scaling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a0a019f7",
   "metadata": {},
   "source": [
    "# Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a375d6d",
   "metadata": {},
   "source": [
    "## biomass_history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "158d743e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-22 19:57:17.575461: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2023-08-22 19:57:18.854747: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from tensorflow.keras.layers import Dense,Input,LSTM, Flatten, ConvLSTM2D\n",
    "from tensorflow.keras.models import Sequential,Model\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "33160e79",
   "metadata": {},
   "outputs": [],
   "source": [
    "biomass_history = pd.read_csv('dataset/Biomass_History.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d424c702",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index        0\n",
       "Latitude     0\n",
       "Longitude    0\n",
       "2010         0\n",
       "2011         0\n",
       "2012         0\n",
       "2013         0\n",
       "2014         0\n",
       "2015         0\n",
       "2016         0\n",
       "2017         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Check if there are any null values\n",
    "biomass_history .isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "15239056",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2418 entries, 0 to 2417\n",
      "Data columns (total 11 columns):\n",
      " #   Column     Non-Null Count  Dtype  \n",
      "---  ------     --------------  -----  \n",
      " 0   Index      2418 non-null   int64  \n",
      " 1   Latitude   2418 non-null   float64\n",
      " 2   Longitude  2418 non-null   float64\n",
      " 3   2010       2418 non-null   float64\n",
      " 4   2011       2418 non-null   float64\n",
      " 5   2012       2418 non-null   float64\n",
      " 6   2013       2418 non-null   float64\n",
      " 7   2014       2418 non-null   float64\n",
      " 8   2015       2418 non-null   float64\n",
      " 9   2016       2418 non-null   float64\n",
      " 10  2017       2418 non-null   float64\n",
      "dtypes: float64(10), int64(1)\n",
      "memory usage: 207.9 KB\n"
     ]
    }
   ],
   "source": [
    "biomass_history.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "fff66148",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='2010', ylabel='Count'>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# df.head()\n",
    "sns.histplot(biomass_history,x='2010')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "683a2ed3",
   "metadata": {},
   "outputs": [],
   "source": [
    "#converting float64 to float32 and int64 to int16\n",
    "columns = list(biomass_history.columns)\n",
    "for column in columns:\n",
    "    if(type(biomass_history[column][0]) == np.float64):\n",
    "        biomass_history[column] = biomass_history[column].astype('float16')\n",
