main.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Preparação dos Dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "from time import time\n",
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import pearsonr\n",
    "import h5py\n",
    "\n",
    "# Módulo do software da implementação do MOD17\n",
    "import mod17\n",
    "from mod17 import MOD17, PFT_VALID\n",
    "from mod17.utils import restore_bplut, pft_remap\n",
    "\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_absolute_error, mean_squared_error\n",
    "from statsmodels.tsa.api import ExponentialSmoothing, SimpleExpSmoothing, Holt\n",
    "from pymoo.core.problem import ElementwiseProblem\n",
    "from pymoo.algorithms.soo.nonconvex.ga import GA\n",
    "from pymoo.optimize import minimize\n",
    "from pymoo.constraints.as_penalty import ConstraintsAsPenalty\n",
    "from pymoo.operators.mutation.pm import PolynomialMutation\n",
    "from pymoo.operators.crossover.sbx import SBX\n",
    "from pymoo.util.display.column import Column\n",
    "from pymoo.util.display.output import Output\n",
    "from pymoo.core.callback import Callback\n",
    "from IPython.display import display, HTML\n",
    "from typing import List, Union\n",
    "\n",
    "def Reco_Horario(serie_reco:pd.Series) -> float:   \n",
    "    return serie_reco.between_time('20:00', '06:00').sum()\n",
    "\n",
    "def Gpp_Horario(serie_gpp: pd.Series) -> float:   \n",
    "    return serie_gpp.between_time('07:00', '19:00').sum()\n",
    "\n",
    "def create_flux_data(dataframe: pd.DataFrame, var: List[Union[str, str]],kg_c = True) -> pd.DataFrame:\n",
    "    \"\"\"\n",
    "    Cria um DataFrame com os dados de fluxo de carbono de acordo com a variável alvo especificada.\n",
    "    \n",
    "    Args:\n",
    "        data_df (pd.DataFrame): DataFrame que conterá os dados de fluxo de carbono.\n",
    "        dataframe (pd.DataFrame): DataFrame contendo os dados originais.\n",
    "        var (List[str, str]): Lista contendo o nome da variável alvo e sua respectiva coluna nos dados originais.\n",
    "        \n",
    "    Returns:\n",
    "        pd.DataFrame: DataFrame contendo os dados de fluxo de carbono atualizados.\n",
    "    \"\"\"\n",
    "    # Mapeamento das funções para calcular as variáveis alvo\n",
    "    var_names = {\n",
    "        'gpp': Gpp_Horario,\n",
    "        'reco': Reco_Horario\n",
    "        # Adicione aqui outras variáveis alvo, se necessário\n",
    "    }\n",
    "    \n",
    "    data_df = pd.DataFrame(index=np.unique(dataframe.index.date)) # index = datas do dataframe (DADOS NÃO SÃO PUXADOS)\n",
    "    \n",
    "    data_df[f'{var[0]}'] = pd.Series(dtype='float') # Cria uma coluna com o nome da variável alvo\n",
    "    \n",
    "    for date in data_df.index:\n",
    "        \"\"\"\n",
    "        Loop através de cada data para calcular os valores desejados.\n",
    "        \"\"\"\n",
    "        # Obtendo a data no formato YYYY/MM/DD\n",
    "        \n",
    "        date_str = date.strftime(\"%Y-%m-%d\")\n",
    "        \n",
    "        # Obtendo o valor da variável alvo do dataframe original\n",
    "        var_alvo = dataframe[var[1]].loc[date_str]\n",
    "        \n",
    "        # Chamando a função apropriada para calcular a variável alvo\n",
    "        var_alvo = var_names.get(var[0].lower())(var_alvo)\n",
    "        \n",
    "        # Atualizando o DataFrame com o valor calculado\n",
    "        data_df.loc[date, var[0]] = var_alvo\n",
    "    \n",
    "    # Aplicando transformação para kgC\n",
    "    if kg_c:\n",
    "        data_df[var[0]] = data_df[var[0]].apply(to_kgC)\n",
    "    data_df.index = data_df.reset_index()['index'].apply(lambda x:pd.to_datetime(x))\n",
    "    \n",
    "    return data_df\n",
    "\n",
    "def to_kgC(x):\n",
    "    \"\"\"Retorna valores convertidos para kg C m² s\"\"\"\n",
    "    umol2mol = 1e-06\n",
    "    Cmol = 0.012011\n",
    "    minutesinseconds = 1800\n",
    "    #days2seconds = 86400\n",
    "    kg2g = 1000 # Precisamos em kg, não em g, por isso não multipliquei por 1000\n",
    "    \n",
    "    return x * umol2mol * Cmol * minutesinseconds # cálculo dado pelo professor\n",
    "\n",
    "\n",
    "\n",
    "# ======= Preparação dos dados originais da torre de fluxo do Peru =================================================================\n",
    "# am_peru_h = pd.read_csv('amf_pe/AMF_PE-QFR_FLUXNET_FULLSET_HH_2018-2019_3-5.csv',na_values=-9999) # Dados da torre pro mesmo dia\n",
    "# am_peru_h.index = am_peru_h['TIMESTAMP_START'].apply(lambda x:datetime.strptime(str(x), '%Y%m%d%H%M%S'))\n",
    "# gpp_peru_h = am_peru_h['GPP_DT_VUT_REF']['2018':'2019'].copy() # Gpp da torre em Hora\n",
    "# gpp_peru_h.to_csv('AMF_PE-QFR_FLUXNET_FULLSET_HH_2018-2019_3-5.csv') # Transformar para csv apenas com o GPP a fim de reduzir o tamanho original do arquivo\n",
    "\n",
    "\n",
    "# ======= Preparação dos dados originais da torre de fluxo de Santarem =================================================================\n",
    "# am_santarem_h = pd.read_csv('./santarem_torre/ameriflux-santarem-dados-original.csv',na_values=-9999.0) # Dados da AMERIFLUX\n",
    "# am_santarem_h.loc[0, 'Time'] = '00:00:00'\n",
    "# am_santarem_h.index = pd.to_datetime(am_santarem_h['Date'] + ' ' + am_santarem_h['Time'], format='%Y-%m-%d %H:%M:%S')\n",
    "# am_santarem_h[['Reco_uStar', 'GPP_uStar_f','NEE_uStar_f']]\n",
    "# am_santarem_h.to_csv('./santarem_torre/ameriflux-santarem-dados-original.csv') # Transformar para csv apenas com o GPP a fim de reduzir o tamanho original do arquivo\n",
    "\n",
    "\n",
    "# Carregando os dados de GPP da torre de fluxo do Peru\n",
    "gpp_peru_h = pd.read_csv('amf_pe/AMF_PE-QFR_FLUXNET_FULLSET_HH_2018-2019_3-5.csv')\n",
    "gpp_peru_h.index = gpp_peru_h['TIMESTAMP_START'].apply(lambda x:datetime.strptime(str(x), '%Y-%m-%d %H:%M:%S'))\n",
    "gpp_peru_h = gpp_peru_h['GPP_DT_VUT_REF']['2018':'2019']\n",
    "gpp_peru_g = create_flux_data(gpp_peru_h.to_frame(),['Gpp','GPP_DT_VUT_REF']) * 1000\n",
    "\n",
    "# Carregando os dados de GPP da torre de fluxo do Santarém\n",
    "gpp = pd.read_csv('./santarem_torre/ameriflux-santarem-dados-original.csv').rename(columns={'Unnamed: 0':'TIMESTAMP_START'})\n",
    "gpp.index = gpp['TIMESTAMP_START'].apply(lambda x:datetime.strptime(str(x), '%Y-%m-%d %H:%M:%S'))\n",
    "gpp = gpp['GPP_uStar_f']\n",
    "gpp_santarem_gc = create_flux_data(gpp.to_frame(),['Gpp','GPP_uStar_f']) * 1000\n",
    "\n",
    "# Aplicando um filtro para redução de ruído das séries temporais\n",
    "smoothed_gpp_peru_torre = ExponentialSmoothing(gpp_peru_g, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues\n",
    "smoothed_gpp_santarem_torre = ExponentialSmoothing(gpp_santarem_gc, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Cálculo do GPP com os Parâmetros da BPLUT (MERRA-2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Santarém"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def constructor_2(path):\n",
    "    \"\"\"Função para construir os Dataframes dos dados metereológicos do MERRA-2\"\"\"\n",
    "    nome = path.split('/')[1].split('_')[0]\n",
    "    df = pd.read_csv(path, parse_dates=['date'])\n",
    "    df = df.rename({'albedo':nome},axis=1)\n",
    "    hora_inicio = df['date'].min().floor('d')\n",
    "    df.index = hora_inicio + pd.to_timedelta(df.index, unit='h')\n",
    "    df = df.drop(['date','.geo'], axis=1)\n",
    "    df = df.drop('system:index',axis=1)\n",
    "    return df\n",
    "\n",
    "\n",
    "# Leitura e pré-processamento dos dados de temperatura\n",
    "tday = constructor_2('merra_Santarem/T10M_SANTAREM_Time_Series.csv') - 273.15 #( K° -> C°)\n",
    "tmin = tday['T10M'].resample('D').min().values  # Reamostrando para Temperatura mínima diária\n",
    "tmean_santarem = tday['T10M'].resample('D').mean().values  # Temperatura média diária\n",
    "tmean_vpd = tday['T10M'].between_time('9:00', '21:00').resample('D').mean().values\n",
    "             \n",
    "# Leitura e pré-processamento dos dados de razão de mistura de vapor de água\n",
    "qv10m = constructor_2('merra_Santarem/QV10M_SANTAREM_Time_Series.csv') # unidade: \"Mass fraction\" (kg/kg)\n",
    "qv10m = pd.to_numeric(qv10m['QV10M']).between_time('9:00', '21:00').resample('D').mean().values\n",
    "\n",
    "# Leitura e pré-processamento dos dados de pressão atmosférica\n",
    "ps = constructor_2('merra_Santarem/PS_SANTAREM_Time_Series.csv') # unidade: Pa  # Removendo vírgulas dos valores\n",
    "ps['PS'] = pd.to_numeric(ps['PS'])\n",
    "ps = ps.between_time('9:00', '21:00').resample('D').mean()\n",
    "ps = pd.to_numeric(ps['PS']).values\n",
    "\n",
    "# Leitura e pré-processamento dos dados de radiação solar incidente\n",
    "SWGNT = constructor_2('merra_Santarem/SWGNT_SANTAREM_Time_Series.csv')['SWGNT'].resample('D').mean().values # unidade: W/m^2\n",
    "\n",
    "# Função para converter datas no formato MODIS\n",
    "def modis_date(data_str):\n",
    "    ano = int(data_str[1:5])\n",
    "    dia_ano = int(data_str[5:])\n",
    "    data = datetime.strptime('{}-{}'.format(ano, dia_ano), '%Y-%j')\n",
    "    return data\n",
    "\n",
    "# Leitura e pré-processamento dos dados de Fração de Absorção de Luz Fotossinteticamente Ativa (FPAR)\n",
    "fpar = pd.read_csv('santarem_mod/filtered_scaled_Fpar_500m.csv', header=None,na_values='F') # unidade: (%)\n",
    "fpar.index = fpar[2].apply(modis_date)  # Convertendo a data no formato MODIS\n",
    "\n",
    "# Arrumando o dataframe\n",
    "# ================================\n",
    "fpar = fpar.drop([0,1,2,3,4], axis=1)\n",
    "fpar.columns = range(len(fpar.columns))\n",
    "fpar = fpar[144]\n",
    "\n",
    "# ================================\n",
    "\n",
    "# Tratamento de dados faltando usando interpolação linear (8 dias -> diário)\n",
    "fpar = fpar.resample('D').interpolate('linear')['2009-01-01':'2011-12-31']\n",
    "fpar = fpar.values\n",
    "\n",
    "drivers_santarem = [\n",
    "    fpar[:][...,None],\n",
    "    tmin[:][...,None],\n",
    "    MOD17.vpd(qv10m, ps, tmean_vpd)[...,None],\n",
    "    MOD17.par(SWGNT)[:][...,None]\n",
    "]\n",
    "drivers_santarem[2] = np.where(drivers_santarem[2] < 0, 0, drivers_santarem[2]) # VPD < 0 = 0\n",
    "\n",
    "# Gerando GPP a partir dos dados do bioma \"Evergreen Broadleaf Forest\" como parâmetro diretamente\n",
    "parametros = pd.read_csv('mod17_bplut/MOD17_BPLUT_CX.X_MERRA_NASA.csv')['EBF=1'].values\n",
    "\n",
    "parametros = [0.001405, -8.0, 9.09, 1000.0, 4000.0, 26.9, 2.0, 2.0, 1.1, 0.162, 0.00604, 0.00519, 0.00397] ## Igual ao acima\n",
    "\n",
    "gpp_ebf = np.nanmean(MOD17._gpp(parametros,drivers_santarem[0],drivers_santarem[1],drivers_santarem[2],drivers_santarem[3]), axis = -1) # Considerar como GPP do algorítmo\n",
    "\n",
    "gpp_ebf_santarem = pd.DataFrame(gpp_ebf,index=gpp_santarem_gc.index)\n",
    "\n",
    "smoothed_gpp_alg_santarem = ExponentialSmoothing(gpp_ebf_santarem, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Peru"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def constructor_2(path):\n",
    "    nome = path.split('/')[1].split('_')[0]\n",
    "    df = pd.read_csv(path, parse_dates=['date'])\n",
    "    df = df.rename({'albedo':nome},axis=1)\n",
    "    # Definindo a coluna 'system:time_start' como índice\n",
    "    hora_inicio = df['date'].min().floor('d')\n",
    "    df.index = hora_inicio + pd.to_timedelta(df.index, unit='h')\n",
    "    # Removendo a coluna 'system:time_start'\n",
    "    df = df.drop(['date','.geo'], axis=1)\n",
    "    df = df.drop('system:index',axis=1)\n",
    "    return df\n",
    "\n",
    "# Leitura e pré-processamento dos dados de temperatura\n",
    "tday = constructor_2('merra/T10M_2018.csv') - 273.15 #( K° -> C°)\n",
    "tmin = tday['T10M'].resample('D').min().values  # Reamostrando para Temperatura mínima diária\n",
    "tmean_peru = tday['T10M'].resample('D').mean().values  # Temperatura média diária\n",
    "tmean_vpd = tday['T10M'].between_time('10:00', '22:00').resample('D').mean().values\n",
    "             \n",
    "# Leitura e pré-processamento dos dados de razão de mistura de vapor de água\n",
    "qv10m = constructor_2('merra/QV10M_2018.csv') # unidade: \"Mass fraction\" (kg/kg)\n",
    "qv10m = pd.to_numeric(qv10m['QV10M']).between_time('10:00', '22:00').resample('D').mean().values\n",
    "\n",
