n",
       "      <td>0.015</td>\n",
       "      <td>Maintenance respiration</td>\n",
       "      <td>maintenance respiration rate coefficient of st...</td>\n",
       "      <td>day-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>SPA</td>\n",
       "      <td>0.000</td>\n",
       "      <td>Crop green area</td>\n",
       "      <td>specific pod area</td>\n",
       "      <td>ha / kg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>SPAN</td>\n",
       "      <td>37.000</td>\n",
       "      <td>Crop green area</td>\n",
       "      <td>life span of leaves growing at 35 Celsius</td>\n",
       "      <td>day</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>TBASE</td>\n",
       "      <td>3.000</td>\n",
       "      <td>Crop green area</td>\n",
       "      <td>lower threshold temperature for physiological ...</td>\n",
       "      <td>˚C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>TBASEM</td>\n",
       "      <td>3.000</td>\n",
       "      <td>Crop green area</td>\n",
       "      <td>lower threshold temperature for emergence</td>\n",
       "      <td>˚C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>TDWI</td>\n",
       "      <td>0.510</td>\n",
       "      <td>Initial state</td>\n",
       "      <td>initial total crop dry weight</td>\n",
       "      <td>kg / ha</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>TEFFMX</td>\n",
       "      <td>20.000</td>\n",
       "      <td>Emergence</td>\n",
       "      <td>max. effective temperature for emergence</td>\n",
       "      <td>˚C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>TSUM1</td>\n",
       "      <td>800.000</td>\n",
       "      <td>Phenological development</td>\n",
       "      <td>temperature sum from emergence to anthesis</td>\n",
       "      <td>˚C day</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>TSUM2</td>\n",
       "      <td>1400.000</td>\n",
       "      <td>Phenological development</td>\n",
       "      <td>temperature sum from anthesis to maturity</td>\n",
       "      <td>˚C day</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>TSUMEM</td>\n",
       "      <td>90.000</td>\n",
       "      <td>Emergence</td>\n",
       "      <td>temperature sum from sowing to emergence</td>\n",
       "      <td>˚C day</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>VERNBASE</td>\n",
       "      <td>14.000</td>\n",
       "      <td>Phenological development</td>\n",
       "      <td>base vernalization requirement</td>\n",
       "      <td>day</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>VERNDVS</td>\n",
       "      <td>0.300</td>\n",
       "      <td>Phenological development</td>\n",
       "      <td>critical DVS for vernalization to switch off</td>\n",
       "      <td>-</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>VERNSAT</td>\n",
       "      <td>70.000</td>\n",
       "      <td>Phenological development</td>\n",
       "      <td>saturated vernalization requirement</td>\n",
       "      <td>day</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Name  Typical Value                                Category  \\\n",
       "0       CFET          1.000                               Water use   \n",
       "1        CVL          0.720  Conversion of assimilates into biomass   \n",
       "2        CVO          0.820  Conversion of assimilates into biomass   \n",
       "3        CVR          0.720  Conversion of assimilates into biomass   \n",
       "4        CVS          0.690  Conversion of assimilates into biomass   \n",
       "5        DLC         16.000                               Phenology   \n",
       "6        DLO          8.000                               Phenology   \n",
       "7      PERDL          0.030                             Death rates   \n",
       "8        Q10          2.000                 Maintenance respiration   \n",
       "9        RDI         10.000                               Water use   \n",
       "10     RDMCR        120.000                               Water use   \n",
       "11    RGRLAI          0.016                           Initial state   \n",
       "12       RML          0.030                 Maintenance respiration   \n",
       "13       RMO          0.003                 Maintenance respiration   \n",
       "14       RMR          0.015                 Maintenance respiration   \n",
       "15       RMS          0.015                 Maintenance respiration   \n",
       "16       SPA          0.000                         Crop green area   \n",
       "17      SPAN         37.000                         Crop green area   \n",
       "18     TBASE          3.000                         Crop green area   \n",
       "19    TBASEM          3.000                         Crop green area   \n",
       "20      TDWI          0.510                           Initial state   \n",
       "21    TEFFMX         20.000                               Emergence   \n",
       "22     TSUM1        800.000                Phenological development   \n",
       "23     TSUM2       1400.000                Phenological development   \n",
       "24    TSUMEM         90.000                               Emergence   \n",
       "25  VERNBASE         14.000                Phenological development   \n",
       "26   VERNDVS          0.300                Phenological development   \n",
       "27   VERNSAT         70.000                Phenological development   \n",
       "\n",
       "                                          Description     Unit  \n",
       "0            correction factor for evapotranspiration        -  \n",
       "1   efficiency conversion of assimilates into leaf...  kg / kg  \n",
       "2   efficiency conversion of assimilates into stor...  kg / kg  \n",