    "    if(type(biomass_history[column][0]) == np.int64):\n",
    "        biomass_history[column] = biomass_history[column].astype('int16')\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "354ad923",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Index</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>2010</th>\n",
       "      <th>2011</th>\n",
       "      <th>2012</th>\n",
       "      <th>2013</th>\n",
       "      <th>2014</th>\n",
       "      <th>2015</th>\n",
       "      <th>2016</th>\n",
       "      <th>2017</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>24.671875</td>\n",
       "      <td>71.3125</td>\n",
       "      <td>8.476562</td>\n",
       "      <td>8.867188</td>\n",
       "      <td>9.203125</td>\n",
       "      <td>6.023438</td>\n",
       "      <td>10.789062</td>\n",
       "      <td>6.648438</td>\n",
       "      <td>7.386719</td>\n",
       "      <td>5.179688</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>24.671875</td>\n",
       "      <td>71.4375</td>\n",
       "      <td>24.031250</td>\n",
       "      <td>28.546875</td>\n",
       "      <td>25.859375</td>\n",
       "      <td>21.640625</td>\n",
       "      <td>34.406250</td>\n",
       "      <td>27.359375</td>\n",
       "      <td>40.437500</td>\n",
       "      <td>42.125000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>24.671875</td>\n",
       "      <td>71.5000</td>\n",
       "      <td>44.843750</td>\n",
       "      <td>66.125000</td>\n",
       "      <td>56.968750</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>70.937500</td>\n",
       "      <td>42.531250</td>\n",
       "      <td>59.187500</td>\n",
       "      <td>73.187500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>24.671875</td>\n",
       "      <td>71.5625</td>\n",
       "      <td>59.968750</td>\n",
       "      <td>80.812500</td>\n",
       "      <td>78.937500</td>\n",
       "      <td>63.156250</td>\n",
       "      <td>93.500000</td>\n",
       "      <td>70.187500</td>\n",
       "      <td>74.562500</td>\n",
       "      <td>101.062500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>24.671875</td>\n",
       "      <td>71.6250</td>\n",
       "      <td>14.656250</td>\n",
       "      <td>19.328125</td>\n",
       "      <td>21.921875</td>\n",
       "      <td>17.906250</td>\n",
       "      <td>19.531250</td>\n",
       "      <td>19.171875</td>\n",
       "      <td>16.531250</td>\n",
       "      <td>26.093750</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Index   Latitude  Longitude       2010       2011       2012       2013  \\\n",
       "0      0  24.671875    71.3125   8.476562   8.867188   9.203125   6.023438   \n",
       "1      1  24.671875    71.4375  24.031250  28.546875  25.859375  21.640625   \n",
       "2      2  24.671875    71.5000  44.843750  66.125000  56.968750  53.000000   \n",