    "# Leitura e pré-processamento dos dados de pressão atmosférica\n",
    "ps = constructor_2('merra/PS_2018.csv') # unidade: Pa  # Removendo vírgulas dos valores\n",
    "ps['PS'] = pd.to_numeric(ps['PS'])\n",
    "ps = ps.between_time('10:00', '22:00').resample('D').mean()\n",
    "ps = pd.to_numeric(ps['PS']).values\n",
    "\n",
    "# Leitura e pré-processamento dos dados de radiação solar incidente\n",
    "SWGNT = constructor_2('merra/SWGNT_2018.csv')['SWGNT'].resample('D').mean().values # unidade: W/m^2\n",
    "\n",
    "# Função para converter datas no formato MODIS\n",
    "def modis_date(data_str):\n",
    "    ano = int(data_str[1:5])\n",
    "    dia_ano = int(data_str[5:])\n",
    "    data = datetime.strptime('{}-{}'.format(ano, dia_ano), '%Y-%j')\n",
    "    return data\n",
    "\n",
    "# Leitura e pré-processamento dos dados de Fração de Absorção de Luz Fotossinteticamente Ativa (FPAR)\n",
    "fpar = pd.read_csv('mod_peru/filtered_scaled_Fpar_500m.csv', header=None,na_values='F') # unidade: (%)\n",
    "fpar.index = fpar[2].apply(modis_date)  # Convertendo a data no formato MODIS\n",
    "\n",
    "# Arrumando o dataframe\n",
    "# ================================\n",
    "fpar = fpar.drop([0,1,2,3,4], axis=1)\n",
    "fpar.columns = range(len(fpar.columns))\n",
    "fpar = fpar[144]\n",
    "\n",
    "# ================================\n",
    "\n",
    "# Tratamento de dados faltando usando interpolação linear (8 dias -> diário)\n",
    "fpar = fpar.resample('D').interpolate('linear')['2018':'2019']\n",
    "fpar = fpar.values\n",
    "\n",
    "drivers_peru = [\n",
    "    fpar[:][...,None],\n",
    "    tmin[:][...,None],\n",
    "    MOD17.vpd(qv10m, ps, tmean_vpd)[...,None],\n",
    "    MOD17.par(SWGNT)[:][...,None]\n",
    "]\n",
    "drivers_peru[2] = np.where(drivers_peru[2] < 0, 0, drivers_peru[2]) # VPD < 0 = 0\n",
    "\n",
    "# Gerando GPP a partir dos dados do bioma \"Evergreen Broadleaf Forest\" como parâmetro diretamente\n",
    "parametros = pd.read_csv('mod17_bplut/MOD17_BPLUT_CX.X_MERRA_NASA.csv')['EBF=1'].values\n",
    "\n",
    "parametros = [0.001405, -8.0, 9.09, 1000.0, 4000.0, 26.9, 2.0, 2.0, 1.1, 0.162, 0.00604, 0.00519, 0.00397] # Igual ao acima\n",
    "\n",
    "gpp_ebf = np.nanmean(MOD17._gpp(parametros,drivers_peru[0],drivers_peru[1],drivers_peru[2],drivers_peru[3]), axis = -1) # Considerar como GPP do algorítmo\n",
    "\n",
    "gpp_ebf_peru = pd.DataFrame(gpp_ebf,index=gpp_peru_g.index)\n",
    "\n",
    "smoothed_gpp_alg_peru = ExponentialSmoothing(gpp_ebf_peru, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Dados de GPP do MOD17"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Santarém"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Função para converter datas no formato MODIS\n",
    "def modis_date(data_str):\n",
    "    ano = int(data_str[1:5])\n",
    "    dia_ano = int(data_str[5:])\n",
    "    data = datetime.strptime('{}-{}'.format(ano, dia_ano), '%Y-%j')\n",
    "    return data\n",
    "\n",
    "# Leitura e pré-processamento dos dados de Fração de Absorção de Luz Fotossinteticamente Ativa (FPAR)\n",
    "Gpp_mod = pd.read_csv('./santarem_mod/Gpp_500m.csv', header=None,na_values='F') # unidade: (%)\n",
    "Gpp_mod.index = Gpp_mod[2].apply(modis_date)  # Convertendo a data no formato MODIS\n",
    "\n",
    "# Arrumando o dataframe\n",
    "# ================================\n",
    "Gpp_mod = Gpp_mod.drop([0,1,2,3,4], axis=1)\n",
    "Gpp_mod.columns = range(len(Gpp_mod.columns))\n",
    "Gpp_mod = Gpp_mod[144].to_frame().rename(columns={144:'Gpp_modis'})\n",
    "santarem_gpp_mod = Gpp_mod['2009-01-01':'2011-12-27'] * 0.0001 * 1000\n",
    "santarem_gpp_mod = santarem_gpp_mod['Gpp_modis']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Peru"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Função para converter datas no formato MODIS\n",
    "def modis_date(data_str):\n",
    "    ano = int(data_str[1:5])\n",
    "    dia_ano = int(data_str[5:])\n",
    "    data = datetime.strptime('{}-{}'.format(ano, dia_ano), '%Y-%j')\n",
    "    return data\n",
    "\n",
    "# Leitura e pré-processamento dos dados de Fração de Absorção de Luz Fotossinteticamente Ativa (FPAR)\n",
    "Gpp_mod = pd.read_csv('./mod_peru/Gpp_500m.csv', header=None,na_values='F') # unidade: (%)\n",
    "Gpp_mod.index = Gpp_mod[2].apply(modis_date)  # Convertendo a data no formato MODIS\n",
    "\n",
    "# Arrumando o dataframe\n",
    "# ================================\n",
    "Gpp_mod = Gpp_mod.drop([0,1,2,3,4], axis=1)\n",
    "Gpp_mod.columns = range(len(Gpp_mod.columns))\n",
    "Gpp_mod = Gpp_mod[144].to_frame().rename(columns={144:'Gpp_modis'})\n",
    "peru_gpp_mod = Gpp_mod['2018':'2019'] * 0.0001 * 1000\n",
    "peru_gpp_mod = peru_gpp_mod['Gpp_modis']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Agregando os dados Para a Resolução Temporal do MOD17"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "observacoes_santarem = smoothed_gpp_santarem_torre\n",
    "observacoes_peru = smoothed_gpp_peru_torre\n",
    "\n",
    "index_santarem = observacoes_santarem.index\n",
    "index_peru = observacoes_peru.index\n",
    "\n",
    "#Organizando os dados das observacões das torres para amostras de 8 dias\n",
    "#Mesma resolução temporal do MOD17\n",
    "gpp_gc_8d_2009_teste = observacoes_santarem['2009':'2009'].resample('8D').sum()\n",
    "gpp_gc_8d_2010_teste = observacoes_santarem['2010':'2010'].resample('8D').sum()\n",
    "gpp_gc_8d_2011_teste = observacoes_santarem['2011':'2011'].resample('8D').sum()\n",
    "observacoes_santarem = pd.concat([gpp_gc_8d_2009_teste,gpp_gc_8d_2010_teste,gpp_gc_8d_2011_teste],axis=0)\n",
    "\n",
    "gpp_gc_8d_2018_teste = observacoes_peru['2018':'2018'].resample('8D').sum()\n",
    "gpp_gc_8d_2019_teste = observacoes_peru['2019':'2019'].resample('8D').sum()\n",
    "observacoes_peru = pd.concat([gpp_gc_8d_2018_teste,gpp_gc_8d_2019_teste],axis=0)\n",
    "\n",
    "#Calculando RMSE entre os dados do satélite e torre\n",
    "rmse_mod_santarem = np.sqrt(mean_squared_error(y_true=observacoes_santarem, y_pred=santarem_gpp_mod))\n",
    "rmse_mod_peru = np.sqrt(mean_squared_error(y_true=observacoes_peru, y_pred=peru_gpp_mod))\n",
    "\n",
    "#Organizando os dados de GPP calculado com o algoritmo do mod17 amostras de 8 dias\n",
    "#Mesma resolução temporal do MOD17\n",
    "gpp_ebf_8d_2009_santarem = smoothed_gpp_alg_santarem['2009':'2009'].resample('8D').sum()\n",
    "gpp_ebf_8d_2010_santarem = smoothed_gpp_alg_santarem['2010':'2010'].resample('8D').sum()\n",
    "gpp_ebf_8d_2011_santarem = smoothed_gpp_alg_santarem['2011':'2011'].resample('8D').sum()\n",
    "smoothed_gpp_algpad_santarem = pd.concat([gpp_ebf_8d_2009_santarem,gpp_ebf_8d_2010_santarem,gpp_ebf_8d_2011_santarem],axis=0)\n",
    "\n",
    "gpp_ebf_8d_2018_teste = smoothed_gpp_alg_peru['2018':'2018'].resample('8D').sum()\n",
    "gpp_ebf_8d_2019_teste = smoothed_gpp_alg_peru['2019':'2019'].resample('8D').sum()\n",
    "smoothed_gpp_algpad_peru = pd.concat([gpp_ebf_8d_2018_teste,gpp_ebf_8d_2019_teste],axis=0)\n",
    "\n",
    "#Calculando o RMSE o algorotimo do mod17 e dados da torre.\n",
    "rmse_algpad_santarem = np.sqrt(mean_squared_error(y_true=observacoes_santarem, y_pred=smoothed_gpp_algpad_santarem))\n",
    "rmse_algpad_peru = np.sqrt(mean_squared_error(y_true=observacoes_peru, y_pred=smoothed_gpp_algpad_peru))\n",
    "[rmse_algpad_peru,rmse_algpad_santarem]\n",
    "\n",
    "initial_population = np.array([0.001405, -8.0, 9.09, 1000.0, 4000.0, 26.9, 2.0, 2.0, 1.1, 0.162, 0.00604, 0.00519, 0.00397])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Algoritmo Genético"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "IdOzgXBzbdFc",
    "outputId": "4ac0eb10-9e85-47b3-8109-9f7f02e5d41e",
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "=================================================================================================================================\n",
      "n_gen  |  n_eval  |     LUEmax    |    Tmin_min   |    Tmin_max   |    VPD_min    |    VPD_max    | RMSE Santarem |   RMSE Peru  \n",
      "=================================================================================================================================\n",
      "     1 |        1 |  0.0014050000 | -8.000000E+00 |  9.0900000000 |  1.000000E+03 |  4.000000E+03 |  2.051345E+01 |  2.310078E+01\n",
      "     2 |     1001 |  0.0015522035 | -8.000000E+00 |  9.0900000000 |  1.000000E+03 |  4.336867E+03 |  1.850748E+01 |  2.107683E+01\n",
      "     3 |     2001 |  0.0016031566 | -8.000000E+00 |  9.0061613116 |  1.000000E+03 |  4.654378E+03 |  1.829135E+01 |  2.077856E+01\n",
      "     4 |     3001 |  0.0015526112 | -8.000000E+00 |  9.0900000000 |  9.615598E+02 |  5.355818E+03 |  1.768853E+01 |  2.107224E+01\n",
      "     5 |     4001 |  0.0015859777 | -9.341189E+00 |  9.0511805165 |  9.876374E+02 |  5.266934E+03 |  1.784888E+01 |  2.085467E+01\n",
      "     6 |     5001 |  0.0015526112 | -8.000000E+00 |  9.0900000000 |  9.616104E+02 |  5.741705E+03 |  1.754983E+01 |  2.107318E+01\n",
      "     7 |     6001 |  0.0015782967 | -8.004391E+00 |  9.0900000000 |  9.999583E+02 |  5.744809E+03 |  1.766644E+01 |  2.089914E+01\n",
      "     8 |     7001 |  0.0015742982 | -7.539468E+00 |  9.9037266134 |  1.000000E+03 |  5.747045E+03 |  1.763693E+01 |  2.092310E+01\n",
      "     9 |     8001 |  0.0015529040 | -8.382883E+00 |  8.7745932396 |  9.980724E+02 |  6.045533E+03 |  1.746313E+01 |  2.107455E+01\n",
      "    10 |     9001 |  0.0015599018 | -8.520793E+00 |  9.0916122531 |  1.000000E+03 |  6.211376E+03 |  1.748700E+01 |  2.102125E+01\n",
      "    11 |    10001 |  0.0016018121 | -8.604136E+00 |  9.0900000000 |  7.203996E+02 |  5.930352E+03 |  1.772798E+01 |  2.069907E+01\n",
      "    12 |    11001 |  0.0016018121 | -8.604136E+00 |  9.0900000000 |  7.203996E+02 |  5.930352E+03 |  1.772798E+01 |  2.069907E+01\n",
      "    13 |    12001 |  0.0016018121 | -8.604136E+00 |  9.0900000000 |  7.203996E+02 |  5.930352E+03 |  1.772798E+01 |  2.069907E+01\n",
      "    14 |    13001 |  0.0015715949 | -9.166078E+00 |  9.1052943793 |  6.798508E+02 |  6.491493E+03 |  1.744486E+01 |  2.086591E+01\n",
      "    15 |    14001 |  0.0015756853 | -5.491354E+00 |  1.058653E+01 |  6.119488E+02 |  6.345926E+03 |  1.745623E+01 |  2.082500E+01\n",
      "    16 |    15001 |  0.0015865174 | -8.676728E+00 |  7.6549978617 |  5.919113E+02 |  6.336649E+03 |  1.746772E+01 |  2.074726E+01\n",
      "    17 |    16001 |  0.0015865174 | -7.954476E+00 |  7.6549978617 |  5.933342E+02 |  6.342304E+03 |  1.746686E+01 |  2.074733E+01\n",
      "    18 |    17001 |  0.0016022075 | -7.821342E+00 |  9.0900000000 |  5.988435E+02 |  6.271173E+03 |  1.753173E+01 |  2.064853E+01\n",
      "    19 |    18001 |  0.0016072193 | -7.962085E+00 |  1.046641E+01 |  5.688718E+02 |  6.428306E+03 |  1.749237E+01 |  2.061817E+01\n",
      "    20 |    19001 |  0.0016072193 | -7.962085E+00 |  1.046641E+01 |  5.688718E+02 |  6.428306E+03 |  1.749237E+01 |  2.061817E+01\n",
      "    21 |    20001 |  0.0016380146 | -7.401553E+00 |  1.157905E+01 |  4.918144E+02 |  6.490605E+03 |  1.753822E+01 |  2.047547E+01\n",
      "    22 |    21001 |  0.0016668932 | -8.737133E+00 |  7.8231330130 |  3.583367E+02 |  6.477808E+03 |  1.751024E+01 |  2.041280E+01\n",
      "    23 |    22001 |  0.0016668932 | -8.737133E+00 |  7.8231330130 |  3.583367E+02 |  6.477808E+03 |  1.751024E+01 |  2.041280E+01\n",
      "    24 |    23001 |  0.0016673257 | -8.755982E+00 |  6.1070918523 |  3.261337E+02 |  6.435638E+03 |  1.748162E+01 |  2.043861E+01\n",
      "    25 |    24001 |  0.0016687396 | -8.332895E+00 |  8.3294578804 |  3.161805E+02 |  6.490304E+03 |  1.746890E+01 |  2.044210E+01\n",
      "    26 |    25001 |  0.0016687396 | -8.332895E+00 |  8.3294578804 |  3.161805E+02 |  6.490304E+03 |  1.746890E+01 |  2.044210E+01\n",
      "    27 |    26001 |  0.0016697298 | -7.765454E+00 |  1.057196E+01 |  3.075151E+02 |  6.488247E+03 |  1.746420E+01 |  2.044638E+01\n",
      "    28 |    27001 |  0.0016895594 | -6.544010E+00 |  6.0446236850 |  2.622553E+02 |  6.494333E+03 |  1.749407E+01 |  2.040039E+01\n",
      "    29 |    28001 |  0.0016895594 | -6.544010E+00 |  6.0446236850 |  2.622553E+02 |  6.494333E+03 |  1.749407E+01 |  2.040039E+01\n",
      "    30 |    29001 |  0.0016895594 | -6.544010E+00 |  6.0446236850 |  2.622553E+02 |  6.494333E+03 |  1.749407E+01 |  2.040039E+01\n",