       "3   efficiency conversion of assimilates into root...  kg / kg  \n",
       "4   efficiency conversion of assimilates into stem...  kg / kg  \n",
       "5   critical day length for development (lower thr...    hours  \n",
       "6                  optimum day length for development    hours  \n",
       "7   max. relative death rate of leaves due to wate...        -  \n",
       "8   increase of the respiration rate per 10 °C tem...        -  \n",
       "9                               initial rooting depth       cm  \n",
       "10               crop-dependent maximum rooting depth       cm  \n",
       "11          max. relative increase in leaf area index    day-1  \n",
       "12  maintenance respiration rate coefficient of le...    day-1  \n",
       "13  maintenance respiration rate coefficient of st...    day-1  \n",
       "14  maintenance respiration rate coefficient of ro...    day-1  \n",
       "15  maintenance respiration rate coefficient of st...    day-1  \n",
       "16                                  specific pod area  ha / kg  \n",
       "17          life span of leaves growing at 35 Celsius      day  \n",
       "18  lower threshold temperature for physiological ...       ˚C  \n",
       "19          lower threshold temperature for emergence       ˚C  \n",
       "20                      initial total crop dry weight  kg / ha  \n",
       "21           max. effective temperature for emergence       ˚C  \n",
       "22         temperature sum from emergence to anthesis   ˚C day  \n",
       "23          temperature sum from anthesis to maturity   ˚C day  \n",
       "24           temperature sum from sowing to emergence   ˚C day  \n",
       "25                     base vernalization requirement      day  \n",
       "26       critical DVS for vernalization to switch off        -  \n",
       "27                saturated vernalization requirement      day  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = pd.read_excel(\"./data/ScalarParametersOfWofost-Potential.xlsx\")\n",
    "display(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sobol’ Sequences versus Random numbers and regular grid\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"http://drive.google.com/uc?export=view&id=1ezjA8aa50P08EUybwdDyo8JKCP7lU7zw\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "gnfye784nEKx"
   },
   "outputs": [],
   "source": [
    "from SALib.sample import saltelli\n",
    "from SALib.analyze import sobol\n",
    "from SALib.test_functions import Ishigami\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "8NA-bZpWQi0f"
   },
   "source": [
    "__Docs [SALib](https://salib.readthedocs.io/en/latest/#)__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "l4W7QknHnRud"
   },
   "source": [
    "In this example, we will perform a Sobol’ sensitivity analysis of the _Ishigami_ function, shown below. The _Ishigami_ function is commonly used to test uncertainty and sensitivity analysis methods because it exhibits strong nonlinearity and nonmonotonicity."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "iUoSwcBJnE5n"
   },
   "source": [
    "$f(x)=\\sin \\left(x_{1}\\right)+ \\text{a}\\, \\operatorname{sin}^{2}\\left(x_{2}\\right)+ \\text{b}\\,  x_{3}^{4} \\sin \\left(x_{1}\\right)$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "DBW4m3p8ff-u"
   },
   "outputs": [],
   "source": [
    "problem = {\n",
    "  'num_vars': 3,\n",
    "  'names': ['x1', 'x2', 'x3'],\n",
    "  'bounds': [[-np.pi, np.pi]]*3\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "id": "sYxbSlYbsiyz",
    "outputId": "43781abd-3325-48fd-fa3e-f0de80fe24bd"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8000, 3)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generate samples\n",
    "param_values = saltelli.sample(problem, 10, calc_second_order=True)\n",
    "param_values.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Vr5zlk3off-y"
   },
   "source": [
    "Here, `param_values` is a NumPy matrix. If we run `param_values.shape`, we see that the matrix is **8000 by 3**. The Saltelli sampler generated 8000 samples. The Saltelli sampler generates $N∗(2D+2)$ samples, where in this example $N$ is 1000 (the argument we supplied) and $D$ is 3 (the number of model inputs). The keyword argument `calc_second_order=False` will exclude second-order indices, resulting in a smaller sample matrix with $N∗(D+2)$ rows instead.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 187
    },
    "colab_type": "code",
    "id": "bQ9e0-w_ff-z",
    "outputId": "0a6e2cd1-eb3b-4e96-ef44-96a5af1cf384"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parameter S1 S1_conf ST ST_conf\n",
      "x1 0.307975 0.060295 0.560137 0.096384\n",
      "x2 0.447767 0.050511 0.438722 0.039817\n",
      "x3 -0.004255 0.057172 0.242845 0.024133\n",
      "\n",
      "Parameter_1 Parameter_2 S2 S2_conf\n",
      "x1 x2 0.012205 0.086730\n",
      "x1 x3 0.251526 0.085752\n",
      "x2 x3 -0.009954 0.065483\n"
     ]
    }
   ],
   "source": [
    "# Run model (example)\n",
    "Y = Ishigami.evaluate(param_values)\n",
    "\n",
    "# Perform analysis\n",
    "Si = sobol.analyze(problem, Y, print_to_console=True)\n",
    "# Returns a dictionary with keys 'S1', 'S1_conf', 'ST', and 'ST_conf'\n",
    "# (first and total-order indices with bootstrap confidence intervals)\n",
    "T_Si, first_Si, (idx, second_Si) = sobol.Si_to_pandas_dict(Si)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>S1</th>\n",
       "      <th>ST</th>\n",
       "      <th>S1_conf</th>\n",
       "      <th>ST_conf</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.307975</td>\n",
       "      <td>0.560137</td>\n",
       "      <td>0.060295</td>\n",
       "      <td>0.096384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.447767</td>\n",
       "      <td>0.438722</td>\n",