       "3      3  24.671875    71.5625  59.968750  80.812500  78.937500  63.156250   \n",
       "4      4  24.671875    71.6250  14.656250  19.328125  21.921875  17.906250   \n",
       "\n",
       "        2014       2015       2016        2017  \n",
       "0  10.789062   6.648438   7.386719    5.179688  \n",
       "1  34.406250  27.359375  40.437500   42.125000  \n",
       "2  70.937500  42.531250  59.187500   73.187500  \n",
       "3  93.500000  70.187500  74.562500  101.062500  \n",
       "4  19.531250  19.171875  16.531250   26.093750  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "biomass_history.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "568759ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Scaling the columns from 2010 to 2017\n",
    "\n",
    "ct = ColumnTransformer([('std',MinMaxScaler(),[str(x) for x in np.arange(2010,2018)]),\n",
    "                       ],remainder='passthrough')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "28b2ce56",
   "metadata": {},
   "outputs": [],
   "source": [
    "transformed_df = ct.fit_transform(biomass_history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a6349eb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "transformed_df= pd.DataFrame(transformed_df,columns=['2010','2011','2012','2013' ,'2014','2015','2016','2017','Index','Latitude','Longitude'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b93c8d88",
   "metadata": {},
   "outputs": [],
   "source": [
    "transformed_df = transformed_df[columns]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "428baeb3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='2010', ylabel='Count'>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(transformed_df, x='2010')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "aeb79721",
   "metadata": {},
   "outputs": [],
   "source": [
    "biomass_history_scaled = transformed_df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "3b06303e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Index</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>2010</th>\n",
       "      <th>2011</th>\n",
       "      <th>2012</th>\n",
       "      <th>2013</th>\n",
       "      <th>2014</th>\n",
       "      <th>2015</th>\n",
       "      <th>2016</th>\n",
       "      <th>2017</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>24.671875</td>\n",
       "      <td>71.3125</td>\n",
       "      <td>0.009598</td>\n",
       "      <td>0.011497</td>\n",
       "      <td>0.010551</td>\n",
       "      <td>0.004887</td>\n",
       "      <td>0.012367</td>\n",
       "      <td>0.008621</td>\n",
       "      <td>0.014374</td>\n",
       "      <td>0.005863</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Index   Latitude  Longitude      2010      2011      2012      2013  \\\n",
       "0    0.0  24.671875    71.3125  0.009598  0.011497  0.010551  0.004887   \n",
       "\n",
       "       2014      2015      2016      2017  \n",