      "    31 |    30001 |  0.0016895594 | -6.544010E+00 |  6.0446236850 |  2.622553E+02 |  6.494333E+03 |  1.749407E+01 |  2.040039E+01\n",
      "    32 |    31001 |  0.0016323945 | -1.236387E+01 |  2.411816E+01 |  5.056570E+02 |  6.487958E+03 |  1.752864E+01 |  2.027670E+01\n",
      "    33 |    32001 |  0.0016694462 | -6.631469E+00 |  2.349362E+01 |  3.173664E+02 |  6.490764E+03 |  1.747318E+01 |  2.021205E+01\n",
      "    34 |    33001 |  0.0016895594 | -6.614329E+00 |  2.435592E+01 |  2.618116E+02 |  6.499191E+03 |  1.749583E+01 |  2.017989E+01\n",
      "    35 |    34001 |  0.0016840958 | -8.328013E+00 |  2.411816E+01 |  2.895348E+02 |  6.498510E+03 |  1.750379E+01 |  2.015561E+01\n",
      "    36 |    35001 |  0.0016895594 | -6.530360E+00 |  2.379167E+01 |  2.622553E+02 |  6.498133E+03 |  1.749417E+01 |  2.014863E+01\n",
      "    37 |    36001 |  0.0016744043 | -2.337635E+00 |  2.387467E+01 |  3.176429E+02 |  6.491161E+03 |  1.749420E+01 |  2.012579E+01\n",
      "    38 |    37001 |  0.0016873002 | -3.094953E+00 |  2.411816E+01 |  3.116418E+02 |  6.499505E+03 |  1.755130E+01 |  2.006434E+01\n",
      "    39 |    38001 |  0.0016873002 | -3.094953E+00 |  2.411816E+01 |  3.116418E+02 |  6.499505E+03 |  1.755130E+01 |  2.006434E+01\n",
      "    40 |    39001 |  0.0016873002 | -3.094953E+00 |  2.411816E+01 |  3.116418E+02 |  6.499505E+03 |  1.755130E+01 |  2.006434E+01\n",
      "    41 |    40001 |  0.0016873002 | -2.101159E+00 |  2.411816E+01 |  3.116418E+02 |  6.493465E+03 |  1.755170E+01 |  2.005544E+01\n",
      "    42 |    41001 |  0.0016903625 | -2.060257E+00 |  2.395247E+01 |  2.648599E+02 |  6.490834E+03 |  1.750179E+01 |  2.010012E+01\n",
      "    43 |    42001 |  0.0016977043 | -2.211166E+00 |  2.396905E+01 |  2.963130E+02 |  6.499181E+03 |  1.758759E+01 |  2.001284E+01\n",
      "    44 |    43001 |  0.0017031028 | -2.216906E+00 |  2.385873E+01 |  2.576178E+02 |  6.493643E+03 |  1.755674E+01 |  2.003569E+01\n",
      "    45 |    44001 |  0.0017074256 | -2.270090E+00 |  2.388590E+01 |  2.382342E+02 |  6.493658E+03 |  1.755119E+01 |  2.003742E+01\n",
      "    46 |    45001 |  0.0017029863 | -2.045708E+00 |  2.397063E+01 |  2.578839E+02 |  6.499570E+03 |  1.755635E+01 |  2.003205E+01\n",
      "    47 |    46001 |  0.0017029863 | -2.043317E+00 |  2.396994E+01 |  2.578839E+02 |  6.499675E+03 |  1.755634E+01 |  2.003202E+01\n",
      "    48 |    47001 |  0.0017126461 | -2.004893E+00 |  2.397444E+01 |  2.287152E+02 |  6.496564E+03 |  1.756624E+01 |  2.001756E+01\n",
      "    49 |    48001 |  0.0017126461 | -2.002848E+00 |  2.398257E+01 |  2.287152E+02 |  6.496513E+03 |  1.756625E+01 |  2.001747E+01\n",
      "    50 |    49001 |  0.0017126461 | -2.002848E+00 |  2.398257E+01 |  2.287152E+02 |  6.496513E+03 |  1.756625E+01 |  2.001747E+01\n",
      "    51 |    50001 |  0.0017126461 | -2.002848E+00 |  2.398257E+01 |  2.287152E+02 |  6.496513E+03 |  1.756625E+01 |  2.001747E+01\n",
      "    52 |    51001 |  0.0017230153 | -2.074965E+00 |  2.398326E+01 |  1.863052E+02 |  6.499937E+03 |  1.755919E+01 |  2.002450E+01\n",
      "    53 |    52001 |  0.0017125754 | -2.016374E+00 |  2.395634E+01 |  2.238624E+02 |  6.499981E+03 |  1.755803E+01 |  2.002556E+01\n",
      "    54 |    53001 |  0.0017222846 | -2.004095E+00 |  2.398125E+01 |  1.877233E+02 |  6.499298E+03 |  1.755725E+01 |  2.002577E+01\n",
      "    55 |    54001 |  0.0017222846 | -2.004095E+00 |  2.398125E+01 |  1.877233E+02 |  6.499298E+03 |  1.755725E+01 |  2.002577E+01\n",
      "    56 |    55001 |  0.0017222846 | -2.001602E+00 |  2.398125E+01 |  1.877233E+02 |  6.499298E+03 |  1.755725E+01 |  2.002575E+01\n",
      "    57 |    56001 |  0.0017222846 | -2.001602E+00 |  2.398125E+01 |  1.877233E+02 |  6.499298E+03 |  1.755725E+01 |  2.002575E+01\n",
      "    58 |    57001 |  0.0017242649 | -2.005622E+00 |  2.398042E+01 |  1.831054E+02 |  6.499914E+03 |  1.756134E+01 |  2.002164E+01\n",
      "    59 |    58001 |  0.0017227019 | -2.004097E+00 |  2.398608E+01 |  1.879980E+02 |  6.499829E+03 |  1.756005E+01 |  2.002291E+01\n",
      "    60 |    59001 |  0.0017227019 | -2.002768E+00 |  2.398608E+01 |  1.879980E+02 |  6.499829E+03 |  1.756005E+01 |  2.002289E+01\n",
      "    61 |    60001 |  0.0017235111 | -2.000857E+00 |  2.398327E+01 |  1.859698E+02 |  6.499793E+03 |  1.756151E+01 |  2.002143E+01\n",
      "    62 |    61001 |  0.0017232560 | -2.000971E+00 |  2.398346E+01 |  1.863082E+02 |  6.499917E+03 |  1.756057E+01 |  2.002235E+01\n",
      "    63 |    62001 |  0.0017243865 | -2.000549E+00 |  2.398125E+01 |  1.822787E+02 |  6.499998E+03 |  1.756074E+01 |  2.002217E+01\n",
      "    64 |    63001 |  0.0017243865 | -2.000549E+00 |  2.398125E+01 |  1.822787E+02 |  6.499998E+03 |  1.756074E+01 |  2.002217E+01\n",
      "    65 |    64001 |  0.0017243865 | -2.000549E+00 |  2.398125E+01 |  1.822787E+02 |  6.499998E+03 |  1.756074E+01 |  2.002217E+01\n",
      "    66 |    65001 |  0.0017243925 | -2.000676E+00 |  2.398278E+01 |  1.822787E+02 |  6.499999E+03 |  1.756078E+01 |  2.002213E+01\n",
      "    67 |    66001 |  0.0017244017 | -2.000387E+00 |  2.398671E+01 |  1.822787E+02 |  6.499999E+03 |  1.756084E+01 |  2.002206E+01\n",
      "    68 |    67001 |  0.0017244017 | -2.000387E+00 |  2.398671E+01 |  1.822787E+02 |  6.499999E+03 |  1.756084E+01 |  2.002206E+01\n",
      "    69 |    68001 |  0.0017244017 | -2.000387E+00 |  2.398671E+01 |  1.822787E+02 |  6.499999E+03 |  1.756084E+01 |  2.002206E+01\n",
      "    70 |    69001 |  0.0017244017 | -2.000387E+00 |  2.398671E+01 |  1.822787E+02 |  6.499999E+03 |  1.756084E+01 |  2.002206E+01\n",
      "    71 |    70001 |  0.0017244017 | -2.000387E+00 |  2.398670E+01 |  1.822787E+02 |  6.499999E+03 |  1.756084E+01 |  2.002205E+01\n",
      "    72 |    71001 |  0.0017242923 | -2.000539E+00 |  2.398612E+01 |  1.823507E+02 |  6.499991E+03 |  1.756033E+01 |  2.002257E+01\n",
      "    73 |    72001 |  0.0017242871 | -2.000120E+00 |  2.398611E+01 |  1.823462E+02 |  6.499991E+03 |  1.756029E+01 |  2.002260E+01\n",
      "    74 |    73001 |  0.0017243044 | -2.000067E+00 |  2.398612E+01 |  1.822787E+02 |  6.499993E+03 |  1.756028E+01 |  2.002260E+01\n",
      "    75 |    74001 |  0.0017243044 | -2.000067E+00 |  2.398612E+01 |  1.822787E+02 |  6.499993E+03 |  1.756028E+01 |  2.002260E+01\n",
      "    76 |    75001 |  0.0017245191 | -2.000079E+00 |  2.398632E+01 |  1.806971E+02 |  6.499990E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    77 |    76001 |  0.0017245191 | -2.000079E+00 |  2.398632E+01 |  1.806971E+02 |  6.499990E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    78 |    77001 |  0.0017245191 | -2.000079E+00 |  2.398632E+01 |  1.806971E+02 |  6.499990E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    79 |    78001 |  0.0017245191 | -2.000079E+00 |  2.398632E+01 |  1.806971E+02 |  6.500000E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    80 |    79001 |  0.0017245191 | -2.000079E+00 |  2.398632E+01 |  1.806971E+02 |  6.500000E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    81 |    80001 |  0.0017245191 | -2.000079E+00 |  2.398632E+01 |  1.806971E+02 |  6.500000E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    82 |    81001 |  0.0017245191 | -2.000007E+00 |  2.398632E+01 |  1.806971E+02 |  6.500000E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    83 |    82001 |  0.0017245191 | -2.000007E+00 |  2.398632E+01 |  1.806971E+02 |  6.500000E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    84 |    83001 |  0.0017245191 | -2.000007E+00 |  2.398632E+01 |  1.806971E+02 |  6.500000E+03 |  1.755905E+01 |  2.002383E+01\n",
      "    85 |    84001 |  0.0017245271 | -2.000001E+00 |  2.398629E+01 |  1.806810E+02 |  6.499999E+03 |  1.755907E+01 |  2.002381E+01\n",
      "    86 |    85001 |  0.0017245271 | -2.000001E+00 |  2.398629E+01 |  1.806810E+02 |  6.499999E+03 |  1.755907E+01 |  2.002381E+01\n",
      "    87 |    86001 |  0.0017245271 | -2.000001E+00 |  2.398629E+01 |  1.806810E+02 |  6.499999E+03 |  1.755907E+01 |  2.002381E+01\n",
      "    88 |    87001 |  0.0017245190 | -2.000001E+00 |  2.398632E+01 |  1.808005E+02 |  6.500000E+03 |  1.755921E+01 |  2.002367E+01\n",
      "    89 |    88001 |  0.0017245271 | -2.000001E+00 |  2.398632E+01 |  1.806976E+02 |  6.499998E+03 |  1.755909E+01 |  2.002379E+01\n",
      "    90 |    89001 |  0.0017246868 | -2.000005E+00 |  2.398614E+01 |  1.808134E+02 |  6.500000E+03 |  1.756018E+01 |  2.002270E+01\n",
      "    91 |    90001 |  0.0017246868 | -2.000005E+00 |  2.398614E+01 |  1.808134E+02 |  6.500000E+03 |  1.756018E+01 |  2.002270E+01\n",
      "    92 |    91001 |  0.0017246940 | -2.000005E+00 |  2.398632E+01 |  1.808193E+02 |  6.499996E+03 |  1.756023E+01 |  2.002265E+01\n",
      "    93 |    92001 |  0.0017246917 | -2.000005E+00 |  2.398632E+01 |  1.806724E+02 |  6.499995E+03 |  1.755999E+01 |  2.002289E+01\n",
      "    94 |    93001 |  0.0017246856 | -2.000000E+00 |  2.398629E+01 |  1.806513E+02 |  6.500000E+03 |  1.755993E+01 |  2.002295E+01\n",
      "    95 |    94001 |  0.0017246856 | -2.000000E+00 |  2.398629E+01 |  1.806513E+02 |  6.500000E+03 |  1.755993E+01 |  2.002295E+01\n",
      "    96 |    95001 |  0.0017246856 | -2.000000E+00 |  2.398629E+01 |  1.806513E+02 |  6.500000E+03 |  1.755993E+01 |  2.002295E+01\n",
      "    97 |    96001 |  0.0017246952 | -2.000003E+00 |  2.398633E+01 |  1.806513E+02 |  6.500000E+03 |  1.755998E+01 |  2.002290E+01\n",
      "    98 |    97001 |  0.0017246856 | -2.000005E+00 |  2.398634E+01 |  1.806513E+02 |  6.500000E+03 |  1.755993E+01 |  2.002295E+01\n",
      "    99 |    98001 |  0.0017246892 | -2.000001E+00 |  2.398634E+01 |  1.805981E+02 |  6.499999E+03 |  1.755986E+01 |  2.002302E+01\n",
      "   100 |    99001 |  0.0017246892 | -2.000001E+00 |  2.398634E+01 |  1.805981E+02 |  6.499999E+03 |  1.755986E+01 |  2.002302E+01\n",
      "   101 |   100001 |  0.0017246892 | -2.000000E+00 |  2.398634E+01 |  1.805981E+02 |  6.499999E+03 |  1.755986E+01 |  2.002302E+01\n",
      "   102 |   101001 |  0.0017246949 | -2.000002E+00 |  2.398633E+01 |  1.799202E+02 |  6.500000E+03 |  1.755884E+01 |  2.002403E+01\n",
      "   103 |   102001 |  0.0017246949 | -2.000002E+00 |  2.398633E+01 |  1.799202E+02 |  6.500000E+03 |  1.755884E+01 |  2.002403E+01\n",
      "   104 |   103001 |  0.0017246903 | -2.000001E+00 |  2.398632E+01 |  1.800813E+02 |  6.500000E+03 |  1.755907E+01 |  2.002381E+01\n",
      "   105 |   104001 |  0.0017246903 | -2.000001E+00 |  2.398632E+01 |  1.800863E+02 |  6.500000E+03 |  1.755907E+01 |  2.002380E+01\n",
      "   106 |   105001 |  0.0017246969 | -2.000000E+00 |  2.398634E+01 |  1.800720E+02 |  6.500000E+03 |  1.755909E+01 |  2.002379E+01\n",
      "   107 |   106001 |  0.0017246969 | -2.000000E+00 |  2.398634E+01 |  1.800720E+02 |  6.500000E+03 |  1.755909E+01 |  2.002379E+01\n",
      "   108 |   107001 |  0.0017246969 | -2.000000E+00 |  2.398634E+01 |  1.800720E+02 |  6.500000E+03 |  1.755909E+01 |  2.002379E+01\n",
      "   109 |   108001 |  0.0017247033 | -2.000001E+00 |  2.398632E+01 |  1.801350E+02 |  6.500000E+03 |  1.755922E+01 |  2.002365E+01\n",
      "   110 |   109001 |  0.0017247033 | -2.000000E+00 |  2.398634E+01 |  1.801296E+02 |  6.500000E+03 |  1.755922E+01 |  2.002366E+01\n",
      "   111 |   110001 |  0.0017247033 | -2.000000E+00 |  2.398634E+01 |  1.801296E+02 |  6.500000E+03 |  1.755922E+01 |  2.002366E+01\n",
      "   112 |   111001 |  0.0017259326 | -2.000000E+00 |  2.398633E+01 |  1.766338E+02 |  6.500000E+03 |  1.756078E+01 |  2.002207E+01\n",
      "   113 |   112001 |  0.0017259326 | -2.000000E+00 |  2.398633E+01 |  1.766338E+02 |  6.500000E+03 |  1.756078E+01 |  2.002207E+01\n",
      "   114 |   113001 |  0.0017258540 | -2.000000E+00 |  2.398633E+01 |  1.766338E+02 |  6.500000E+03 |  1.756034E+01 |  2.002252E+01\n",
      "   115 |   114001 |  0.0017258418 | -2.000000E+00 |  2.398633E+01 |  1.766338E+02 |  6.500000E+03 |  1.756027E+01 |  2.002259E+01\n",
      "   116 |   115001 |  0.0017258418 | -2.000000E+00 |  2.398633E+01 |  1.766338E+02 |  6.500000E+03 |  1.756027E+01 |  2.002259E+01\n",
      "   117 |   116001 |  0.0017258418 | -2.000000E+00 |  2.398633E+01 |  1.766338E+02 |  6.500000E+03 |  1.756027E+01 |  2.002259E+01\n",
      "   118 |   117001 |  0.0017258418 | -2.000000E+00 |  2.398633E+01 |  1.766338E+02 |  6.500000E+03 |  1.756027E+01 |  2.002259E+01\n",
      "   119 |   118001 |  0.0017258126 | -2.000006E+00 |  2.398634E+01 |  1.766216E+02 |  6.500000E+03 |  1.756008E+01 |  2.002277E+01\n",
      "   120 |   119001 |  0.0017258126 | -2.000006E+00 |  2.398634E+01 |  1.766216E+02 |  6.500000E+03 |  1.756008E+01 |  2.002277E+01\n",
      "   121 |   120001 |  0.0017257876 | -2.000006E+00 |  2.398634E+01 |  1.766216E+02 |  6.500000E+03 |  1.755994E+01 |  2.002292E+01\n",
      "   122 |   121001 |  0.0017258004 | -2.000000E+00 |  2.398634E+01 |  1.766932E+02 |  6.500000E+03 |  1.756012E+01 |  2.002273E+01\n",
      "   123 |   122001 |  0.0017258017 | -2.000000E+00 |  2.398634E+01 |  1.767514E+02 |  6.500000E+03 |  1.756022E+01 |  2.002263E+01\n",
      "   124 |   123001 |  0.0017258017 | -2.000000E+00 |  2.398634E+01 |  1.767514E+02 |  6.500000E+03 |  1.756022E+01 |  2.002263E+01\n",
      "   125 |   124001 |  0.0017258017 | -2.000000E+00 |  2.398634E+01 |  1.767514E+02 |  6.500000E+03 |  1.756022E+01 |  2.002263E+01\n",
      "   126 |   125001 |  0.0017257787 | -2.000000E+00 |  2.398634E+01 |  1.767761E+02 |  6.500000E+03 |  1.756013E+01 |  2.002273E+01\n",
      "   127 |   126001 |  0.0017257787 | -2.000000E+00 |  2.398634E+01 |  1.767761E+02 |  6.500000E+03 |  1.756013E+01 |  2.002273E+01\n",
      "   128 |   127001 |  0.0017257787 | -2.000000E+00 |  2.398634E+01 |  1.767761E+02 |  6.500000E+03 |  1.756013E+01 |  2.002273E+01\n",
      "   129 |   128001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   130 |   129001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   131 |   130001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   132 |   131001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   133 |   132001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   134 |   133001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   135 |   134001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   136 |   135001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   137 |   136001 |  0.0017257761 | -2.000001E+00 |  2.398634E+01 |  1.767670E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   138 |   137001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767594E+02 |  6.500000E+03 |  1.756008E+01 |  2.002277E+01\n",
      "   139 |   138001 |  0.0017257748 | -2.000000E+00 |  2.398634E+01 |  1.767562E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   140 |   139001 |  0.0017257748 | -2.000000E+00 |  2.398634E+01 |  1.767562E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   141 |   140001 |  0.0017257748 | -2.000000E+00 |  2.398634E+01 |  1.767562E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   142 |   141001 |  0.0017257761 | -2.000000E+00 |  2.398634E+01 |  1.767676E+02 |  6.500000E+03 |  1.756010E+01 |  2.002275E+01\n",
      "   143 |   142001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   144 |   143001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   145 |   144001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   146 |   145001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   147 |   146001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   148 |   147001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   149 |   148001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   150 |   149001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   151 |   150001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   152 |   151001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   153 |   152001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   154 |   153001 |  0.0017257755 | -2.000000E+00 |  2.398634E+01 |  1.767627E+02 |  6.500000E+03 |  1.756009E+01 |  2.002276E+01\n",
      "   155 |   154001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   156 |   155001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   157 |   156001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   158 |   157001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   159 |   158001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   160 |   159001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   161 |   160001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767575E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   162 |   161001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767575E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   163 |   162001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   164 |   163001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   165 |   164001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   166 |   165001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   167 |   166001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   168 |   167001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   169 |   168001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   170 |   169001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   171 |   170001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   172 |   171001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   173 |   172001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   174 |   173001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   175 |   174001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   176 |   175001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   177 |   176001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   178 |   177001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   179 |   178001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   180 |   179001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   181 |   180001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   182 |   181001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   183 |   182001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   184 |   183001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   185 |   184001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   186 |   185001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   187 |   186001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   188 |   187001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   189 |   188001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   190 |   189001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   191 |   190001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   192 |   191001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   193 |   192001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   194 |   193001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   195 |   194001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   196 |   195001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   197 |   196001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   198 |   197001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   199 |   198001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   200 |   199001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   201 |   200001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   202 |   201001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   203 |   202001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   204 |   203001 |  0.0017257747 | -2.000000E+00 |  2.398634E+01 |  1.767567E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   205 |   204001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   206 |   205001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   207 |   206001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   208 |   207001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   209 |   208001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   210 |   209001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   211 |   210001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   212 |   211001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   213 |   212001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   214 |   213001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   215 |   214001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   216 |   215001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   217 |   216001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   218 |   217001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   219 |   218001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   220 |   219001 |  0.0017257742 | -2.000000E+00 |  2.398634E+01 |  1.767578E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   221 |   220001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   222 |   221001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   223 |   222001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   224 |   223001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   225 |   224001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   226 |   225001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   227 |   226001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   228 |   227001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   229 |   228001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   230 |   229001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   231 |   230001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   232 |   231001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   233 |   232001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   234 |   233001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   235 |   234001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   236 |   235001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   237 |   236001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   238 |   237001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   239 |   238001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   240 |   239001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   241 |   240001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   242 |   241001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   243 |   242001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   244 |   243001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   245 |   244001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   246 |   245001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   247 |   246001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   248 |   247001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   249 |   248001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   250 |   249001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   251 |   250001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   252 |   251001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   253 |   252001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   254 |   253001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   255 |   254001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   256 |   255001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   257 |   256001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   258 |   257001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   259 |   258001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   260 |   259001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   261 |   260001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   262 |   261001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   263 |   262001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   264 |   263001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   265 |   264001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   266 |   265001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   267 |   266001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   268 |   267001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   269 |   268001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   270 |   269001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   271 |   270001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   272 |   271001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   273 |   272001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   274 |   273001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   275 |   274001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   276 |   275001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   277 |   276001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   278 |   277001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   