       "      <td>0.050511</td>\n",
       "      <td>0.039817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.004255</td>\n",
       "      <td>0.242845</td>\n",
       "      <td>0.057172</td>\n",
       "      <td>0.024133</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         S1        ST   S1_conf   ST_conf\n",
       "0  0.307975  0.560137  0.060295  0.096384\n",
       "1  0.447767  0.438722  0.050511  0.039817\n",
       "2 -0.004255  0.242845  0.057172  0.024133"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Si_dict = dict(Si) \n",
    "sensitivity_indices = ['S1', 'ST', 'S1_conf', 'ST_conf']\n",
    "Si_df = pd.DataFrame()\n",
    "for SI in sensitivity_indices:\n",
    "    Si_df = Si_df.append(pd.Series(Si_dict[SI]), ignore_index=True)\n",
    "\n",
    "Si_df = Si_df.T\n",
    "Si_df.columns = sensitivity_indices\n",
    "Si_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xb2Ag_6g_TAd"
   },
   "source": [
    "Consider the model output as\n",
    "\\begin{eqnarray*}\n",
    "Y=f(X)=f\\left(X_{1}, \\ldots, X_{p}\\right),\n",
    "\\end{eqnarray*}\n",
    "where $f$ in our case part of agro-model simulator, $X$ are $p$ varied input parameters and $Y$ is the predicted output. Following the techniques by Sobol we represent the multi-variate random function $f$ using Hoeffding decomposition:\n",
    "\\begin{equation}\n",
    "f(X_1,\\dots,X_p) = f_0 + \\sum_i^p f_i + \\sum_i^p\\sum_{j>i}^p f_{ij} + \\dots + f_{1\\dots p},\n",
    "\\end{equation}\n",
    "where $f_0$ is a constant term, $f_i = f_i(X_i)$ denotes main effects, $f_{ij} = f_{ij}(X_i, X_j)$ and others describe higher-order interactions. These terms can be written as\n",
    "\\begin{equation*}\n",
    "\\begin{split}\n",
    "f_0 &= E(Y),\\\\\n",
    "f_i &= E_{X_{\\sim i}}(Y | X_i) - E(Y),\\\\\n",
    "f_{ij} &= E_{X_{\\sim ij}}(Y | X_i, X_j) - f_i - f_j - f_0,\\\\\n",
    "\\dots\n",
    "\\end{split}\n",
    "\\end{equation*}\n",
    "where $E$ is mathematical expectation and $X_{\\sim i}$ denotes all parameters except $i^\\text{th}$. Under the assumption that the input parameters are independent, total variance $V(Y)$ of the crop yield can be decomposed as follows:\n",
    "\\begin{equation*}\n",
    "V(Y) = \\sum_i^p V_i + \\sum_i^p\\sum_{j>i}^p V_{ij} + \\dots + V_{12\\dots p},\n",
    "\\end{equation*}\n",
    "where partial variances are\n",
    "\\begin{equation*}\n",
    "\\begin{split}\n",
    "V_i &= V[f_i(X_i)] = V_{X_i}\\left[E_{X_{\\sim i}}(Y | X_i)\\right],\\\\\n",
    "V_{ij} &= V[f_{ij}(X_i,X_j)] = V_{X_iX_j}\\left[E_{X_{\\sim ij}}(Y | X_i, X_j)\\right] - V_i - V_j,\\\\\n",
    "\\dots\n",
    "\\end{split}\n",
    "\\end{equation*}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "lRL6RZj6_kFd"
   },
   "source": [
    "## Sobol index (first order, second order, total index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "DaeCBQQF_b5f"
   },
   "source": [
    "This way, sensitivity indices (SI) can be introduced as \n",
    "\\begin{equation}\n",
    "\\Large\n",
    "S_i = \\frac{V_i}{V(Y)},~S_{ij} = \\frac{V_{ij}}{V(Y)},~\\dots\n",
    "\\end{equation}\n",
    "In order to incorporate all of the interactions for a particular parameter, one can compute the total effect index:\n",
    "\\begin{equation}\n",
    "S_{T_i} = \\frac{E_{X_{\\sim i}}\\left[V_{X_i}(Y|X_{\\sim i})\\right]}{V(Y)} = 1 - \\frac{V_{X_{\\sim i}}\\left[E_{X_i}(Y | X_{\\sim i})\\right]}{V(Y)}\n",
    "\\end{equation}\n",
    "\n",
    "\n",
    "From this assumption we can conclude:\n",
    "\\begin{equation}\n",
    "\\Large\n",
    "0 \\leq S_i \\leq S_{T_i} \\leq 1\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "JCKO1zaU_4Kk"
   },
   "source": [
    "More -\n",
    "* [Wiki](https://en.wikipedia.org/wiki/Sobol_sequence)\n",
    "* [Habr](https://habr.com/ru/post/440892/)\n",
    "* Feature selection [Skoltech ML 2020](https://github.com/adasegroup/ML2020_lectures/blob/master/lecture9/Lecture_9_Model_Feature_Selection_Sensitivity.pdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "IM9HLmySff98"
   },
   "source": [
    "# Sensitivity analysis of WOFOST model "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "mBpW6Vi_81Ba"
   },
   "source": [
    "<img src=\"http://drive.google.com/uc?export=view&id=1OPZ0xIFpc8Ku3wtt6XffKy2QKLashDWP\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model input info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define location, crop and season\n",
    "latitude = 52.0\n",
    "longitude = 5.0\n",
    "crop_name = 'sugarbeet'\n",
    "variety_name = 'Sugarbeet_601'\n",
    "campaign_start_date = '2006-01-01'\n",
    "emergence_date = \"2006-03-31\"\n",
    "harvest_date = \"2006-10-20\"\n",
    "max_duration = 300"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "- 2006-01-01:\n",
      "    CropCalendar:\n",
      "        crop_name: sugarbeet\n",
      "        variety_name: Sugarbeet_601\n",
      "        crop_start_date: 2006-03-31\n",
      "        crop_start_type: emergence\n",
      "        crop_end_date: 2006-10-20\n",
      "        crop_end_type: harvest\n",
      "        max_duration: 300\n",
      "    TimedEvents: null\n",
      "    StateEvents: null\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Here we define the agromanagement for sugar beet\n",
    "agro_yaml = \"\"\"\n",
    "- {start}:\n",
    "    CropCalendar:\n",
    "        crop_name: {cname}\n",
    "        variety_name: {vname}\n",
    "        crop_start_date: {startdate}\n",
    "        crop_start_type: emergence\n",
    "        crop_end_date: {enddate}\n",
    "        crop_end_type: harvest\n",
    "        max_duration: {maxdur}\n",
    "    TimedEvents: null\n",
    "    StateEvents: null\n",
    "\"\"\".format(cname=crop_name, vname=variety_name, \n",
    "           start=campaign_start_date, startdate=emergence_date, \n",