       "0  0.012367  0.008621  0.014374  0.005863  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#how does the biomass history look like for a particualr coordinate\n",
    "#24.671875,71.3125\n",
    "df = biomass_history_scaled.loc[(biomass_history_scaled['Latitude']==24.671875) & (biomass_history_scaled['Longitude']==71.3125)]\n",
    "# sns.lineplot(data=df.iloc[0,3:], x=[str(x) for x in np.arange(2010,2018)])\n",
    "sns.lineplot(x=[str(x) for x in np.arange(2010,2018)], y=df.iloc[0,3:])\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8112a367",
   "metadata": {},
   "source": [
    "## Distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ea1060fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "distance_matrix = pd.read_csv('dataset/Distance_Matrix.csv')\n",
    "distance_matrix.drop(['Unnamed: 0'],axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "a2a6d74b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='1', ylabel='Count'>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(distance_matrix,x='1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "6ce96f76",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler_mms = MinMaxScaler()\n",
    "scaled_distances = pd.DataFrame(scaler_mms.fit_transform(distance_matrix),\n",
    "                                 columns = distance_matrix.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "952dc853",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='1', ylabel='Count'>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAGwCAYAAABPSaTdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAto0lEQVR4nO3df3RU9Z3/8dcEyACaH4SQXzUhQFeCyo8KGuNPIhQILi1Hdv3Bj8ZCQW3ANTmrmBUNYLdhqUWONuKxK2DPgrTuQaTIYiEIqATUYBbQwAqCg5KEDpYMJGGSkPv9o4f5OpIASW4yMx+ej3PuOdx7P/PO+17RvLzzufc6LMuyBAAAYKiwQDcAAADQkQg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbADAACMRtgBAABG6xroBoJBU1OTjh8/roiICDkcjkC3AwAALoNlWTp9+rSSkpIUFtby9RvCjqTjx48rOTk50G0AAIA2OHbsmK655poW9xN2JEVEREj6+8mKjIwMcDcAAOByeDweJScn+36Pt4SwI/m+uoqMjCTsAAAQYi41BYUJygAAwGiEHQAAYDTCDgAAMBphBwAAGI2wAwAAjEbYAQAARiPsAAAAoxF2AACA0Qg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbADAACMRtgBAABG6xroBgAA7eNyueR2u22pFRsbq5SUFFtqAcGCsAMAIczlciktbZDq6mptqdejR08dOFBO4IFRCDsAEMLcbrfq6mqVPr1AkYmp7arlqTiq3csXyO12E3ZgFMIOABggMjFVMSkDA90GEJSYoAwAAIxG2AEAAEYj7AAAAKMFNOwUFhbqpptuUkREhOLi4jRx4kQdPHjQb8zZs2eVk5Oj3r176+qrr9akSZNUVVXlN8blcumee+5Rz549FRcXpyeeeEKNjY2deSgAACBIBTTsbN++XTk5Odq1a5c2b96shoYGjRkzRjU1Nb4xubm5+vOf/6w333xT27dv1/Hjx3Xvvff69p87d0733HOP6uvrtXPnTr3++utauXKlnn322UAcEgAACDIBvRtr06ZNfusrV65UXFycSktLdeedd6q6ulqvvfaaVq9erbvvvluStGLFCg0aNEi7du3SLbfcor/85S/6/PPPtWXLFsXHx2vYsGF67rnnNHfuXM2fP1/h4eGBODQAABAkgmrOTnV1tSQpJiZGklRaWqqGhgaNHj3aNyYtLU0pKSkqKSmRJJWUlGjw4MGKj4/3jRk7dqw8Ho8+++yzZn+O1+uVx+PxWwAAgJmCJuw0NTXp8ccf12233aYbbrhBklRZWanw8HBFR0f7jY2Pj1dlZaVvzHeDzvn95/c1p7CwUFFRUb4lOTnZ5qMBAADBImjCTk5Ojvbv3681a9Z0+M/Kz89XdXW1bzl27FiH/0wAABAYQfEE5dmzZ2vDhg3asWOHrrnmGt/2hIQE1dfX69SpU35Xd6qqqpSQkOAb89FHH/nVO3+31vkx3+d0OuV0Om0+CgAAEIwCemXHsizNnj1bb731lrZu3ap+/fr57R8+fLi6deum4uJi37aDBw/K5XIpIyNDkpSRkaF9+/bpxIkTvjGbN29WZGSkrrvuus45EAAAELQCemUnJydHq1ev1ttvv62IiAjfHJuoqCj16NFDUVFRmjFjhvLy8hQTE6PIyEjNmTNHGRkZuuWWWyRJY8aM0XXXXadp06Zp8eLFqqys1Lx585STk8PVGwAAENiws2zZMknSyJEj/bavWLFCDz30kCTphRdeUFhYmCZNmiSv16uxY8fq5Zdf9o3t0qWLNmzYoEcffVQZGRm66qqrlJ2drYULF3bWYQAAgCAW0LBjWdYlx3Tv3l1FRUUqKipqcUzfvn21ceNGO1sDAACGCJq7sQAAADoCYQcAABiNsAMAAIxG2AEAAEYj7AAAAKMRdgAAgNEIOwAAwGhB8W4sALjSuFwuud3udtcpLy+3oRvAbIQdAOhkLpdLaWmDVFdXa1vNBm+9bbUA0xB2AKCTud1u1dXVKn16gSITU9tVq2Jfifavf1WNjY32NAcYiLADAAESmZiqmJSB7arhqThqTzOAwZigDAAAjEbYAQAARiPsAAAAoxF2AACA0Qg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbADAACMRtgBAABG43URAHCZeFM5EJoIOwBwGXhTORC6CDsAcBl4UzkQugg7ANAKvKkcCD1MUAYAAEYj7AAAAKMRdgAAgNEIOwAAwGiEHQAAYDTCDgAAMBq3ngMwnh1PPuapx0DoCmjY2bFjh37zm9+otLRUFRUVeuuttzRx4kTffofD0eznFi9erCeeeEKSlJqaqq+++spvf2FhoZ566qkO6xtA6LD7ycc89RgIPQENOzU1NRo6dKimT5+ue++994L9FRUVfuv/8z//oxkzZmjSpEl+2xcuXKiZM2f61iMiIjqmYQAhx64nH/PUYyB0BTTsZGVlKSsrq8X9CQkJfutvv/22MjMz1b9/f7/tERERF4wFgO9q75OPeeoxELpCZs5OVVWV3nnnHb3++usX7Fu0aJGee+45paSkaPLkycrNzVXXri0fmtfrldfr9a17PJ4O6RlA2/GGcQB2CZmw8/rrrysiIuKCr7see+wx3XjjjYqJidHOnTuVn5+viooKLVmypMVahYWFWrBgQUe3DKCNeMM4ADuFTNhZvny5pkyZou7du/ttz8vL8/15yJAhCg8P18MPP6zCwkI5nc5ma+Xn5/t9zuPxKDk5uWMaB9BqvGEcgJ1CIuy8//77OnjwoP74xz9ecmx6eroaGxt19OhRDRzY/PfzTqezxSAEIHjwhnEAdgiJhwq+9tprGj58uIYOHXrJsWVlZQoLC1NcXFwndAYAAIJdQK/snDlzRocOHfKtHzlyRGVlZYqJiVFKSoqkv3/F9Oabb+q3v/3tBZ8vKSnR7t27lZmZqYiICJWUlCg3N1dTp05Vr169Ou04AABA8Apo2Pnkk0+UmZnpWz8/jyY7O1srV66UJK1Zs0aWZenBBx+84PNOp1Nr1qzR/Pnz5fV61a9fP+Xm5vrNxwEAAFe2gIadkSNHyrKsi46ZNWuWZs2a1ey+G2+8Ubt27eqI1gAAgCFCYs4OAABAWxF2AACA0ULi1nMAQOex66nTsbGxvptNgEAi7AAAJEl11SclOTR16lRb6vXo0VMHDpQTeBBwhB0AgCSpofa0JEvDJs9Vn35p7arlqTiq3csXyO12E3YQcIQdAICfq+NS2v3kaiCYMEEZAAAYjbADAACMRtgBAABGI+wAAACjEXYAAIDRCDsAAMBohB0AAGA0wg4AADAaYQcAABiNsAMAAIxG2AEAAEYj7AAAAKMRdgAAgNEIOwAAwGiEHQAAYDTCDgAAMBphBwAAGI2wAwAAjEbYAQAARiPsAAAAoxF2AACA0Qg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbADAACMFtCws2PHDk2YMEFJSUlyOBxat26d3/6HHnpIDofDbxk3bpzfmG+//VZTpkxRZGSkoqOjNWPGDJ05c6YTjwIAAASzgIadmpoaDR06VEVFRS2OGTdunCoqKnzLG2+84bd/ypQp+uyzz7R582Zt2LBBO3bs0KxZszq6dQAAECK6BvKHZ2VlKSsr66JjnE6nEhISmt1XXl6uTZs26eOPP9aIESMkSS+99JLGjx+v559/XklJSc1+zuv1yuv1+tY9Hk8bjwAAAAS7oJ+zs23bNsXFxWngwIF69NFHdfLkSd++kpISRUdH+4KOJI0ePVphYWHavXt3izULCwsVFRXlW5KTkzv0GAAAQOAEddgZN26c/vCHP6i4uFj/8R//oe3btysrK0vnzp2TJFVWViouLs7vM127dlVMTIwqKytbrJufn6/q6mrfcuzYsQ49DgAAEDgB/RrrUh544AHfnwcPHqwhQ4ZowIAB2rZtm0aNGtXmuk6nU06n044WAQBAkAvqKzvf179/f8XGxurQoUOSpISEBJ04ccJvTGNjo7799tsW5/kAAIArS0iFna+//lonT55UYmKiJCkjI0OnTp1SaWmpb8zWrVvV1NSk9PT0QLUJAACCSEC/xjpz5ozvKo0kHTlyRGVlZYqJiVFMTIwWLFigSZMmKSEhQYcPH9aTTz6pH/7whxo7dqwkadCgQRo3bpxmzpypV155RQ0NDZo9e7YeeOCBFu/EAuDP5XLJ7XbbUis2NlYpKSm21AIAuwQ07HzyySfKzMz0refl5UmSsrOztWzZMu3du1evv/66Tp06paSkJI0ZM0bPPfec33ybVatWafbs2Ro1apTCwsI0adIkvfjii51+LEAocrlcSksbpLq6Wlvq9ejRUwcOlBN44FNeXt7uGoRotFdAw87IkSNlWVaL+999991L1oiJidHq1avtbAu4YrjdbtXV1Sp9eoEiE1PbVctTcVS7ly+Q2+3mFxNUV31SkkNTp05tdy1CNNorqO/GAtA5IhNTFZMyMNBtwCANtaclWRo2ea769Etrcx1CNOxA2AEAdJir41II0gi4kLobCwAAoLUIOwAAwGiEHQAAYDTCDgAAMBphBwAAGI2wAwAAjEbYAQAARiPsAAAAoxF2AACA0Qg7AADAaLwuAoCt7HjLtR01AOA8wg4AW9j5luvzGrz1ttUCcOUi7ACwhV1vuZakin0l2r/+VTU2NtrTHIArGmEHgK3seMu1p+KoPc0AgJigDAAADEfYAQAARiPsAAAAoxF2AACA0Qg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbADAACMRtgBAABG43URQIhyuVxyu93tqsHbxQFcCQg7QAhyuVxKSxukurpaW+rxdnEAJiPsACHI7Xarrq5W6dMLFJmY2uY6vF0cwJWAsAOEsMjE1Ha9YZy3iwO4EjBBGQAAGI2wAwAAjBbQsLNjxw5NmDBBSUlJcjgcWrdunW9fQ0OD5s6dq8GDB+uqq65SUlKSfvazn+n48eN+NVJTU+VwOPyWRYsWdfKRAACAYBXQsFNTU6OhQ4eqqKjogn21tbXas2ePnnnmGe3Zs0dr167VwYMH9ZOf/OSCsQsXLlRFRYVvmTNnTme0DwAAQkBAJyhnZWUpKyur2X1RUVHavHmz37bf/e53uvnmm+VyuZSSkuLbHhERoYSEhMv+uV6vV16v17fu8Xha2TkAAAgVITVnp7q6Wg6HQ9HR0X7bFy1apN69e+tHP/qRfvOb31zyNtrCwkJFRUX5luTk5A7sGgAABFLI3Hp+9uxZzZ07Vw8++KAiIyN92x977DHdeOONiomJ0c6dO5Wfn6+KigotWbKkxVr5+fnKy8vzrXs8HgIPAACGComw09DQoPvuu0+WZWnZsmV++74bWoYMGaLw8HA9/PDDKiwslNPpbLae0+lscR8AADBL0H+NdT7ofPXVV9q8ebPfVZ3mpKenq7GxUUePHu2cBgEAQFAL6is754POF198offee0+9e/e+5GfKysoUFhamuLi4TugQAAAEu4CGnTNnzujQoUO+9SNHjqisrEwxMTFKTEzUP/3TP2nPnj3asGGDzp07p8rKSklSTEyMwsPDVVJSot27dyszM1MREREqKSlRbm6upk6dql69egXqsAAAQBAJaNj55JNPlJmZ6Vs/P/8mOztb8+fP1/r16yVJw4YN8/vce++9p5EjR8rpdGrNmjWaP3++vF6v+vXrp9zcXL95PAAA4MoW0LAzcuRIWZbV4v6L7ZOkG2+8Ubt27bK7LQAAYJCgn6AMAADQHoQdAABgNMIOAAAwWpvCTv/+/XXy5MkLtp86dUr9+/dvd1MAAAB2aVPYOXr0qM6dO3fBdq/Xq2+++abdTQEAANilVXdjnb8VXJLeffddRUVF+dbPnTun4uJipaam2tYcAABAe7Uq7EycOFGS5HA4lJ2d7bevW7duSk1N1W9/+1vbmgMAAGivVoWdpqYmSVK/fv308ccfKzY2tkOaAgAAsEubHip45MgRu/sAAADoEG1+gnJxcbGKi4t14sQJ3xWf85YvX97uxgAAAOzQprCzYMECLVy4UCNGjFBiYqIcDofdfQEAANiiTWHnlVde0cqVKzVt2jS7+wEAALBVm56zU19fr1tvvdXuXgAAAGzXprDzi1/8QqtXr7a7FwAAANu16Wuss2fP6tVXX9WWLVs0ZMgQdevWzW//kiVLbGkOAACgvdoUdvbu3athw4ZJkvbv3++3j8nKAAAgmLQp7Lz33nt29wEAANAh2jRnBwAAIFS06cpOZmbmRb+u2rp1a5sbAgAAsFObws75+TrnNTQ0qKysTPv377/gBaEAAACB1Kaw88ILLzS7ff78+Tpz5ky7GgIAALCTrXN2pk6dynuxAABAULE17JSUlKh79+52lgQAAGiXNn2Nde+99/qtW5aliooKffLJJ3rmmWdsaQwAAMAObQo7UVFRfuthYWEaOHCgFi5cqDFjxtjSGAAAgB3aFHZWrFhhdx8AAAAdok1h57zS0lKVl5dLkq6//nr96Ec/sqUpAAAAu7Qp7Jw4cUIPPPCAtm3bpujoaEnSqVOnlJmZqTVr1qhPnz529ggYw+Vyye12t7vO+f/JAABcWpvCzpw5c3T69Gl99tlnGjRokCTp888/V3Z2th577DG98cYbtjYJmMDlciktbZDq6mptq9ngrbetFgCYqk1hZ9OmTdqyZYsv6EjSddddp6KiIiYoAy1wu92qq6tV+vQCRSamtqtWxb4S7V//qhobG+1pDgAM1qaw09TUpG7dul2wvVu3bmpqamp3U4DJIhNTFZMysF01PBVH7WkGAK4AbXqo4N13361/+Zd/0fHjx33bvvnmG+Xm5mrUqFGXXWfHjh2aMGGCkpKS5HA4tG7dOr/9lmXp2WefVWJionr06KHRo0friy++8Bvz7bffasqUKYqMjFR0dLRmzJjBKysAAIBPm8LO7373O3k8HqWmpmrAgAEaMGCA+vXrJ4/Ho5deeumy69TU1Gjo0KEqKipqdv/ixYv14osv6pVXXtHu3bt11VVXaezYsTp79qxvzJQpU/TZZ59p8+bN2rBhg3bs2KFZs2a15bAAAICB2vQ1VnJysvbs2aMtW7bowIEDkqRBgwZp9OjRraqTlZWlrKysZvdZlqWlS5dq3rx5+ulPfypJ+sMf/qD4+HitW7dODzzwgMrLy7Vp0yZ9/PHHGjFihCTppZde0vjx4/X8888rKSmpLYcHAAAM0qorO1u3btV1110nj8cjh8OhH//4x5ozZ47mzJmjm266Sddff73ef/99Wxo7cuSIKisr