279 |   278001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   280 |   279001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   281 |   280001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   282 |   281001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   283 |   282001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   284 |   283001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   285 |   284001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   286 |   285001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   287 |   286001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   288 |   287001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   289 |   288001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   290 |   289001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   291 |   290001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   292 |   291001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   293 |   292001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   294 |   293001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   295 |   294001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   296 |   295001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   297 |   296001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   298 |   297001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   299 |   298001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   300 |   299001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   301 |   300001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   302 |   301001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   303 |   302001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   304 |   303001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   305 |   304001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   306 |   305001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   307 |   306001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   308 |   307001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   309 |   308001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   310 |   309001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   311 |   310001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   312 |   311001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   313 |   312001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   314 |   313001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   315 |   314001 |  0.0017257744 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.756007E+01 |  2.002278E+01\n",
      "   316 |   315001 |  0.0017257476 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   317 |   316001 |  0.0017257476 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   318 |   317001 |  0.0017257476 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   319 |   318001 |  0.0017257476 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   320 |   319001 |  0.0017257476 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   321 |   320001 |  0.0017257476 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   322 |   321001 |  0.0017257476 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   323 |   322001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.755992E+01 |  2.002293E+01\n",
      "   324 |   323001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.755992E+01 |  2.002293E+01\n",
      "   325 |   324001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.755992E+01 |  2.002293E+01\n",
      "   326 |   325001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.755992E+01 |  2.002293E+01\n",
      "   327 |   326001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.755992E+01 |  2.002293E+01\n",
      "   328 |   327001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.755992E+01 |  2.002293E+01\n",
      "   329 |   328001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.755992E+01 |  2.002293E+01\n",
      "   330 |   329001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767577E+02 |  6.500000E+03 |  1.755992E+01 |  2.002293E+01\n",
      "   331 |   330001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767602E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   332 |   331001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767602E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   333 |   332001 |  0.0017257480 | -2.000000E+00 |  2.398634E+01 |  1.767602E+02 |  6.500000E+03 |  1.755993E+01 |  2.002293E+01\n",
      "   334 |   333001 |  0.0017257452 | -2.000000E+00 |  2.398634E+01 |  1.767576E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   335 |   334001 |  0.0017257452 | -2.000000E+00 |  2.398634E+01 |  1.767576E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   336 |   335001 |  0.0017257452 | -2.000000E+00 |  2.398634E+01 |  1.767576E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   337 |   336001 |  0.0017257452 | -2.000000E+00 |  2.398634E+01 |  1.767576E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   338 |   337001 |  0.0017257452 | -2.000000E+00 |  2.398634E+01 |  1.767576E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   339 |   338001 |  0.0017257452 | -2.000000E+00 |  2.398634E+01 |  1.767576E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   340 |   339001 |  0.0017257493 | -2.000000E+00 |  2.398634E+01 |  1.767609E+02 |  6.500000E+03 |  1.755994E+01 |  2.002292E+01\n",
      "   341 |   340001 |  0.0017257493 | -2.000000E+00 |  2.398634E+01 |  1.767609E+02 |  6.500000E+03 |  1.755994E+01 |  2.002292E+01\n",
      "   342 |   341001 |  0.0017257493 | -2.000000E+00 |  2.398634E+01 |  1.767609E+02 |  6.500000E+03 |  1.755994E+01 |  2.002292E+01\n",
      "   343 |   342001 |  0.0017257493 | -2.000000E+00 |  2.398634E+01 |  1.767609E+02 |  6.500000E+03 |  1.755994E+01 |  2.002292E+01\n",
      "   344 |   343001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   345 |   344001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   346 |   345001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   347 |   346001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   348 |   347001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   349 |   348001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   350 |   349001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   351 |   350001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   352 |   351001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   353 |   352001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   354 |   353001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   355 |   354001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   356 |   355001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   357 |   356001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   358 |   357001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   359 |   358001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   360 |   359001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   361 |   360001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   362 |   361001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   363 |   362001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   364 |   363001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   365 |   364001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   366 |   365001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   367 |   366001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   368 |   367001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   369 |   368001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   370 |   369001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   371 |   370001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   372 |   371001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   373 |   372001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   374 |   373001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   375 |   374001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   376 |   375001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   377 |   376001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   378 |   377001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   379 |   378001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   380 |   379001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   381 |   380001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   382 |   381001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   383 |   382001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   384 |   383001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   385 |   384001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   386 |   385001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   387 |   386001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   388 |   387001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   389 |   388001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   390 |   389001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   391 |   390001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   392 |   391001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   393 |   392001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   394 |   393001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   395 |   394001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   396 |   395001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   397 |   396001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   398 |   397001 |  0.0017257451 | -2.000000E+00 |  2.398634E+01 |  1.767568E+02 |  6.500000E+03 |  1.755991E+01 |  2.002295E+01\n",
      "   399 |   398001 |  0.0017257495 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755994E+01 |  2.002292E+01\n",
      "   400 |   399001 |  0.0017257495 | -2.000000E+00 |  2.398634E+01 |  1.767608E+02 |  6.500000E+03 |  1.755994E+01 |  2.002292E+01\n",
      "Function 'exec_model' executed in 7182.4234s\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Porcentagem de melhora: 5.030684710754182\n",
      "Porcentagem de melhora: 9.728784322503403\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def get_data(x):\n",
    "    '''Função para calcular os dados de GPP com os parametros em otimização pelo Algoritmo Genético'''\n",
    "    global observacoes_santarem\n",
    "    global observacoes_peru\n",
    "    previsoes_santarem = np.nanmean(MOD17._gpp(x, drivers_santarem[0], drivers_santarem[1], drivers_santarem[2], drivers_santarem[3]), axis=-1)\n",
    "    previsoes_santarem = pd.DataFrame(previsoes_santarem,index=index_santarem)\n",
    "    previsoes_santarem = ExponentialSmoothing(previsoes_santarem, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues\n",
    "\n",
    "    previsoes_peru = np.nanmean(MOD17._gpp(x, drivers_peru[0], drivers_peru[1], drivers_peru[2], drivers_peru[3]), axis=-1)\n",
    "    previsoes_peru = pd.DataFrame(previsoes_peru,index=index_peru)\n",
    "    previsoes_peru = ExponentialSmoothing(previsoes_peru, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues\n",
    "\n",
    "    gpp_kgc_8d_2009_teste = previsoes_santarem['2009':'2009'].resample('8D').sum()\n",
    "    gpp_kgc_8d_2010_teste = previsoes_santarem['2010':'2010'].resample('8D').sum()\n",
    "    gpp_kgc_8d_2011_teste = previsoes_santarem['2011':'2011'].resample('8D').sum()\n",
    "    previsoes_santarem = pd.concat([gpp_kgc_8d_2009_teste,gpp_kgc_8d_2010_teste,gpp_kgc_8d_2011_teste],axis=0) # previsões santarem\n",
    "\n",
    "    gpp_kgc_8d_2018_teste = previsoes_peru['2018':'2018'].resample('8D').sum()\n",
    "    gpp_kgc_8d_2019_teste = previsoes_peru['2019':'2019'].resample('8D').sum()\n",
    "    previsoes_peru = pd.concat([gpp_kgc_8d_2018_teste,gpp_kgc_8d_2019_teste],axis=0) # previsões peru\n",
    "\n",
    "    return previsoes_santarem, previsoes_peru\n",
    "\n",
    "class MyOutput(Output):\n",
    "    '''Classe utilizada para printar os dados da otimização em andamento pelo AG'''\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.LUEmax = Column(\"LUEmax\", width=13)\n",
    "        self.Tmin_min = Column(\"Tmin_min\", width=13)\n",
    "        self.Tmin_max = Column(\"Tmin_max\", width=13)\n",
    "        self.VPD_min = Column(\"VPD_min\", width=13)\n",
    "        self.VPD_max = Column(\"VPD_max\", width=13)\n",