    "           enddate=harvest_date, maxdur=max_duration)\n",
    "agro = yaml.safe_load(agro_yaml)\n",
    "print(agro_yaml)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# Weather data for Netherlands\n",
    "wdp = NASAPowerWeatherDataProvider(latitude=latitude, longitude=longitude)\n",
    "\n",
    "from pcse.fileinput import csvweatherdataprovider\n",
    "# Download the weather from CSV file if NASA POWER server doesn't response\n",
    "# wdp = csvweatherdataprovider.CSVWeatherDataProvider('data/meteo/NASA_weather_latitude_52.0_longitude_5.0.csv')\n",
    "\n",
    "# Parameter sets for crop, soil and site\n",
    "# Standard crop parameter library\n",
    "cropd = YAMLCropDataProvider()\n",
    "# We don't need soil for potential production, so we use dummy values\n",
    "soild = DummySoilDataProvider()\n",
    "# Some site parameters\n",
    "sited = WOFOST71SiteDataProvider(WAV=50, CO2=360.)\n",
    "\n",
    "# Retrieve all parameters in the form of a single object. \n",
    "# In order to see all parameters for the selected crop already, we\n",
    "# synchronise data provider cropd with the crop/variety: \n",
    "firstkey = list(agro[0])[0]\n",
    "cropcalendar = agro[0][firstkey]['CropCalendar'] \n",
    "cropd.set_active_crop(cropcalendar['crop_name'], cropcalendar['variety_name'])\n",
    "params = ParameterProvider(cropdata=cropd, sitedata=sited, soildata=soild)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the target variable\n",
    "target_variable = \"TWSO\"\n",
    "\n",
    "# For each scalar parameter, determine a sensible interval \n",
    "\n",
    "Crop_parameters = ['TSUM1','TSUM2','SPAN','Q10','TDWI']\n",
    "Model_parameters = {\n",
    "    'num_vars':5,\n",
    "    'names': Crop_parameters,\n",
    "    'bounds':[[500, 800],\n",
    "            [1200, 1600],\n",
    "            [28, 37],\n",
    "            [1.8, 2.2],\n",
    "            [0.4, 0.6]]}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "EZAWCjDu7xVT"
   },
   "source": [
    "## Generate input parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cuput0rjff-C"
   },
   "outputs": [],
   "source": [
    "param_values = saltelli.sample(Model_parameters, 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Ha0_KvGHff-E"
   },
   "source": [
    "$n = N \\times (D \\times 2 +2)$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "id": "jwv9Lslkff-F",
    "outputId": "be898344-e4ab-4636-b5ab-625ca5fde263"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(120, 5)"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_values.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "16a213fe870742429bad29cd6014b23c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "nsets = len(param_values)\n",
    "\n",
    "\n",
    "# Loop over the samples of the parameter values and run WOFOST for each of the samples\n",
    "target_results = []\n",
    "pbar=tqdm()\n",
    "pbar.reset(total=nsets)\n",
    "for i, parameter in enumerate(param_values):\n",
    "    pbar.update()\n",
    "    params.clear_override()\n",
    "    for name, value in zip(Model_parameters[\"names\"], parameter):\n",
    "        params.set_override(name, value)\n",
    "    wofost = Wofost71_PP(params, wdp, agro)\n",
    "    wofost.run_till_terminate()\n",
    "    r = wofost.get_summary_output()\n",
    "    target_result = r[0][target_variable]\n",
    "#     print('Parameter:', parameter, 'Yield:', target_result)\n",
    "    if target_result is None:\n",
    "        print(\"Target variable is not available in summary output!\")\n",
    "        break\n",
    "    target_results.append(target_result)\n",
    "\n",
    "pbar.refresh()\n",
    "target_results = np.array(target_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "-MJQX33IMZh9"
   },
   "outputs": [],
   "source": [
    "\n",
    "Si = sobol.analyze(Model_parameters, target_results, print_to_console=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "colab_type": "code",
    "id": "dLlY6CSpff-g",
    "outputId": "3fd9db37-0422-49b0-9c28-9434aa780085"
   },
   "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>S1</th>\n",
       "      <th>ST</th>\n",
       "      <th>S1_conf</th>\n",
       "      <th>ST_conf</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Parameters</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>TSUM1</th>\n",
       "      <td>0.913165</td>\n",
       "      <td>1.042718</td>\n",
       "      <td>0.518546</td>\n",
       "      <td>0.806036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TSUM2</th>\n",
       "      <td>-0.048769</td>\n",
       "      <td>0.013741</td>\n",
       "      <td>0.077270</td>\n",
       "      <td>0.018143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SPAN</th>\n",
       "      <td>0.310392</td>\n",
       "      <td>0.312428</td>\n",
       "      <td>0.254017</td>\n",
       "      <td>0.251188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Q10</th>\n",
       "      <td>0.069177</td>\n",
       "      <td>0.024272</td>\n",
       "      <td>0.133090</td>\n",
       "      <td>0.016854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TDWI</th>\n",
       "      <td>0.036392</td>\n",
       "      <td>0.001314</td>\n",
       "      <td>0.031586</td>\n",
       "      <td>0.001273</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  S1        ST   S1_conf   ST_conf\n",
       "Parameters                                        \n",
       "TSUM1       0.913165  1.042718  0.518546  0.806036\n",
       "TSUM2      -0.048769  0.013741  0.077270  0.018143\n",
       "SPAN        0.310392  0.312428  0.254017  0.251188\n",
       "Q10         0.069177  0.024272  0.133090  0.016854\n",
       "TDWI        0.036392  0.001314  0.031586  0.001273"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Si_dict = dict(Si) \n",