/QJUVFSU0tPTVVJSIunv7+KKjo72BR1JGj16tMLCwrR79+4Wa3u9Xnk8Hr8FAACYqVVhZ+nSpZo5c6YiIyMv2BcVFaWHH35YS5YssaWxyspKSVJ8fLzf9vj4eN++yspKxcXF+e3v2rWrYmJifGOaU1hYqKioKN+SnJxsS88AACD4tCrs/O///q/GjRvX4v4xY8aotLS03U11tPz8fFVXV/uWY8eOBbolAADQQVoVdqqqqpq95fy8rl276q9//Wu7m5KkhIQE38/8fg/n9yUkJOjEiRN++xsbG/Xtt9/6xjTH6XQqMjLSbwEAAGZqVdj5wQ9+oP3797e4f+/evUpMTGx3U5LUr18/JSQkqLi42LfN4/Fo9+7dysjIkCRlZGTo1KlTfleTtm7dqqamJqWnp9vSBwAACG2tCjvjx4/XM88843fr93l1dXUqKCjQP/7jP152vTNnzqisrExlZWWS/j4puaysTC6XSw6HQ48//rh+9atfaf369dq3b59+9rOfKSkpSRMnTpT09zvAxo0bp5kzZ+qjjz7Shx9+qNmzZ+uBBx7gTiwAACCplbeez5s3T2vXrtW1116r2bNna+DAvz8F9sCBAyoqKtK5c+f09NNPX3a9Tz75RJmZmb71vLw8SVJ2drZWrlypJ598UjU1NZo1a5ZOnTql22+/XZs2bVL37t19n1m1apVmz56tUaNGKSwsTJMmTdKLL77YmsMCAAAGa1XYiY+P186dO/Xoo48qPz9flmVJkhwOh8aOHauioqIL7p66mJEjR/pqNMfhcGjhwoVauHBhi2NiYmK0evXqyz8IAABwRWn1QwX79u2rjRs36m9/+5sOHToky7L0D//wD+rVq1dH9AcAANAubXqCsiT16tVLN910k529AAAA2K7NYQcAgM5SXl5uS53Y2FilpKTYUguhg7ADAAhaddUnJTk0depUW+r16NFTBw6UE3iuMIQdAEDQaqg9LcnSsMlz1adfWrtqeSqOavfyBXK73YSdKwxhBwAQ9K6OS1FMysBAt4EQ1aqHCgIAAIQawg4AADAaYQcAABiNsAMAAIxG2AEAAEYj7AAAAKNx6zlwCS6XS263u9117HoCLACgdQg7wEW4XC6lpQ1SXV2tbTUbvPW21QIAXBphB7gIt9uturpapU8vUGRiartqVewr0f71r6qxsdGe5gAAl4WwA1yGyMTUdj+91VNx1J5mAACtQtiBkZhnAwA4j7AD4zDPBgDwXYQdGId5NgCA7yLswFjMswEASDxUEAAAGI6wAwAAjEbYAQAARiPsAAAAoxF2AACA0Qg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbADAACMRtgBAABGI+wAAACjBX3YSU1NlcPhuGDJycmRJI0cOfKCfY888kiAuwYAAMEi6N96/vHHH+vcuXO+9f379+vHP/6x/vmf/9m3bebMmVq4cKFvvWfPnp3aIwAACF5BH3b69Onjt75o0SINGDBAd911l29bz549lZCQ0NmtAQCAEBD0X2N9V319vf7rv/5L06dPl8Ph8G1ftWqVYmNjdcMNNyg/P1+1tbUXreP1euXxePwWAABgpqC/svNd69at06lTp/TQQw/5tk2ePFl9+/ZVUlKS9u7dq7lz5+rgwYNau3Zti3UKCwu1YMGCTugYAAAEWkiFnddee01ZWVlKSkrybZs1a5bvz4MHD1ZiYqJGjRqlw4cPa8CAAc3Wyc/PV15enm/d4/EoOTm54xoHAAABEzJh56uvvtKWLVsuesVGktLT0yVJhw4dajHsOJ1OOZ1O23sEAADBJ2Tm7KxYsUJxcXG65557LjqurKxMkpSYmNgJXQEAgGAXEld2mpqatGLFCmVnZ6tr1//f8uHDh7V69WqNHz9evXv31t69e5Wbm6s777xTQ4YMCWDHAAAgWIRE2NmyZYtcLpemT5/utz08PFxbtmzR0qVLVVNTo+TkZE2aNEnz5s0LUKcAACDYhETYGTNmjCzLumB7cnKytm/fHoCOAABAqAiZOTsAAABtQdgBAABGI+wAAACjEXYAAIDRQmKCMgAAdikvL7elTmxsrFJSUmyphY5F2AEAXBHqqk9Kcmjq1Km21OvRo6cOHCgn8IQAwg4A4IrQUHtakqVhk+eqT7+0dtXyVBzV7uUL5Ha7CTshgLADALiiXB2XopiUgYFuA52ICcoAAMBohB0AAGA0wg4AADAaYQcAABiNsAMAAIxG2AEAAEYj7AAAAKMRdgAAgNEIOwAAwGg8QRlBw+Vyye12t7uOXS/5AwCYgbCDoOByuZSWNkh1dbW21Wzw1ttWCwAQugg7CAput1t1dbVKn16gyMTUdtWq2Fei/etfVWNjoz3NAQBCGmEHQSUyMbXdL+jzVBy1pxkAgBEIO2g3O+baMM8GANBRCDtoF7vn2jDPBgBgN8IO2sWuuTbMswEAdBTCDmzR3rk2zLMBAHQUHioIAACMRtgBAABGI+wAAACjEXYAAIDRCDsAAMBohB0AAGA0wg4AADBaUIed+fPny+Fw+C1paWm+/WfPnlVOTo569+6tq6++WpMmTVJVVVUAOwYAAMEmqMOOJF1//fWqqKjwLR988IFvX25urv785z/rzTff1Pbt23X8+HHde++9AewWAAAEm6B/gnLXrl2VkJBwwfbq6mq99tprWr16te6++25J0ooVKzRo0CDt2rVLt9xyS4s1vV6vvF6vb93j8djfOAAACApBf2Xniy++UFJSkvr3768pU6bI5XJJkkpLS9XQ0KDRo0f7xqalpSklJUUlJSUXrVlYWKioqCjfkpyc3KHHAAAAAieow056erpWrlypTZs2admyZTpy5IjuuOMOnT59WpWVlQoPD1d0dLTfZ+Lj41VZWXnRuvn5+aqurvYtx44d68CjAAAAgRTUX2NlZWX5/jxkyBClp6erb9+