    "        self.RMSE_Santarem = Column(\"RMSE Santarem\", width=13)\n",
    "        self.RMSE_Peru = Column(\"RMSE Peru\", width=13)\n",
    "        self.columns += [self.LUEmax, self.Tmin_min,self.Tmin_max,self.VPD_min,self.VPD_max,self.RMSE_Santarem,self.RMSE_Peru]\n",
    "\n",
    "\n",
    "    def update(self, algorithm):\n",
    "\n",
    "        global observacoes_santarem\n",
    "        global observacoes_peru\n",
    "\n",
    "        super().update(algorithm)\n",
    "        self.LUEmax.set(algorithm.pop.get(\"X\")[0][0])\n",
    "        self.Tmin_min.set(algorithm.pop.get(\"X\")[0][1])\n",
    "        self.Tmin_max.set(algorithm.pop.get(\"X\")[0][2])\n",
    "        self.VPD_min.set(algorithm.pop.get(\"X\")[0][3])\n",
    "        self.VPD_max.set(algorithm.pop.get(\"X\")[0][4])\n",
    "\n",
    "        previsoes_santarem, previsoes_peru = get_data(algorithm.pop.get(\"X\")[0])\n",
    "\n",
    "        rmse_santarem = np.round(np.sqrt(mean_squared_error(y_true=observacoes_santarem, y_pred=previsoes_santarem)),5)\n",
    "        rmse_peru = np.round(np.sqrt(mean_squared_error(y_true=observacoes_peru, y_pred=previsoes_peru)),5)\n",
    "\n",
    "        self.RMSE_Santarem.set(rmse_santarem)\n",
    "        self.RMSE_Peru.set(rmse_peru)\n",
    "\n",
    "\n",
    "class MyCustomProblem(ElementwiseProblem,):\n",
    "    '''Classe que implementa o problema a ser otimizado'''\n",
    "    def __init__(self):\n",
    "        super().__init__(\n",
    "            n_var=5, #número de variaveis a serem otimizadas\n",
    "            n_obj=1, #número de objetivos do GA\n",
    "            n_constr=1,\n",
    "            xl=np.array([0.3 / 1000, -35, 6, 60, 1500]),  #valores mínimos \n",
    "            xu=np.array([3.0 / 1000, -2, 30, 1000, 6500]),#valores máximos\n",
    "        )\n",
    "\n",
    "\n",
    "    def _evaluate(self, x, out, *args, **kwargs):\n",
    "        '''Função utilizada para cálculo da fitness de cada individuo da população do AG'''\n",
    "        global observacoes_santarem\n",
    "        global observacoes_peru\n",
    "\n",
    "        previsoes_santarem, previsoes_peru = get_data(x)\n",
    "\n",
    "        rmse_santarem = np.sqrt(mean_squared_error(y_true=observacoes_santarem, y_pred=previsoes_santarem))\n",
    "        rmse_peru = np.sqrt(mean_squared_error(y_true=observacoes_peru, y_pred=previsoes_peru))\n",
    "\n",
    "        comparar_mod_santarem = (rmse_santarem > rmse_mod_santarem)\n",
    "        comparar_mod_peru = (rmse_peru > rmse_mod_peru)\n",
    "\n",
    "        # Calcule a média das diferenças de RMSE entre o algoritmo e o modelo para ambas as regiões\n",
    "        mean_diff = np.mean([comparar_mod_santarem, comparar_mod_peru])\n",
    "\n",
    "        # Calcular a média aritmética dos valores RMSE para Santarém e Peru\n",
    "        mean_rmse = np.mean([rmse_santarem,rmse_peru])\n",
    "\n",
    "        # Armazenar na saída\n",
    "        out[\"F\"] = mean_rmse\n",
    "\n",
    "        out[\"G\"] = mean_diff\n",
    "\n",
    "class MyCallback(Callback):\n",
    "    '''Classe utilizada para salvar dados do GA sem precisar usar o parâmetro save_history que ocupa muita memória e deixa o AG mais lento'''   \n",
    "    def __init__(self) -> None:\n",
    "        super().__init__()\n",
    "        self.n_evals = []\n",
    "        self.opt = []\n",
    "\n",
    "    def notify(self, algorithm):\n",
    "        self.n_evals.append(algorithm.evaluator.n_eval)\n",
    "        self.opt.append(algorithm.opt[0].F)       \n",
    "\n",
    "\n",
    "# Defina o problema\n",
    "problem = MyCustomProblem()\n",
    "callback = MyCallback()\n",
    "\n",
    "# Defina o algoritmo GA\n",
    "algorithm = GA(\n",
    "    pop_size=1000,\n",
    "    eliminate_duplicates=True,\n",
    "    sampling=initial_population[:5],\n",
    "    mutation=PolynomialMutation(prob=0.5, eta=20),\n",
    "    crossover=SBX(prob=0.5,eta=20)\n",
    ")\n",
    "\n",
    "\n",
    "def timer_func(func):\n",
    "    # This function shows the execution time of\n",
    "    # the function object passed\n",
    "    def wrap_func(*args, **kwargs):\n",
    "        t1 = time()\n",
    "        result = func(*args, **kwargs)\n",
    "        t2 = time()\n",
    "        print(f'Function {func.__name__!r} executed in {(t2-t1):.4f}s')\n",
    "        return result\n",
    "    return wrap_func\n",
    "\n",
    "\n",
    "@timer_func\n",
    "def exec_model():\n",
    "    # Execute a otimização\n",
    "    res = minimize(ConstraintsAsPenalty(problem,penalty=50.0),\n",
    "                   algorithm,\n",
    "                   ('n_gen', 400),\n",
    "                   seed=1,\n",
    "                   output=MyOutput(),\n",
    "                   callback=callback,\n",
    "                   save_history=False,\n",
    "                   verbose=True)\n",
    "    return res\n",
    "\n",
    "res = exec_model()\n",
    "\n",
    "#Plotando o gráfico de convergência\n",
    "plt.title(\"Convergence\")\n",
    "plt.plot(callback.n_evals, callback.opt, \"--\")\n",
    "plt.ylabel(\"Média RMSEs\")\n",
    "plt.xlabel(\"N_Evals\")\n",
    "\n",
    "plt.show()\n",
    "plt.savefig(\"convergence.png\", dpi=300)\n",
    "previsoes_santarem_GA, previsoes_peru_GA = get_data(list(res.X))\n",
    "rmse_st = np.sqrt(mean_squared_error(y_true=observacoes_santarem,y_pred=previsoes_santarem_GA))\n",
    "rmse_pe = np.sqrt(mean_squared_error(y_true=observacoes_peru,y_pred=previsoes_peru_GA))\n",
    "\n",
    "print(f'Porcentagem de melhora: {-((rmse_pe - rmse_mod_peru) / rmse_mod_peru) * 100}')\n",
    "print(f'Porcentagem de melhora: {-((rmse_st - rmse_mod_santarem) / rmse_mod_santarem) * 100}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotando o gráfico de convergência em relação ao número de gerações\n",
    "plt.title(\"Convergence\")\n",
    "plt.plot(range(1,401), callback.opt, \"--\")\n",
    "plt.ylabel(\"Média RMSEs\")\n",
    "plt.xlabel(\"N_Evals\")\n",
    "\n",
    "plt.show()\n",
    "plt.savefig(\"convergence2.png\", dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Resultados"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Análise do GPP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Porcentagem de melhora: 5.037479460922268\n",
      "Porcentagem de melhora: 9.721157201631298\n"
     ]
    }
   ],
   "source": [
    "def get_data(x):\n",
    "    global observacoes_santarem\n",
    "    global observacoes_peru\n",
    "    previsoes_santarem = np.nanmean(MOD17._gpp(x, drivers_santarem[0], drivers_santarem[1], drivers_santarem[2], drivers_santarem[3]), axis=-1)\n",
    "    previsoes_santarem = pd.DataFrame(previsoes_santarem,index=index_santarem)\n",
    "    previsoes_santarem = ExponentialSmoothing(previsoes_santarem, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues\n",
    "\n",
    "    previsoes_peru = np.nanmean(MOD17._gpp(x, drivers_peru[0], drivers_peru[1], drivers_peru[2], drivers_peru[3]), axis=-1)\n",
    "    previsoes_peru = pd.DataFrame(previsoes_peru,index=index_peru)\n",
    "    previsoes_peru = ExponentialSmoothing(previsoes_peru, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues\n",
    "\n",
    "    gpp_kgc_8d_2009_teste = previsoes_santarem['2009':'2009'].resample('8D').sum()\n",
    "    gpp_kgc_8d_2010_teste = previsoes_santarem['2010':'2010'].resample('8D').sum()\n",
    "    gpp_kgc_8d_2011_teste = previsoes_santarem['2011':'2011'].resample('8D').sum()\n",
    "    previsoes_santarem = pd.concat([gpp_kgc_8d_2009_teste,gpp_kgc_8d_2010_teste,gpp_kgc_8d_2011_teste],axis=0) # previsões santarem\n",
    "\n",
    "    gpp_kgc_8d_2018_teste = previsoes_peru['2018':'2018'].resample('8D').sum()\n",
    "    gpp_kgc_8d_2019_teste = previsoes_peru['2019':'2019'].resample('8D').sum()\n",
    "    previsoes_peru = pd.concat([gpp_kgc_8d_2018_teste,gpp_kgc_8d_2019_teste],axis=0) # previsões peru\n",
    "\n",
    "    return previsoes_santarem, previsoes_peru\n",
    "\n",
    "parametros_padrao = initial_population\n",
    "melhor_params = [0.001726, -2.0, 23.99, 176.79, 6500.0] # Parâmetros Encontrados pelo Algoritmo Genético (Arredondados) [round(res.X[0],6)] + list(np.round(res.X[1:],2)) \n",
    "previsoes_santarem_GA, previsoes_peru_GA = get_data(melhor_params)\n",
    "rmse_st = np.sqrt(mean_squared_error(y_true=observacoes_santarem,y_pred=previsoes_santarem_GA))\n",
    "rmse_pe = np.sqrt(mean_squared_error(y_true=observacoes_peru,y_pred=previsoes_peru_GA))\n",
    "\n",
    "print(f'Porcentagem de melhora: {-((rmse_pe - rmse_mod_peru) / rmse_mod_peru) * 100}')\n",
    "print(f'Porcentagem de melhora: {-((rmse_st - rmse_mod_santarem) / rmse_mod_santarem) * 100}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Erro Absoluto Diário"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "tags": []
   },
   "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>Erro Diário $GPP_{MERRA-2}$ (Novo)</th>\n",
       "      <th>Erro Diário $GPP_{MOD17}$</th>\n",
       "      <th>Erro Diário $GPP_{MERRA-2}$</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>138.000000</td>\n",
       "      <td>138.000000</td>\n",
       "      <td>138.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>13.013979</td>\n",
       "      <td>15.015599</td>\n",
       "      <td>15.837828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>11.834473</td>\n",
       "      <td>12.411476</td>\n",
       "      <td>13.084553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.076658</td>\n",
       "      <td>0.161443</td>\n",
       "      <td>0.179949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4.480301</td>\n",
       "      <td>5.256445</td>\n",
       "      <td>6.196906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>9.754109</td>\n",
       "      <td>11.918580</td>\n",
       "      <td>12.455563</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>15.687689</td>\n",
       "      <td>22.359194</td>\n",
       "      <td>22.940316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>66.794937</td>\n",
       "      <td>66.816703</td>\n",
       "      <td>68.397932</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Erro Diário $GPP_{MERRA-2}$ (Novo)  Erro Diário $GPP_{MOD17}$  \\\n",
       "count                          138.000000                 138.000000   \n",
       "mean                            13.013979                  15.015599   \n",
       "std                             11.834473                  12.411476   \n",
       "min                              0.076658                   0.161443   \n",
       "25%                              4.480301                   5.256445   \n",
       "50%                              9.754109                  11.918580   \n",
       "75%                             15.687689                  22.359194   \n",
       "max                             66.794937                  66.816703   \n",
       "\n",
       "       Erro Diário $GPP_{MERRA-2}$  \n",
       "count                   138.000000  \n",
       "mean                     15.837828  \n",
       "std                      13.084553  \n",
       "min                       0.179949  \n",
       "25%                       6.196906  \n",
       "50%                      12.455563  \n",
       "75%                      22.940316  \n",
       "max                      68.397932  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>Erro Diário $GPP_{MERRA-2}$ (Novo)</th>\n",
       "      <th>Erro Diário $GPP_{MOD17}$</th>\n",
       "      <th>Erro Diário $GPP_{MERRA-2}$</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>92.000000</td>\n",
       "      <td>92.000000</td>\n",
       "      <td>92.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>15.355726</td>\n",
       "      <td>17.206737</td>\n",
       "      <td>18.179579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>12.918023</td>\n",
       "      <td>12.250556</td>\n",
       "      <td>14.331126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.238985</td>\n",
       "      <td>0.169084</td>\n",
       "      <td>0.352087</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.221893</td>\n",
       "      <td>6.612729</td>\n",
       "      <td>6.761037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>11.346356</td>\n",
       "      <td>15.018315</td>\n",
       "      <td>13.543690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>23.334623</td>\n",