    "sensitivity_indices = ['S1', 'ST', 'S1_conf', 'ST_conf']\n",
    "Si_df = pd.DataFrame()\n",
    "for SI in sensitivity_indices:\n",
    "    Si_df = Si_df.append(pd.Series(Si_dict[SI]), ignore_index=True)\n",
    "\n",
    "Si_df = Si_df.T\n",
    "Si_df.columns = sensitivity_indices\n",
    "Si_df['Parameters']=Crop_parameters\n",
    "Si_df.set_index('Parameters', inplace=True)\n",
    "Si_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "TIeMDNzga2_y"
   },
   "source": [
    "Is it ok?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "QqllR3h_RK7d"
   },
   "source": [
    "\n",
    "\\begin{equation}\n",
    "\\Large\n",
    "0 \\leq S_i \\leq S_{T_i} \\leq 1\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "PaaUpFfQff-2"
   },
   "source": [
    "# Visualization for sensitivity analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xyAx8-8Fff-4"
   },
   "source": [
    "Plots by [Water programming group](https://waterprogramming.wordpress.com/2019/08/27/a-python-implementation-of-grouped-radial-convergence-plots-to-visualize-sobol-sensitivity-analysis-results/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "5CbapN-Off-2"
   },
   "source": [
    "How to repeat: [Repo of SampleVIS](https://github.com/charlesrouge/SampleVis)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "qdag8yBlff-4"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from SALib.analyze import sobol\n",
    "from SALib.sample import saltelli\n",
    "from src.fishery import fish_game\n",
    "import matplotlib.pyplot as plt\n",
    "import itertools\n",
    "import math"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "TkyL56lTff-6"
   },
   "source": [
    "### Why number of samples is important?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "mcLu7IlTff-7",
    "outputId": "16c68f29-eb6e-4674-b9bf-57ad321aad61"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n= 50\n",
      "n= 100\n",
      "n= 150\n",
      "n= 200\n",
      "n= 250\n",
      "n= 300\n",
      "n= 350\n",
      "n= 400\n",
      "n= 450\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fe7ed03e390>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set up dictionary with system parameters\n",
    "problem = {\n",
    "  'num_vars': 6,\n",
    "  'names': ['a', 'b', 'c','h',\n",
    "            'K','m'],\n",
    "  'bounds': [[ 0.002, 2],\n",
    "             [0.005, 1],\n",
    "             [0.2, 1],\n",
    "             [0.001, 1],\n",
    "             [100, 5000],\n",
    "             [0.1, 1.5]]\n",
    "}\n",
    "\n",
    "# Array with n's to use\n",
    "nsamples = np.arange(50, 500, 50)\n",
    "\n",
    "# Arrays to store the index estimates\n",
    "S1_estimates = np.zeros([problem['num_vars'],len(nsamples)])\n",
    "ST_estimates = np.zeros([problem['num_vars'],len(nsamples)])\n",
    "\n",
    "# Loop through all n values, create sample, evaluate model and estimate S1 & ST\n",
    "for i in range(len(nsamples)):\n",
    "    print('n= '+ str(nsamples[i]))\n",
    "    # Generate samples\n",
    "    sampleset = saltelli.sample(problem, nsamples[i],calc_second_order=False)\n",
    "    # Run model for all samples\n",
    "    output = [fish_game(*sampleset[j,:]) for j in range(len(sampleset))]\n",
    "    # Perform analysis\n",
    "    results = sobol.analyze(problem, np.asarray(output), calc_second_order=False,print_to_console=False)\n",
    "    # Store estimates\n",
    "    ST_estimates[:,i]=results['ST']\n",
    "    S1_estimates[:,i]=results['S1']\n",
    "\n",
    "np.save('ST_estimates.npy', ST_estimates)\n",
    "np.save('S1_estimates.npy', S1_estimates)\n",
    "\n",
    "S1_estimates = np.load('S1_estimates.npy')\n",
    "ST_estimates = np.load('ST_estimates.npy')\n",
    "\n",
    "# Generate figure showing evolution of indices\n",
    "fig =  plt.figure(figsize=(18,9))\n",
    "ax1 = fig.add_subplot(1,2,1)\n",
    "handles = []\n",
    "for j in range(problem['num_vars']):\n",
    "    handles += ax1.plot(nsamples, S1_estimates[j,:], linewidth=5)\n",
    "ax1.set_title('Evolution of S1 index estimates', fontsize=20)\n",
    "ax1.set_ylabel('S1', fontsize=18)\n",
    "ax1.set_xlabel('Number of samples (n)', fontsize=18)\n",
    "ax1.tick_params(axis='both', which='major', labelsize=14)\n",
    "ax2 = fig.add_subplot(1,2,2)\n",
    "for j in range(problem['num_vars']):\n",
    "    ax2.plot(nsamples, ST_estimates[j,:], linewidth=5)\n",
    "ax2.set_title('Evolution of ST index estimates', fontsize=20)\n",
    "ax2.set_ylabel('ST', fontsize=18)\n",
    "ax2.tick_params(axis='both', which='major', labelsize=14)\n",
    "ax2.set_xlabel('Number of samples (n)', fontsize=18)\n",
    "fig.legend(handles, problem['names'], loc = 'right', fontsize=11)\n",
    "plt.show()\n",
    "#plt.savefig('indexevolution.png')\n",
    "\n",
    "# Calculate parameter rankings\n",
    "S1_ranks = np.zeros_like(S1_estimates)\n",
    "ST_ranks = np.zeros_like(ST_estimates)\n",
    "for i in range(len(nsamples)):\n",
    "    orderS1 = np.argsort(S1_estimates[:,i])\n",
    "    orderST = np.argsort(ST_estimates[:,i])\n",
    "    S1_ranks[:,i] = orderS1.argsort()\n",
    "    ST_ranks[:,i] = orderST.argsort()\n",
    "    \n",
    "# Generate figure showing evolution of ranks\n",
    "fig =  plt.figure(figsize=(18,9))\n",
    "ax1 = fig.add_subplot(1,2,1)\n",
    "handles = []\n",
    "for j in range(problem['num_vars']):\n",
    "    handles += ax1.plot(nsamples, S1_ranks[j,:], linewidth=3)\n",
    "ax1.set_title('Parameter ranking based on S1', fontsize=20)\n",
    "ax1.set_ylabel('S1', fontsize=18)\n",
    "ax1.set_xlabel('Number of samples (n)', fontsize=18)\n",
    "ax1.set_yticklabels(np.arange(problem['num_vars']+1, 0, -1))\n",
    "ax1.tick_params(axis='both', which='major', labelsize=14)\n",
