Automatic parameter estimation for GR4J.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Automatic parameter estimation for GR4J. Uses scipy optimize packages.\n",
    "\n",
    "Following exemples found in https://stackoverflow.com/questions/19664865/migrating-from-pulp-to-scipy\n",
    "\n",
    "Preparing the functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import scipy.optimize as opt\n",
    "\n",
    "# GR4J lives in a directory above this code. Not sure this is the best practice...\n",
    "import os, sys\n",
    "sys.path.append(os.path.relpath(\"../gr4j\"))\n",
    "from gr4j import gr4j\n",
    "\n",
    "# discharge conversion functions\n",
    "def m3_to_mm(q, area):\n",
    "    '''Convert m3/s discharge data to mm/day\n",
    "    q: list of daily discharge records\n",
    "    area: basin area (km2)\n",
    "    return: list of converted discharge records\n",
    "    '''\n",
    "    return [i * 86.4 / area for i in q]\n",
    "\n",
    "## parameter transformation functions\n",
    "# recieve a dictionary of parameters\n",
    "# returns a dictionary of transformed (real) parameter\n",
    "\n",
    "def real2transf(params):\n",
    "    '''Convert real parameter to transformed'''\n",
    "    x1 = np.log(params['X1'])\n",
    "    x2 = np.arcsinh(params['X2'])\n",
    "    x3 = np.log(params['X3'])\n",
    "    x4 = np.log(params['X4']-0.5)\n",
    "    return {'X1': x1,\n",
    "            'X2': x2,\n",
    "            'X3': x3,\n",
    "            'X4': x4\n",
    "           }\n",
    "\n",
    "def transf2real(params):\n",
    "    '''Convert transformed parameters to real'''\n",
    "    x1 = np.exp(params['X1'])\n",
    "    x2 = np.sinh(params['X2'])\n",
    "    x3 = np.exp(params['X3'])\n",
    "    x4 = np.exp(params['X4'])+0.5\n",
    "    return {'X1': x1,\n",
    "            'X2': x2,\n",
    "            'X3': x3,\n",
    "            'X4': x4\n",
    "           }\n",
    "\n",
    "\n",
    "# Objective function (Nash-Sutcliffe)\n",
    "def nash(qobs, qsim):\n",
    "    top = sum((qsim - qobs)**2)\n",
    "    bottom = sum((qobs.mean() - qobs)**2)\n",
    "    return 1 - top/bottom\n",
    "\n",
    "# function to be minimized, using real parameter values\n",
    "def fun(guess, prec, etp, qobs):\n",
    "    params = {'X1': guess[0],\n",
    "              'X2': guess[1],\n",
    "              'X3': guess[2],\n",
    "              'X4': guess[3]\n",
    "             }\n",
    "       \n",
    "    states = {'production_store': 0.60 * params['X1'],\n",
    "              'routing_store': 0.70 * params['X3']\n",
    "             }\n",
    "\n",
    "    qsim = gr4j(prec, etp, params, states)\n",
    "    n = nash(qobs, qsim)\n",
    "    return -n\n",
    "\n",
    "# function to be minimized, using transformed parameters values\n",
    "def fun_transf(guess, prec, etp, qobs):\n",
    "    transf_params = {'X1': guess[0],\n",
    "              'X2': guess[1],\n",
    "              'X3': guess[2],\n",
    "              'X4': guess[3]\n",
    "             }\n",
    "    \n",
    "    params = transf2real(transf_params)\n",
    "    \n",
    "    states = {'production_store': 0.60 * params['X1'],\n",
    "              'routing_store': 0.70 * params['X3']\n",
    "             }\n",
    "\n",
    "    qsim = gr4j(prec, etp, params, states)\n",
    "    n = nash(qobs, qsim)\n",
    "    return -n\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Testing the automatic parameter estimation using the dataset that comes with the GR4J_en.xlsx model spreadsheet. This data is from _Le Léguer à Belle-Isle-en-Terre_ watershed. The original model was obtained from https://webgr.irstea.fr/en/modeles/journalier-gr4j-2/fonctionnement_gr4j/.\n",
    "\n",
    "A more detailed description of the dataset can be found on the dataset directory of this repository.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nash coefficient from the spreadsheet: 0.919\n",
      "Nash coefficient obtained using transformed parameter optimization: 0.528\n",
      "Nash coefficient obtained using real parameter optimization: 0.764\n"
     ]
    }
   ],
   "source": [
    "# Reading sample data that comes with the original model - french basin\n",
    "test_data = pd.read_csv('datasets/gr4j_test_data.csv')\n",
    "test_data['Date'] = pd.to_datetime(test_data['Date'])\n",
    "test_data = test_data.set_index('Date')\n",