       "      <td>25.928258</td>\n",
       "      <td>28.845596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>62.426419</td>\n",
       "      <td>58.268753</td>\n",
       "      <td>67.289038</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Erro Diário $GPP_{MERRA-2}$ (Novo)  Erro Diário $GPP_{MOD17}$  \\\n",
       "count                           92.000000                  92.000000   \n",
       "mean                            15.355726                  17.206737   \n",
       "std                             12.918023                  12.250556   \n",
       "min                              0.238985                   0.169084   \n",
       "25%                              5.221893                   6.612729   \n",
       "50%                             11.346356                  15.018315   \n",
       "75%                             23.334623                  25.928258   \n",
       "max                             62.426419                  58.268753   \n",
       "\n",
       "       Erro Diário $GPP_{MERRA-2}$  \n",
       "count                    92.000000  \n",
       "mean                     18.179579  \n",
       "std                      14.331126  \n",
       "min                       0.352087  \n",
       "25%                       6.761037  \n",
       "50%                      13.543690  \n",
       "75%                      28.845596  \n",
       "max                      67.289038  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gpp_alg_santarem = np.nanmean(MOD17._gpp(melhor_params,drivers_santarem[0], drivers_santarem[1], drivers_santarem[2], drivers_santarem[3]), axis = -1)\n",
    "gpp_alg_santarem = pd.DataFrame(gpp_alg_santarem,index=index_santarem)\n",
    "smoothed_gpp_alg_santarem = ExponentialSmoothing(gpp_alg_santarem, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues\n",
    "\n",
    "gpp_ebf_8d_2009_teste_santarem = smoothed_gpp_alg_santarem['2009':'2009'].resample('8D').sum()\n",
    "gpp_ebf_8d_2010_teste_santarem = smoothed_gpp_alg_santarem['2010':'2010'].resample('8D').sum()\n",
    "gpp_ebf_8d_2011_teste_santarem = smoothed_gpp_alg_santarem['2011':'2011'].resample('8D').sum()\n",
    "new_params_santarem = pd.concat([gpp_ebf_8d_2009_teste_santarem,gpp_ebf_8d_2010_teste_santarem,gpp_ebf_8d_2011_teste_santarem],axis=0)\n",
    "\n",
    "\n",
    "gpp_alg_peru = np.nanmean(MOD17._gpp(melhor_params,drivers_peru[0], drivers_peru[1], drivers_peru[2], drivers_peru[3]), axis = -1)\n",
    "gpp_alg_peru = pd.DataFrame(gpp_alg_peru,index=index_peru)\n",
    "smoothed_gpp_alg_peru = ExponentialSmoothing(gpp_alg_peru, initialization_method=\"heuristic\",freq='D').fit(optimized=True).fittedvalues\n",
    "\n",
    "\n",
    "gpp_ebf_8d_2018_teste = smoothed_gpp_alg_peru['2018':'2018'].resample('8D').sum()\n",
    "gpp_ebf_8d_2019_teste = smoothed_gpp_alg_peru['2019':'2019'].resample('8D').sum()\n",
    "new_params_peru = pd.concat([gpp_ebf_8d_2018_teste,gpp_ebf_8d_2019_teste],axis=0)\n",
    "\n",
    "\n",
    "erro_absoluto_modelo1 = abs(observacoes_santarem - new_params_santarem)\n",
    "erro_absoluto_modelo2 = abs(observacoes_santarem - santarem_gpp_mod)\n",
    "erro_absoluto_modelo3_santarem = abs(observacoes_santarem - smoothed_gpp_algpad_santarem)\n",
    "\n",
    "# Calcula o erro absoluto diário para cada modelo\n",
    "eda_modelo1 = erro_absoluto_modelo1\n",
    "eda_modelo2 = erro_absoluto_modelo2\n",
    "eda_modelo3_santarem = erro_absoluto_modelo3_santarem\n",
    "\n",
    "\n",
    "# Cria um DataFrame para armazenar os erros absolutos diários de ambos os modelos\n",
    "erro_absoluto_santarem = pd.DataFrame({\n",
    "    'Erro Diário $GPP_{MERRA-2}$ (Novo)': eda_modelo1,\n",
    "    'Erro Diário $GPP_{MOD17}$': eda_modelo2,\n",
    "    'Erro Diário $GPP_{MERRA-2}$': eda_modelo3_santarem\n",
    "})\n",
    "\n",
    "\n",
    "erro_absoluto_modelo1 = abs(observacoes_peru - new_params_peru)\n",
    "erro_absoluto_modelo2 = abs(observacoes_peru - peru_gpp_mod)\n",
    "erro_absoluto_modelo3_peru = abs(observacoes_peru - smoothed_gpp_algpad_peru)\n",
    "\n",
    "# Calcula o RMSE diário para cada modelo\n",
    "eda_modelo1 = erro_absoluto_modelo1\n",
    "eda_modelo2 = erro_absoluto_modelo2\n",
    "eda_modelo3_peru = erro_absoluto_modelo3_peru\n",
    "\n",
    "# Cria um DataFrame para armazenar os RMSE diários de ambos os modelos\n",
    "erro_absoluto_peru = pd.DataFrame({\n",
    "    'Erro Diário $GPP_{MERRA-2}$ (Novo)': eda_modelo1,\n",
    "    'Erro Diário $GPP_{MOD17}$': eda_modelo2,\n",
    "    'Erro Diário $GPP_{MERRA-2}$': eda_modelo3_peru\n",
    "})\n",
    "\n",
    "\n",
    "display(erro_absoluto_santarem.describe())\n",
    "display(erro_absoluto_peru.describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Boxplot Mensal das Fontes de GPP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Definindo os dados mensais de Santarém\n",
    "observacoes_santarem_monthly = observacoes_santarem.resample('M').mean()\n",
    "new_params_santarem_monthly = new_params_santarem.resample('M').mean()\n",
    "santarem_gpp_mod_monthly = santarem_gpp_mod.resample('M').mean()\n",
    "smoothed_gpp_algpad_santarem_monthly = smoothed_gpp_algpad_santarem.resample('M').mean()\n",
    "\n",
    "# Criando DataFrame para Santarém com os dados resampleados\n",
    "df_santarem = pd.DataFrame({\n",
    "    '$GPP_{Torre}$': observacoes_santarem_monthly,\n",
    "    '$GPP_{MERRA-2}$ (Novo)': new_params_santarem_monthly,\n",
    "    '$GPP_{MERRA-2}$': smoothed_gpp_algpad_santarem_monthly,\n",
    "    '$GPP_{MOD17}$': santarem_gpp_mod_monthly,\n",
    "})\n",
    "\n",
    "# Resetando o índice para ter uma coluna de datas\n",
    "df_santarem.reset_index(inplace=True)\n",
    "\n",
    "# Derretendo o DataFrame para ter todas as variáveis em uma única coluna\n",
    "df_santarem_melted = df_santarem.melt(id_vars='index', var_name='Medição de GPP', value_name='Valor')\n",
    "\n",
    "# Obtendo o ano e o mês de cada data\n",
    "df_santarem_melted['Ano'] = df_santarem_melted['index'].dt.year\n",
    "df_santarem_melted['Mes'] = df_santarem_melted['index'].dt.strftime('%b')\n",
    "\n",
    "\n",
    "# Definindo os dados mensais de Peru\n",
    "observacoes_peru_monthly = observacoes_peru.resample('M').mean()\n",
    "new_params_peru_monthly = new_params_peru.resample('M').mean()\n",
    "peru_gpp_mod_monthly = peru_gpp_mod.resample('M').mean()\n",
    "smoothed_gpp_algpad_peru_monthly = smoothed_gpp_algpad_peru.resample('M').mean()\n",
    "\n",
    "# Criando DataFrame para Peru com os dados resampleados\n",
    "df_peru = pd.DataFrame({\n",
    "    '$GPP_{Torre}$': observacoes_peru_monthly,\n",
    "    '$GPP_{MERRA-2}$ (Novo)': new_params_peru_monthly,\n",
    "    '$GPP_{MERRA-2}$': smoothed_gpp_algpad_peru_monthly,\n",
    "    '$GPP_{MOD17}$': peru_gpp_mod_monthly\n",
    "})\n",
    "\n",
    "# Resetando o índice para ter uma coluna de datas\n",
    "df_peru.reset_index(inplace=True)\n",
    "\n",
    "# Derretendo o DataFrame para ter todas as variáveis em uma única coluna\n",
    "df_peru_melted = df_peru.melt(id_vars='index', var_name='Medição de GPP', value_name='Valor')\n",
    "\n",
    "# Obtendo o ano e o mês de cada data\n",
    "df_peru_melted['Ano'] = df_peru_melted['index'].dt.year\n",
    "df_peru_melted['Mes'] = df_peru_melted['index'].dt.strftime('%b')\n",
    "\n",
    "\n",
    "# Plotando os boxplots em subplots\n",
    "fig, axes = plt.subplots(1, 2, figsize=(20, 6))  # Criando subplots com 2 linhas e 1 coluna\n",
    "\n",
    "# Boxplot para Santarém\n",
    "sns.boxplot(data=df_santarem_melted, x='Mes', y='Valor', hue='Medição de GPP', dodge=True, ax=axes[0])\n",
    "axes[0].set_title('(a) Box Plot Mensal - Santarém')  # Definindo título\n",
    "axes[0].set_ylabel('GPP ($g : C : m^{-2} : 8d^{-1}$)')  # Definindo rótulo do eixo y\n",
    "axes[0].set_xlabel('Mês')  # Definindo rótulo do eixo x\n",
    "axes[0].legend(loc='upper center', fontsize=15, ncol=2)  # Definindo legenda\n",
    "\n",
    "# Boxplot para Peru\n",
    "sns.boxplot(data=df_peru_melted, x='Mes', y='Valor', hue='Medição de GPP', dodge=True, ax=axes[1])\n",
    "axes[1].set_title('(b) Box Plot Mensal - Peru')  # Definindo título\n",
    "axes[1].set_ylabel('')  # Definindo rótulo do eixo y\n",
    "axes[1].set_xlabel('Mês')  # Definindo rótulo do eixo x\n",
    "axes[1].legend(loc='upper center', fontsize=15, ncol=2)  # Definindo legenda\n",
    "\n",
    "plt.tight_layout()  # Ajustando layout\n",
    "#plt.savefig('BoxPlot_Santarem_Peru.png', dpi=300, bbox_inches='tight', pad_inches=0.2)  # Salvando figura\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Derretendo o DataFrame para ter todas as variáveis em uma única colun\n",
    "df_santarem_melted = df_santarem.drop(['$GPP_{MERRA-2}$ (Novo)'],axis=1).melt(id_vars='index', var_name='Medição de GPP', value_name='Valor')\n",
    "\n",
    "# Obtendo o ano e o mês de cada data\n",
    "df_santarem_melted['Ano'] = df_santarem_melted['index'].dt.year\n",
    "df_santarem_melted['Mes'] = df_santarem_melted['index'].dt.strftime('%b')\n",
    "\n",
    "\n",
    "# Derretendo o DataFrame para ter todas as variáveis em uma única coluna\n",
    "df_peru_melted = df_peru.drop(['$GPP_{MERRA-2}$ (Novo)'],axis=1).melt(id_vars='index', var_name='Medição de GPP', value_name='Valor')\n",
    "\n",
    "# Obtendo o ano e o mês de cada data\n",
    "df_peru_melted['Ano'] = df_peru_melted['index'].dt.year\n",
    "df_peru_melted['Mes'] = df_peru_melted['index'].dt.strftime('%b')\n",
    "\n",
    "# Plotando os boxplots em subplots\n",
    "fig, axes = plt.subplots(1, 2, figsize=(20, 6))  # Criando subplots com 2 linhas e 1 coluna\n",
    "\n",
    "# Boxplot para Santarém\n",
    "sns.boxplot(data=df_santarem_melted, x='Mes', y='Valor', hue='Medição de GPP', dodge=True, ax=axes[0])\n",
    "axes[0].set_title('(a) Box Plot Mensal - Santarém')  # Definindo título\n",
    "axes[0].set_ylabel('GPP ($g : C : m^{-2} : 8d^{-1}$)')  # Definindo rótulo do eixo y\n",
    "axes[0].set_xlabel('Mês')  # Definindo rótulo do eixo x\n",
    "axes[0].legend(loc='upper center', fontsize=15, ncol=2)  # Definindo legenda\n",
    "\n",
    "# Boxplot para Peru\n",
    "sns.boxplot(data=df_peru_melted, x='Mes', y='Valor', hue='Medição de GPP', dodge=True, ax=axes[1])\n",
    "axes[1].set_title('(b) Box Plot Mensal - Peru')  # Definindo título\n",
    "axes[1].set_ylabel('')  # Definindo rótulo do eixo y\n",
    "axes[1].set_xlabel('Mês')  # Definindo rótulo do eixo x\n",
    "axes[1].legend(loc='upper center', fontsize=15, ncol=2)  # Definindo legenda\n",
    "\n",
    "plt.tight_layout()  # Ajustando layout\n",
    "#plt.savefig('BoxPlot_Santarem_Peru.png', dpi=300, bbox_inches='tight', pad_inches=0.2)  # Salvando figura\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Análise do NPP"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Cálculo do NPP (Peru)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "index_santarem = pd.date_range(start='2009-01-01', end='2011-12-31', freq='D')\n",
    "index_peru = pd.date_range(start='2018-01-01', end='2019-12-31', freq='D')\n",
    "\n",
    "# Leitura e pré-processamento dos dados de NPP (kgC/m²/year)\n",
    "npp_mod = pd.read_csv('mod_peru/filtered_scaled_Npp_500m.csv', header=None,na_values='F')\n",
    "npp_mod.index = npp_mod[2].apply(modis_date)  # Convertendo a data no formato MODIS\n",
    "\n",
    "# Arrumando o dataframe\n",
    "# ================================\n",
    "npp_mod = npp_mod.drop([0,1,2,3,4], axis=1)\n",
    "npp_mod.columns = range(len(npp_mod.columns))\n",
    "npp_modis_peru = npp_mod[144]\n",
    "\n",
    "\n",
    "am_D = pd.read_csv('amf_pe/AMF_PE-QFR_FLUXNET_FULLSET_DD_2018-2019_3-5.csv',na_values=-9999) # Dados da torre pro mesmo dia\n",
    "am_D.index = am_D['TIMESTAMP'].apply(lambda x:datetime.strptime(str(x), '%Y%m%d'))\n",
    "#print([column for column in am_D.columns if 'NEE' in column])\n",
    "temp_2 = am_D[['NEE_VUT_MEAN','RECO_DT_VUT_MEAN']].copy()\n",
    "nep = -temp_2['NEE_VUT_MEAN'].copy()\n",
    "reco = temp_2['RECO_DT_VUT_MEAN'].copy()\n",
    "npp_peru = (nep + (0.3919 * reco)).resample('Y').sum() # Cálculo do NPP para eddy covariance\n",
    "\n",
    "# Leitura e pré-processamento dos dados de LAI\n",
    "lai_peru = pd.read_csv('mod_peru/filtered_scaled_Lai_500m.csv', header=None,na_values='F') # unidade: m²/m²\n",