    "ax2 = fig.add_subplot(1,2,2)\n",
    "for j in range(problem['num_vars']):\n",
    "    ax2.plot(nsamples, ST_ranks[j,:], linewidth=3)\n",
    "ax2.set_title('Parameter ranking based on ST', fontsize=20)\n",
    "ax2.set_ylabel('ST', fontsize=18)\n",
    "ax2.set_yticklabels(np.arange(problem['num_vars']+1, 0, -1))\n",
    "ax2.tick_params(axis='both', which='major', labelsize=14)\n",
    "ax2.set_xlabel('Number of samples (n)', fontsize=18)\n",
    "fig.legend(handles, problem['names'], loc = 'right', fontsize=14)\n",
    "#plt.show()\n",
    "#plt.savefig('rankingevolution.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "-1Lblvnvff-9"
   },
   "source": [
    "## Radial plot for SA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "5Oa37UTOff-9"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import itertools\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import math\n",
    "from numpy import genfromtxt\n",
    "import matplotlib.patches as mpatches\n",
    "import matplotlib.pyplot as plt\n",
    "sns.set_style('whitegrid', {'axes_linewidth': 0, 'axes.edgecolor': 'white'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "lCM69eEq-Qh1"
   },
   "source": [
    "## Plot function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "B6J2tyLAff-_"
   },
   "outputs": [],
   "source": [
    "def is_significant(value, confidence_interval, threshold=\"conf\"):\n",
    "    if threshold == \"conf\":\n",
    "        return value - abs(confidence_interval) > 0\n",
    "    else:\n",
    "        return value - abs(float(threshold)) > 0\n",
    " \n",
    "def grouped_radial(SAresults, parameters, radSc=2.0, scaling=1, widthSc=0.5, STthick=1, varNameMult=1.3, colors=None, groups=None, gpNameMult=1.5, threshold=\"conf\"):\n",
    "    # Derived from https://github.com/calvinwhealton/SensitivityAnalysisPlots\n",
    "    fig, ax = plt.subplots(1, 1)\n",
    "    color_map = {}\n",
    "     \n",
    "    # initialize parameters and colors\n",
    "    if groups is None:\n",
    "         \n",
    "        if colors is None:\n",
    "            colors = [\"k\"]\n",
    "         \n",
    "        for i, parameter in enumerate(parameters):\n",
    "            color_map[parameter] = colors[i % len(colors)]\n",
    "    else:        \n",
    "        if colors is None:\n",
    "            colors = sns.color_palette(\"deep\", max(3, len(groups)))\n",
    "         \n",
    "        for i, key in enumerate(groups.keys()):\n",
    "            #parameters.extend(groups[key])\n",
    "             \n",
    "            for parameter in groups[key]:\n",
    "                color_map[parameter] = colors[i % len(colors)]\n",
    "     \n",
    "    n = len(parameters)\n",
    "    angles = radSc*math.pi*np.arange(0, n)/n\n",
    "    x = radSc*np.cos(angles)\n",
    "    y = radSc*np.sin(angles)\n",
    "     \n",
    "    # plot second-order indices\n",
    "    for i, j in itertools.combinations(range(n), 2):\n",
    "        #key1 = parameters[i]\n",
    "        #key2 = parameters[j]\n",
    "         \n",
    "        if is_significant(SAresults[\"S2\"][i][j], SAresults[\"S2_conf\"][i][j], threshold):\n",
    "            angle = math.atan((y[j]-y[i])/(x[j]-x[i]))\n",
    "                 \n",
    "            if y[j]-y[i] < 0:\n",
    "                angle += math.pi\n",
    "                 \n",
    "            line_hw = scaling*(max(0, SAresults[\"S2\"][i][j])**widthSc)/2\n",
    "                 \n",
    "            coords = np.empty((4, 2))\n",
    "            coords[0, 0] = x[i] - line_hw*math.sin(angle)\n",
    "            coords[1, 0] = x[i] + line_hw*math.sin(angle)\n",
    "            coords[2, 0] = x[j] + line_hw*math.sin(angle)\n",
    "            coords[3, 0] = x[j] - line_hw*math.sin(angle)\n",
    "            coords[0, 1] = y[i] + line_hw*math.cos(angle)\n",
    "            coords[1, 1] = y[i] - line_hw*math.cos(angle)\n",
    "            coords[2, 1] = y[j] - line_hw*math.cos(angle)\n",
    "            coords[3, 1] = y[j] + line_hw*math.cos(angle)\n",
    " \n",
    "            ax.add_artist(plt.Polygon(coords, color=\"0.75\"))\n",
    "         \n",
    "    # plot total order indices\n",
    "    for i, key in enumerate(parameters):\n",
    "        if is_significant(SAresults[\"ST\"][i], SAresults[\"ST_conf\"][i], threshold):\n",
    "            ax.add_artist(plt.Circle((x[i], y[i]), scaling*(SAresults[\"ST\"][i]**widthSc)/2, color='w'))\n",
    "            ax.add_artist(plt.Circle((x[i], y[i]), scaling*(SAresults[\"ST\"][i]**widthSc)/2, lw=STthick, color='0.4', fill=False))\n",
    "     \n",
    "    # plot first-order indices\n",
    "    for i, key in enumerate(parameters):\n",
    "        if is_significant(SAresults[\"S1\"][i], SAresults[\"S1_conf\"][i], threshold):\n",
    "            ax.add_artist(plt.Circle((x[i], y[i]), scaling*(SAresults[\"S1\"][i]**widthSc)/2, color='0.4'))\n",
    "            \n",
    "    # add labels\n",
    "    for i, key in enumerate(parameters):                \n",
    "        ax.text(varNameMult*x[i], varNameMult*y[i], key, ha='center', va='center',\n",
    "                rotation=angles[i]*360/(2*math.pi) - 90,\n",
    "                color=color_map[key])\n",
    "         \n",
    "    if groups is not None:\n",
    "        for i, group in enumerate(groups.keys()):\n",
    "            print(group)\n",
    "            group_angle = np.mean([angles[j] for j in range(n) if parameters[j] in groups[group]])\n",
    "             \n",
    "            ax.text(gpNameMult*radSc*math.cos(group_angle), gpNameMult*radSc*math.sin(group_angle), group, ha='center', va='center',\n",
    "                rotation=group_angle*360/(2*math.pi) - 90,\n",
    "                color=colors[i % len(colors)])\n",
    "             \n",