    "\n",
    "# Initial guess of TRANSFORMED model parameters\n",
    "# done as a list since it's what the optimization function expects.\n",
    "# GR4J function recieves a dictionary of REAL model parameters\n",
    "# functon takes care of parameter convertion\n",
    "# guess based on median parameter value reported by Perrin et al (2003)\n",
    "# in real and transformed parameter space\n",
    "guess = [350, 0, 90, 1.7]\n",
    "guess_transf = [5.9, 0, 4.5, 0.18] \n",
    "\n",
    "# Calibrating model with the first year of data\n",
    "prec_cal = test_data['Precip (mm)'].loc[:'1991-12-31']\n",
    "etp_cal = test_data['ETP (mm)'].loc[:'1991-12-31']\n",
    "qobs_cal = test_data['Qobs (mm/day)'].loc[:'1991-12-31']\n",
    "qsim_orig_cal = test_data['Qsim (mm/day)'].loc[:'1991-12-31']\n",
    "\n",
    "# bounds based on 80% parameter estimates (from Perrin et al, 2003)\n",
    "# in real and transformed space\n",
    "bnds = ((100, 1200),\n",
    "        (-5, 3),\n",
    "        (20, 300),\n",
    "        (1.1, 2.9)\n",
    "       )\n",
    "\n",
    "bnds_transf = ((4.5, 7),\n",
    "               (-2.31, 1.81),\n",
    "               (3, 5.7),\n",
    "               (-0.51, 0.87)\n",
    "              )\n",
    "\n",
    "# running optimization using transformed or real parameter values\n",
    "result = opt.minimize(fun, guess, method='SLSQP',\n",
    "                      args=(prec_cal, etp_cal, qobs_cal), bounds=bnds)\n",
    "\n",
    "result_transf = opt.minimize(fun_transf, guess_transf, method='SLSQP',\n",
    "                             args=(prec_cal, etp_cal, qobs_cal), bounds=bnds_transf)\n",
    "\n",
    "#result = opt.minimize(fun, guess, method='Nelder-Mead', args=(prec_cal, etp_cal, qobs_cal))\n",
    "\n",
    "# Simulating using the optimized parameters (transformed values)\n",
    "t_params = {'X1': result_transf['x'][0],\n",
    "          'X2': result_transf['x'][1],\n",
    "          'X3': result_transf['x'][2],\n",
    "          'X4': result_transf['x'][3]\n",
    "         }\n",
    "\n",
    "params_t = transf2real(t_params)\n",
    "states_t = {'production_store': 0.60 * params_t['X1'],\n",
    "          'routing_store': 0.70 * params_t['X3']\n",
    "         }\n",
    "\n",
    "# Running for the entire period but verifying for the seccond year\n",
    "prec = test_data['Precip (mm)']\n",
    "etp = test_data['ETP (mm)']\n",
    "qobs = test_data['Qobs (mm/day)']\n",
    "qsim_orig = test_data['Qsim (mm/day)']\n",
    "\n",
    "qsim_t = gr4j(prec, etp, params_t, states_t)\n",
    "qsim_t = pd.DataFrame(qsim_t, columns = ['qsim_t_mod'], index = pd.date_range('1991-1-1', '1992-12-30'))\n",
    "\n",
    "# Simulating using the optimized parameters (real values)\n",
    "params_r = {'X1': result['x'][0],\n",
    "          'X2': result['x'][1],\n",
    "          'X3': result['x'][2],\n",
    "          'X4': result['x'][3]\n",
    "         }\n",
    "\n",
    "states_r = {'production_store': 0.60 * params_r['X1'],\n",
    "          'routing_store': 0.70 * params_r['X3']\n",
    "         }\n",
    "\n",
    "qsim_r = gr4j(prec, etp, params_r, states_r)\n",
    "qsim_r = pd.DataFrame(qsim_r, columns = ['qsim_r_mod'], index = pd.date_range('1991-1-1', '1992-12-30'))\n",
    "\n",
    "data = pd.concat([test_data, qsim_t, qsim_r], axis=1)\n",
    "\n",
    "# checking the adjustment (Nash-Sutcliffe)\n",
    "print(\"Nash coefficient from the spreadsheet: 0.919\")\n",
    "print(\"Nash coefficient obtained using transformed parameter optimization: {:.3f}\".format(nash(\n",
    "    data['Qobs (mm/day)'].loc['1992-1-1':],\n",
    "    data['qsim_t_mod'].loc['1992-1-1':])))\n",
    "print(\"Nash coefficient obtained using real parameter optimization: {:.3f}\".format(nash(\n",
    "    data['Qobs (mm/day)'].loc['1992-1-1':],\n",
    "    data['qsim_r_mod'].loc['1992-1-1':])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model parameters in the spreadsheet were:\n",
      "X1: 320.107; X2: 2.421; X3: 69.628; X4: 1.389\n",
      "Model parameters obtained here (transformed optimization) are:\n",
      "X1: 1096.633; X2: 1.435; X3: 82.399; X4: 1.100\n",
      "Model parameters obtained here (real optimization) are:\n",
      "X1: 379.965; X2: 1.714; X3: 138.283; X4: 1.100\n"
     ]
    }
   ],
   "source": [
    "print('Model parameters in the spreadsheet were:')\n",