    "lai_peru.index = lai_peru[2].apply(modis_date)  # Convertendo a data no formato MODIS\n",
    "\n",
    "# Arrumando o dataframe\n",
    "# ================================\n",
    "lai_peru = lai_peru.drop([0,1,2,3,4], axis=1)\n",
    "lai_peru.columns = range(len(lai_peru.columns))\n",
    "lai_peru = lai_peru[144]\n",
    "\n",
    "# ================================\n",
    "\n",
    "lai_peru = lai_peru.resample('D').interpolate('linear')['2018 ':'2019']\n",
    "\n",
    "\n",
    "npp_algoritmo_peru = MOD17._npp(params=(melhor_params + list(parametros_padrao)[5:]),fpar=drivers_peru[0],tmin=drivers_peru[1],vpd=drivers_peru[2],par=drivers_peru[3],tmean=tmean_peru[:][...,None],years=index_peru.year.values[:][...,None],lai=lai_peru.values[:][...,None])\n",
    "npp_algoritmopad_peru = MOD17._npp(params=parametros_padrao,fpar=drivers_peru[0],tmin=drivers_peru[1],vpd=drivers_peru[2],par=drivers_peru[3],tmean=tmean_peru[:][...,None],years=index_peru.year.values[:][...,None],lai=lai_peru.values[:][...,None])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Cálculo do NPP (Santarém)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Leitura e pré-processamento dos dados de NPP (kgC/m²/year)\n",
    "npp_mod = pd.read_csv('santarem_mod/filtered_scaled_Npp_500m.csv', header=None,na_values='F')\n",
    "npp_mod.index = npp_mod[2].apply(modis_date)  # Convertendo a data no formato MODIS\n",
    "\n",
    "# Arrumando o dataframe\n",
    "# ================================\n",
    "npp_mod = npp_mod.drop([0,1,2,3,4], axis=1)\n",
    "npp_mod.columns = range(len(npp_mod.columns))\n",
    "npp_modis = npp_mod[144]\n",
    "npp_modis_st = npp_modis[['2009-01-01','2010-01-01','2011-01-01']]\n",
    "\n",
    "ameriflux = pd.read_csv('./santarem_torre/ameriflux-santarem-dados-original.csv',na_values=-9999) # Dados da AMERIFLUX\n",
    "ameriflux.loc[0, 'Time'] = '00:00:00'\n",
    "ameriflux.index = pd.to_datetime(ameriflux['Date'] + ' ' + ameriflux['Time'], format='%Y-%m-%d %H:%M:%S')\n",
    "\n",
    "nee = ameriflux['NEE_uStar_f'].copy()\n",
    "reco = ameriflux['Reco_uStar'].copy()\n",
    "gpp = ameriflux['GPP_uStar_f'].copy()\n",
    "\n",
    "nee_kgc = nee.apply(to_kgC)\n",
    "reco_kgc_total = reco.apply(to_kgC)\n",
    "npp_santarem = (-nee_kgc + (0.3919 * reco_kgc_total)).resample('Y').sum()\n",
    "\n",
    "# Leitura e pré-processamento dos dados de LAI\n",
    "lai_santarem = pd.read_csv('santarem_mod/filtered_scaled_Lai_500m.csv', header=None,na_values='F') # unidade: m²/m²\n",
    "lai_santarem.index = lai_santarem[2].apply(modis_date)  # Convertendo a data no formato MODIS\n",
    "\n",
    "# Arrumando o dataframe\n",
    "# ================================\n",
    "lai_santarem = lai_santarem.drop([0,1,2,3,4], axis=1)\n",
    "lai_santarem.columns = range(len(lai_santarem.columns))\n",
    "lai_santarem = lai_santarem[144]\n",
    "\n",
    "# ================================\n",
    "\n",
    "lai_santarem = lai_santarem.resample('D').interpolate('linear')['2009-01-01':'2011-12-31']\n",
    "\n",
    "npp_algoritmo_st = MOD17._npp(params=(melhor_params + list(parametros_padrao)[5:]),fpar=drivers_santarem[0],tmin=drivers_santarem[1],vpd=drivers_santarem[2],par=drivers_santarem[3],tmean=tmean_santarem[:][...,None],years=index_santarem.year.values[:][...,None],lai=lai_santarem.values[:][...,None])\n",
    "npp_algoritmopad_st = MOD17._npp(params=parametros_padrao,fpar=drivers_santarem[0],tmin=drivers_santarem[1],vpd=drivers_santarem[2],par=drivers_santarem[3],tmean=tmean_santarem[:][...,None],years=index_santarem.year.values[:][...,None],lai=lai_santarem.values[:][...,None])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Resultados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Defina a largura das barras\n",
    "largura_barra = 0.2\n",
    "\n",
    "# Obter uma lista de anos únicos presentes nos dados\n",
    "anos_unicos_peru = observacoes_peru.index.year.unique()\n",
    "anos_unicos_santarem = observacoes_santarem.index.year.unique()\n",
    "\n",
    "# Convertendo os anos em formato de string\n",
    "anos_unicos_str_peru = [str(ano) for ano in anos_unicos_peru]\n",
    "anos_unicos_str_santarem = [str(ano) for ano in anos_unicos_santarem]\n",
    "\n",
    "# Defina os deslocamentos das barras para os diferentes conjuntos de dados\n",
    "deslocamento_peru = np.arange(len(anos_unicos_peru))\n",
    "deslocamento_santarem = np.arange(len(anos_unicos_santarem))\n",
    "\n",
    "# Criação da figura e dos subplots\n",
    "fig, axs = plt.subplots(1, 2, figsize=(14, 6))\n",
    "\n",
    "\n",
    "# Plotando as barras agrupadas para cada conjunto de dados (Santarem)\n",
    "axs[0].bar(deslocamento_santarem - largura_barra, npp_modis_st.values, width=largura_barra, color='red', alpha=0.5, label='$NPP_{MOD17}$')\n",
    "axs[0].bar(deslocamento_santarem, npp_santarem, width=largura_barra, color='blue', alpha=0.5, label='$NPP_{EC}$')\n",
    "axs[0].bar(deslocamento_santarem + largura_barra, (npp_algoritmo_st / 1000).reshape(3), width=largura_barra, color='orange', alpha=0.5, label='$NPP_{MERRA-2}$ (Novo)')\n",
    "axs[0].bar(deslocamento_santarem + 2*largura_barra, (npp_algoritmopad_st / 1000).reshape(3), width=largura_barra, color='green', alpha=0.5, label='$NPP_{MERRA-2}$')\n",
    "axs[0].set_ylabel('NPP ($kg \\: C \\: m^{-2} \\: y^{-1}$)')\n",
    "\n",
    "# Plotando as barras agrupadas para cada conjunto de dados (Peru)\n",
    "axs[1].bar(deslocamento_peru - largura_barra, npp_modis_peru['2018':'2019'].values, width=largura_barra, color='red', alpha=0.5, label='$NPP_{MOD17}$')\n",
    "axs[1].bar(deslocamento_peru, npp_peru / 1000, width=largura_barra, color='blue', alpha=0.5, label='$NPP_{EC}$')\n",
    "axs[1].bar(deslocamento_peru + largura_barra, (npp_algoritmo_peru / 1000).reshape(2), width=largura_barra, color='orange', alpha=0.5, label='$NPP_{MERRA-2}$ (Novo)')\n",
    "axs[1].bar(deslocamento_peru + 2*(largura_barra), (npp_algoritmopad_peru / 1000).reshape(2), width=largura_barra, color='green', alpha=0.5, label='$NPP_{MERRA-2}$')\n",
    "\n",
    "for i, valor in enumerate(npp_modis_peru['2018':'2019'].values):\n",
    "    axs[1].text(i - largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_peru / 1000)):\n",
    "    axs[1].text(i, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_algoritmo_peru / 1000).reshape(2)):\n",
    "    axs[1].text(i + largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_algoritmopad_peru / 1000).reshape(2)):\n",
    "    axs[1].text(i + 2*(largura_barra), valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "\n",
    "# Adicionando legendas e rótulos (Peru)\n",
    "axs[1].legend()\n",
    "axs[1].set_ylim(top=2.5)\n",
    "axs[1].set_xlabel('Ano')\n",
    "axs[1].set_xticks(deslocamento_peru)\n",
    "axs[1].set_xticklabels(anos_unicos_str_peru)\n",
    "\n",
    "\n",
    "mae_peru = mean_absolute_error(npp_peru / 1000, npp_algoritmo_peru / 1000)\n",
    "\n",
    "mae_santarem = mean_absolute_error(npp_santarem, npp_algoritmo_st / 1000)\n",
    "\n",
    "mae_pad_peru = mean_absolute_error(npp_peru / 1000, npp_algoritmopad_peru / 1000)\n",
    "\n",
    "mae_pad_santarem = mean_absolute_error(npp_santarem, npp_algoritmopad_st / 1000)\n",
    "\n",
    "mae_modis_peru = mean_absolute_error(npp_peru / 1000, npp_modis_peru['2018':'2019'].values)\n",
    "\n",
    "mae_modis_santarem = mean_absolute_error(npp_santarem, npp_modis_st.values)\n",
    "\n",
    "axs[1].text(0.25, 0.9, 'MAE ($NPP_{MODIS}$):' + ' ' + f'{mae_modis_peru:.2f}', ha='center', va='top', transform=axs[1].transAxes, fontsize=10, color='red')\n",
    "axs[0].text(0.25, 0.9, 'MAE ($NPP_{MODIS}$):' + ' ' + f'{mae_modis_santarem:.2f}', ha='center', va='top', transform=axs[0].transAxes, fontsize=10, color='red')\n",
    "\n",
    "\n",
    "axs[1].text(0.25, 0.95, 'MAE ($NPP_{MERRA-2}$) (Novo):' + ' ' + f'{mae_peru:.3f}', ha='center', va='top', transform=axs[1].transAxes, fontsize=10, color='orange')\n",
    "axs[0].text(0.25, 0.95, 'MAE ($NPP_{MERRA-2}$) (Novo):' + ' ' + f'{mae_santarem:.3f}', ha='center', va='top', transform=axs[0].transAxes, fontsize=10, color='orange')\n",
    "\n",
    "for i, valor in enumerate(npp_modis_st.values):\n",
    "    axs[0].text(i - largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_santarem)):\n",
    "    axs[0].text(i, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_algoritmo_st / 1000).reshape(3)):\n",
    "    axs[0].text(i + largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_algoritmopad_st / 1000).reshape(3)):\n",
    "    axs[0].text(i + 2*largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "\n",
    "# Adicionando legendas e rótulos (Santarem)\n",
    "axs[0].legend()\n",
    "axs[0].set_ylim(top=2.5)\n",
    "axs[0].set_xlabel('Ano')\n",
    "\n",
    "axs[0].set_xticks(deslocamento_santarem)\n",
    "axs[0].set_xticklabels(anos_unicos_str_santarem)\n",
    "\n",
    "#plt.savefig('npp_otimizado.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Criação da figura e dos subplots\n",
    "fig, axs = plt.subplots(1, 2, figsize=(14, 6))\n",
    "\n",
    "\n",
    "# Plotando as barras agrupadas para cada conjunto de dados (Santarem)\n",
    "axs[0].bar(deslocamento_santarem - largura_barra, npp_modis_st.values, width=largura_barra, color='red', alpha=0.5, label='$NPP_{MOD17}$')\n",
    "axs[0].bar(deslocamento_santarem, npp_santarem, width=largura_barra, color='blue', alpha=0.5, label='$NPP_{EC}$')\n",
    "axs[0].bar(deslocamento_santarem + largura_barra, (npp_algoritmopad_st / 1000).reshape(3), width=largura_barra, color='orange', alpha=0.5, label='$NPP_{MERRA-2}$ (Novo)')\n",
    "axs[0].set_ylabel('NPP ($kg \\: C \\: m^{-2} \\: y^{-1}$)')\n",
    "\n",
    "# Plotando as barras agrupadas para cada conjunto de dados (Peru)\n",
    "axs[1].bar(deslocamento_peru - largura_barra, npp_modis_peru['2018':'2019'].values, width=largura_barra, color='red', alpha=0.5, label='$NPP_{MOD17}$')\n",
    "axs[1].bar(deslocamento_peru, npp_peru / 1000, width=largura_barra, color='blue', alpha=0.5, label='$NPP_{EC}$')\n",
    "axs[1].bar(deslocamento_peru + largura_barra, (npp_algoritmopad_peru / 1000).reshape(2), width=largura_barra, color='orange', alpha=0.5, label='$NPP_{MERRA-2}$ (Novo)')\n",
    "\n",
    "for i, valor in enumerate(npp_modis_peru['2018':'2019'].values):\n",
    "    axs[1].text(i - largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_peru / 1000)):\n",
    "    axs[1].text(i, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_algoritmopad_peru / 1000).reshape(2)):\n",
    "    axs[1].text(i + largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "\n",
    "# Adicionando legendas e rótulos (Peru)\n",
    "axs[1].legend()\n",
    "axs[1].set_ylim(top=2.5)\n",
    "axs[1].set_xlabel('Ano')\n",
    "axs[1].set_xticks(deslocamento_peru)\n",
    "axs[1].set_xticklabels(anos_unicos_str_peru)\n",
    "\n",
    "axs[1].text(0.25, 0.9, 'MAE ($NPP_{MODIS}$):' + ' ' + f'{mae_modis_peru:.2f}', ha='center', va='top', transform=axs[1].transAxes, fontsize=10, color='red')\n",
    "axs[0].text(0.25, 0.9, 'MAE ($NPP_{MODIS}$):' + ' ' + f'{mae_modis_santarem:.2f}', ha='center', va='top', transform=axs[0].transAxes, fontsize=10, color='red')\n",
    "\n",
    "\n",
    "axs[1].text(0.25, 0.95, 'MAE ($NPP_{MERRA-2}$):' + ' ' + f'{mae_pad_peru:.3f}', ha='center', va='top', transform=axs[1].transAxes, fontsize=10, color='orange')\n",
    "axs[0].text(0.25, 0.95, 'MAE ($NPP_{MERRA-2}$):' + ' ' + f'{mae_pad_santarem:.3f}', ha='center', va='top', transform=axs[0].transAxes, fontsize=10, color='orange')\n",
    "\n",
    "for i, valor in enumerate(npp_modis_st.values):\n",
    "    axs[0].text(i - largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_santarem)):\n",
    "    axs[0].text(i, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "for i, valor in enumerate((npp_algoritmopad_st / 1000).reshape(3)):\n",
    "    axs[0].text(i + largura_barra, valor, str(round(valor, 2)), ha='center', va='bottom', color='black')\n",
    "\n",
    "# Adicionando legendas e rótulos (Santarem)\n",
    "axs[0].legend()\n",
    "axs[0].set_ylim(top=2.5)\n",
    "axs[0].set_xlabel('Ano')\n",
    "\n",
    "axs[0].set_xticks(deslocamento_santarem)\n",
    "axs[0].set_xticklabels(anos_unicos_str_santarem)\n",
    "\n",
    "#plt.savefig('npp_original.png')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "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": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}