    "    ax.set_facecolor('white')\n",
    "    ax.set_xticks([])\n",
    "    ax.set_yticks([])\n",
    "#     ax.\n",
    "    plt.axis('equal')\n",
    "    plt.axis([-2*radSc, 2*radSc, -2*radSc, 2*radSc])\n",
    "    #plt.show()\n",
    " \n",
    "     \n",
    "    return fig"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "mkmi-0-vff_C"
   },
   "source": [
    "## Range of soil parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "EbKdSoihff_C"
   },
   "outputs": [],
   "source": [
    "problem = {\n",
    "    'num_vars':6,\n",
    "    'names':['SOC', 'Sand', 'Clay', 'pH', 'CN', 'BD'],\n",
    "    'bounds':[[2.58, 6.20],\n",
    "              [0.01, 0.30],\n",
    "              [0.01, 0.30],\n",
    "              [4.6, 6.9],\n",
    "              [10.9, 12.4],\n",
    "              [900, 1350]]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "OOz0S62Off_F"
   },
   "outputs": [],
   "source": [
    "#names for csv files\n",
    "list_of_csv=['soybean-000-2015.csv', 'sugar-beet-2011.csv', 'sugar-beet-2017.csv',\n",
    "'spring-barley-2012.csv', 'sugar-beet-2014.csv']\n",
    "list_of_names=['soybean-000-2015', 'sugar-beat-2011', 'sugar-beat-2017',\n",
    "'spring-barley-2012', 'sugar-beat-2014']\n",
    "list_of_totals=['total_SI_'+x for x in list_of_names]\n",
    "list_of_first=['fisrt_SI_'+x for x in list_of_names]\n",
    "list_of_second=['second_SI_'+x for x in list_of_names]\n",
    "list_of_SI=['SI_'+x for x in list_of_names]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 111
    },
    "colab_type": "code",
    "id": "gz3hUbeuff_I",
    "outputId": "5917f0ad-5554-4761-8bdc-80d47d9d59b8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "soybean-000-2015.csv\n",
      "sugar-beet-2011.csv\n",
      "sugar-beet-2017.csv\n",
      "spring-barley-2012.csv\n",
      "sugar-beet-2014.csv\n"
     ]
    }
   ],
   "source": [
    "for j, i in enumerate(list_of_csv):\n",
    "        all_data_csv = genfromtxt('./data/monica_sensitivity/'+str(i), delimiter=',')\n",
    "        output = all_data_csv[:,2]\n",
    "        print(i)\n",
    "        list_of_SI[j] = sobol.analyze(problem, output, calc_second_order=True, conf_level=0.95, print_to_console=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 327
    },
    "colab_type": "code",
    "id": "-d6zTAAxff_K",
    "outputId": "f835a88e-d5a3-46ea-bf94-fe587b2a6981"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Soil physics\n",
      "Soil chemistry\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "groups={\"Soil physics\" : [\"Sand\", \"Clay\", \"BD\"],\n",
    "        \"Soil chemistry\" : [\"pH\", \"SOC\", \"CN\"]}\n",
    " \n",
    "fig = grouped_radial(list_of_SI[4], ['BD', 'Sand', 'Clay', 'pH', 'CN', 'SOC'], groups=groups, threshold=0.001)\n",
    "red_patch = mpatches.Patch(color='red', label='The red data')\n",
    "plt.title(list_of_names[4], loc='left')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "uz8LoKAWff_Q"
   },
   "source": [
    "# Homework\n",
    "____\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "NRf5Tc2eff_R"
   },
   "source": [
    "\n",
    "## Resources\n",
    "\n",
    "__SA and UQ__\n",
    "\n",
    "\n",
    "1) [Rhodium project](https://github.com/Project-Platypus/Rhodium.git)\n",
    "\n",
    "\n",
    "2) [SALib](https://github.com/SALib/SALib)\n",
    "\n",
    "__Model__\n",
    "\n",
    "3) [PCSE](https://pcse.readthedocs.io/en/stable/index.html)\n",
    "\n",
    "4) How to install PCSE at local machine\n",
    "    `conda env create -f` [py3_pcse.yml](https://github.com/mishagrol/Seminar_Sobol/blob/master/py3_pcse.yml)\n",
    "\n",
    "Аny questions  -  \n",
    "\n",
    "Telegram - `@misha_grol`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pcse.fileinput import YAMLCropDataProvider\n",
    "cropd = YAMLCropDataProvider()\n",
    "# cropd.print_crops_varieties()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 1 – Crop Yield Prediction (PCSE/WOFOST)  \n",
    "\n",
    "You can use the seminars’ notebook:  \n",
    "\n",
    "* Assess the yield of one of the crops for the Moscow region over several years (potatoes, sugar beets)   \n",
    "Crop  - [List of crops](https://github.com/ajwdewit/WOFOST_crop_parameters) \\\n",
    "Weather  - [NASAdataprovider in PCSE](https://pcse.readthedocs.io/en/stable/code.html?highlight=NASA#pcse.db.NASAPowerWeatherDataProvider)   \n",
    "* Generate plots for biomass growth, crop yield and seasonal weather for one year\n",
    "\n",
    "\n",
    "### Agromanagement for Moscow Potato crop and NPK fertilization\n",
    "```\n",
    "\"name\": \"Potato\", \n",
    "\"latitude\": 54.85, \n",
    "\"longitude\": 38.85, \n",
    "\"crop_start\": \"2019-04-20\", \n",
    "\"crop_end\": \"2019-09-15\", \n",
    "\"crop_name\": \"potato\", \n",
    "\"npk_events\": [\"2019-06-22\"], \n",
    "\"npk\": [[90,10,90]]\n",
    "```\n",
    "\n",
    "### Available crop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pcse.fileinput import YAMLCropDataProvider\n",
    "cropd = YAMLCropDataProvider()\n",
    "# cropd.print_crops_varieties()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "3QNf2VAmVrCB"
   },
   "source": [
    "## Part 2 – Sensitivity Analysis (SALib)  \n",
    "\n",
    "1) Perform sensitivity analysis of one of the model blocks (crop, soil, agromanagement *) with SALib. You can choose one of the methods that you consider necessary (Sobol, FAST, ...). \n",
    "\n",
    "    Generate samples – In report provide the size of the resulting matrix and the sample size (N) \n",
    "\n",
    "    Conduct parameter sensitivity analysis  - In report provide S1 and ST indices.  \n",
    "\n",
    "2) Generate plots (Hist, etc.)  \n",
    "\n",
    "*3) Speed-up sensitivity analysis with multiprocessing \n",
    "\n",