    "print('X1: 320.107; X2: 2.421; X3: 69.628; X4: 1.389')\n",
    "print('Model parameters obtained here (transformed optimization) are:')\n",
    "print('X1: {:.3f}; X2: {:.3f}; X3: {:.3f}; X4: {:.3f}'\n",
    "     .format(params_t['X1'], params_t['X2'],\n",
    "             params_t['X3'], params_t['X4']))\n",
    "print('Model parameters obtained here (real optimization) are:')\n",
    "print('X1: {:.3f}; X2: {:.3f}; X3: {:.3f}; X4: {:.3f}'\n",
    "     .format(params_r['X1'], params_r['X2'],\n",
    "             params_r['X3'], params_r['X4']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running the model with the original parameters\n",
      "Nash coefficient equals to the one from the spreadsheet\n",
      "Nash: 0.919\n"
     ]
    }
   ],
   "source": [
    "# running the model with the parameters from the excell spreadsheet exemple\n",
    "params_ex = {'X1': 320.107,\n",
    "          'X2': 2.421,\n",
    "          'X3': 69.628,\n",
    "          'X4': 1.389\n",
    "         }\n",
    "\n",
    "states_ex = {'production_store': 0.60 * params['X1'],\n",
    "          'routing_store': 0.70 * params['X3']\n",
    "         }\n",
    "\n",
    "qsim_ex = gr4j(prec, etp, params_ex, states_ex)\n",
    "\n",
    "print('Running the model with the original parameters')\n",
    "print('Nash coefficient equals to the one from the spreadsheet')\n",
    "print('Nash: {:.3f}'.format(nash(\n",
    "    data['Qobs (mm/day)'].loc['1992-1-1':],\n",
    "    qsim_ex[365:])))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What we've seen is that the python optimization can find some good parameters for the model. However, parameters obtained using the solver routines from excell yielded better results. Some things to evaluate:\n",
    "1. What optimization algorithm does Excell solver utilizes?\n",
    "  * For nonlinear optimization: Generalized Reduced Gradient (GRG2)\n",
    "  * Linear programming: Simplex and dual Simplex from Simplex LP\n",
    "  * non-smooth optimization: various genetic algorithms\n",
    "  * source: https://www.solver.com/excel-solver-algorithms-and-methods-used\n",
    "2. The excell spreadsheed optimizes a transformed coefficient. Check if that changes something\n",
    "  * The documentation in the R package about adding other optimization methods mentions that we should used transformed parameters, to ensure homogenity of displacement in different dimensions of the parameter space\n",
    "  * https://cran.r-project.org/web/packages/airGR/vignettes/V02.1_param_optim.html\n",
    "3. Change options in the python solver\n",
    "4. Deppending on the solver and on the initial guess, the optimizatin does not work and I get a \"IndexError: list index out of range\". Ex.: guess = [400, 3, 50, 2]; solver = SLSQP\n",
    "\n",
    "Some observations with solver = SLSQP:\n",
    "\n",
    "| Guess parameters   | Nash |\n",
    "|--------------------|------|\n",
    "| [303, 3.2, 65, 1]  | 0.768|\n",
    "| [303, 3.2, 50, 0.5]| 0.728| \n",
    "|[320, 2.4, 70, 1.4] | 0.766|\n",
    "|[320.11, 2.42, 69.63, 1.39]| 0.766|\n",
    "\n",
    "Even using the initial guess very close to the results from the Excell spreadsheet, I cannot get a Nash coefficient close to 0.92! Need to investigate solver options a bit more.\n",
    "\n",
    "Anyway, plotting some figures:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsMAAAGfCAYAAAC6KN9TAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSCQSiUQikeRapDEskUgkEolEIsm1/B+WlCTrenNUGgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fca11f920f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,7))\n",
    "\n",
    "plt.plot(data['qsim_r_mod'], 'b--', label = 'Simulated (real)')\n",
    "plt.plot(data['qsim_t_mod'], 'g--', label = 'Simulated (transf)')\n",
    "plt.plot(data['Qobs (mm/day)'], 'k', label = 'Observed discharge')\n",
    "plt.plot(data['Qsim (mm/day)'], 'r--', label = 'Sim. original')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'X1': 7.0900768357760917,\n",
       " 'X2': 1.8184464592320668,\n",
       " 'X3': 5.7037824746562009,\n",
       " 'X4': 0.87546873735389985}"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "params = {'X1': 1200,\n",
    "          'X2': 3,\n",
    "          'X3': 300,\n",
    "          'X4': 2.9\n",
    "         }\n",
    "real2transf(params)"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}