    "*4) Estimate the required number of simulations to obtain reliable values of the sensitivity indices. Try to estimate the sample size at the confidence interval of the sensitivity indices. \n",
    "\n",
    "* Please note that working with discrete data can cause certain difficulties. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "YlZ0GfwNff_U"
   },
   "source": [
    "### bonus"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "0NsBD3_lff_V"
   },
   "source": [
    "__Morris method__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Y7dLOiBiff_V"
   },
   "source": [
    "\n",
    "Generate a sample using the Method of Morris\n",
    "\n",
    "Three variants of Morris' sampling for elementary effects is supported:\n",
    "\n",
    "- Vanilla Morris\n",
    "- Optimised trajectories when ``optimal_trajectories=True`` (using\n",
    "    Campolongo's enhancements from 2007 and optionally Ruano's enhancement\n",
    "    from 2012; ``local_optimization=True``)\n",
    "- Groups with optimised trajectories when ``optimal_trajectories=True`` and\n",
    "    the problem definition specifies groups (note that ``local_optimization``\n",
    "    must be ``False``)\n",
    "\n",
    "At present, optimised trajectories is implemented using either a brute-force\n",
    "approach, which can be very slow, especially if you require more than four\n",
    "trajectories, or a local method based which is much faster. Both methods now\n",
    "implement working with groups of factors.\n",
    "\n",
    "Note that the number of factors makes little difference,\n",
    "but the ratio between number of optimal trajectories and the sample size\n",
    "results in an exponentially increasing number of scores that must be\n",
    "computed to find the optimal combination of trajectories.  We suggest going\n",
    "no higher than 4 from a pool of 100 samples with the brute force approach.\n",
    "With local_optimization = True (which is default),\n",
    "it is possible to go higher than the previously suggested 4 from 100.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ROldqQcHff_W"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parameter                         Mu_Star         Mu    Mu_Star_Conf      Sigma\n",
      "x1                                  2.564     -0.039           0.052      2.690\n",
      "x2                                  1.431     -0.047           0.044      1.619\n",
      "x3                                  0.603     -0.003           0.024      0.704\n",
      "x4                                  0.330      0.004           0.013      0.388\n",
      "x5                                  0.034     -0.000           0.001      0.040\n",
      "x6                                  0.034      0.000           0.001      0.040\n",
      "x7                                  0.033     -0.001           0.001      0.040\n",
      "x8                                  0.034      0.003           0.001      0.039\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import sys\n",
    "\n",
    "from SALib.analyze import morris\n",
    "from SALib.sample.morris import sample\n",
    "from SALib.test_functions import Sobol_G\n",
    "from SALib.util import read_param_file\n",
    "from SALib.plotting.morris import horizontal_bar_plot, covariance_plot, \\\n",
    "    sample_histograms\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "#sys.path.append('../..')\n",
    "\n",
    "# Read the parameter range file and generate samples\n",
    "#problem = read_param_file('/Users/mikhailgasanov/Documents/GIT/SALib/src/SALib/test_functions/params/Sobol_G.txt')\n",
    "# or define manually without a parameter file:\n",
    "problem = {\n",
    " 'num_vars': 8,\n",
    " 'names': ['x1', 'x2', 'x3', 'x4', 'x5', 'x6', 'x7', 'x8'],\n",
    " 'groups': None,\n",
    " 'bounds': [[0.0, 1.0],\n",
    "            [0.0, 1.0],\n",
    "            [0.0, 1.0],\n",
    "            [0.0, 1.0],\n",
    "            [0.0, 1.0],\n",
    "            [0.0, 1.0],\n",
    "            [0.0, 1.0],\n",
    "            [0.0, 1.0]]\n",
    "}\n",
    "# Files with a 4th column for \"group name\" will be detected automatically, e.g.\n",
    "# param_file = '../../src/SALib/test_functions/params/Ishigami_groups.txt'\n",
    "\n",
    "# Generate samples\n",
    "param_values = sample(problem, N=1000, num_levels=4,\n",
    "                      optimal_trajectories=None)\n",
    "\n",
    "# To use optimized trajectories (brute force method),\n",
    "# give an integer value for optimal_trajectories\n",
    "\n",
    "# Run the \"model\" -- this will happen offline for external models\n",
    "Y = Sobol_G.evaluate(param_values)\n",
    "\n",
    "# Perform the sensitivity analysis using the model output\n",
    "# Specify which column of the output file to analyze (zero-indexed)\n",
    "Si = morris.analyze(problem, param_values, Y, conf_level=0.95,\n",
    "                    print_to_console=True,\n",
    "                    num_levels=4, num_resamples=100)\n",
    "# Returns a dictionary with keys 'mu', 'mu_star', 'sigma', and 'mu_star_conf'\n",
    "# e.g. Si['mu_star'] contains the mu* value for each parameter, in the\n",
    "# same order as the parameter file\n",
    "\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2)\n",
    "horizontal_bar_plot(ax1, Si, {}, sortby='mu_star', unit=r\"tCO$_2$/year\")\n",
    "covariance_plot(ax2, Si, {}, unit=r\"tCO$_2$/year\")\n",
    "\n",
    "fig2 = plt.figure()\n",
    "sample_histograms(fig2, param_values, problem, {'color': 'y'})\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "8-FizFxrff_Y"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [
    "KWTtMUZsff84"
   ],
   "include_colab_link": true,
   "name": "Seminar_Sensitivity_analysis.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}