++tOf/qQePXq0ua7T6ZTT6bSjRQAAEOSC+srO90VHR+vaa6/VoUOHlJCQoPr6ep06dcpvTFVVVbNzfAAAwJUppMLOmTNndPjwYSUmJmr48OHq1q2biouLffsPHjwol8uljIyMAHYJAACCSVB/jfWv//qvmjBhgvr27avjx4+roKBAXbp00YMPPqioqCjNmDFDeXl5iomJUWRkpObMmaOMjIyL3okFAACuLEEddr7++ms9+OCDOnnypPr06aPbb79du3btUp8+fSRJL7zwgsLCwjRp0iR5vV6NHTtWL7/8coC7BgAAwSSow86aNWsuur979+4qKipSUVFRJ3UEAABCTUjN2QEAAGgtwg4AADAaYQcAABgtqOfsAAAQzMrLy22pExsbq5SUFFtq4UKEHQAAWqmu+qQkh6ZOnWpLvR49eurAgXICTwch7AAA0EoNtaclWRo2ea769EtrVy1PxVHtXr5AbrebsNNBCDsAALTR1XEpikkZGOg2cAlMUAYAAEYj7AAAAKMRdgAAgNGYswMAQBDgNvaOQ9gBACCAuI294xF2AAAIoI64jf3999/XoEGD2lXLpCtEhB0AAIKAHbex23mVyKQrRIQdAAAMYddVItMedEjYAQDAMDzs0B+3ngMAAKNxZecK5XK55Ha7213HrlslAQDoKISdK5DL5VJa2iDV1dXaVrPBW29bLQAA7ETYuQK53W7V1dUqfXqBIhNT21WrYl+J9q9/VY2NjfY0BwCAzQg7V7DIxNR2T2DzVBy1pxkAADoIE5QBAIDRCDsAAMBohB0AAGA0wg4AADAaYQcAABiNsAMAAIxG2AEAAEYj7AAAAKMRdgAAgNEIOwAAwGhBHXYKCwt10003KSIiQnFxcZo4caIOHjzoN2bkyJFyOBx+yyOPPBKgjgEAQLAJ6rCzfft25eTkaNeuXdq8ebMaGho0ZswY1dTU+I2bOXOmKioqfMvixYsD1DEAAAg2Qf0i0E2bNvmtr1y5UnFxcSotLdWdd97p296zZ08lJCR0dnsAACAEBPWVne+rrq6WJMXExPhtX7VqlWJjY3XDDTcoPz9ftbW1F63j9Xrl8Xj8FgAAYKagvrLzXU1NTXr88cd122236YYbbvBtnzx5svr27aukpCTt3btXc+fO1cGDB7V27doWaxUWFmrBggWd0TYAAAiwkAk7OTk52r9/vz744AO/7bNmzfL9efDgwUpMTNSoUaN0+PBhDRgwoNla+fn5ysvL8617PB4lJyd3TOMAACCgQiLszJ49Wxs2bNCOHTt0zTXXXHRsenq6JOnQoUMthh2n0ymn02l7nwAAIPgEddixLEtz5szRW2+9pW3btqlfv36X/ExZWZkkKTExsYO7AwAAoSCow05OTo5Wr16tt99+WxEREaqsrJQkRUVFqUePHjp8+LBWr16t8ePHq3fv3tq7d69yc3N15513asiQIQHu3n4ul0tut7vddcrLy23oBgCA0BDUYWfZsmWS/v7gwO9asWKFHnroIYWHh2vLli1aunSpampqlJycrEmTJmnevHkB6LZjuVwupaUNUl3dxe80a40Gb71ttQAACFZBHXYsy7ro/uTkZG3fvr2Tugkst9uturpapU8vUGRiartqVewr0f71r6qxsdGe5gAACGJBHXZwocjEVMWkDGxXDU/FUXuaAQAgBITUQwUBAABai7ADAACMRtgBAABGI+wAAACjEXYAAIDRCDsAAMBohB0AAGA0wg4AADAaDxXsYLzPCgCAwCLsdCDeZwUAQOARdjoQ77MCACDwCDudgPdZAQAQOExQBgAARiPsAAAAoxF2AACA0Qg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbADAACMRtgBAABGI+wAAACjEXYAAIDReDcWAABoVnl5uS11YmNjlZKSYkuttiDsAAAAP3XVJyU5NHXqVFvq9ejRUwcOlAcs8BB2AACAn4ba05IsDZs8V336pbWrlqfiqHYvXyC3203YAQAAweXquBTFpAwMdBvtxgRlAABgNMIOAAAwGmEHAAAYjbADAACMZkzYKSoqUmpqqrp376709HR99NFHgW4JAAAEASPCzh//+Efl5eWpoKBAe/bs0dChQzV27FidOHEi0K0BAIAAMyLsLFmyRDNnztTPf/5zXXfddXrllVfUs2dPLV++PNCtAQCAAAv55+zU19ertLRU+fn5vm1hYWEaPXq0SkpKmv2M1+uV1+v1rVdXV0uSPB6Prb2dOXNGkvTtVwfV6K1rVy1PxVeSpOpvvlC3rg7jagVjT1dCrWDsKVhrBWNPV0KtYOzpSqhla0+VLkl//51o9+/Z8/Usy7r4QCvEffPNN5Yka+fOnX7bn3jiCevmm29u9jMFBQWWJBYWFhYWFhYDlmPHjl00K4T8lZ22yM/PV15enm+9qalJ3377rXr37i2Ho30J9rs8Ho+Sk5N17NgxRUZG2lYX/jjPnYdz3Tk4z52D89w5OvI8W5al06dPKykp6aLjQj7sxMbGqkuXLqqqqvLbXlVVpYSEhGY/43Q65XQ6/bZFR0d3VIuKjIzkX6ROwHnuPJzrzsF57hyc587RUec5KirqkmNCfoJyeHi4hg8fruLiYt+2pqYmFRcXKyMjI4CdAQCAYBDyV3YkKS8vT9nZ2RoxYoRuvvlmLV26VDU1Nfr5z38e6NYAAECAGRF27r//fv31r3/Vs88+q8rKSg0bNkybNm1SfHx8QPtyOp0qKCi44Csz2Ivz3Hk4152D89w5OM+dIxjOs8OyLnW/FgAAQOgK+Tk7AAAAF0PYAQAARiPsAAAAoxF2AACA0Qg77VRUVKTU1FR1795d6enp+uijjy46/s0331RaWpq6d++uwYMHa+PGjZ3UaWhrzXn+/e9/rzvuuEO9evVSr169NHr06Ev+c8Hftfbv83lr1qyRw+HQxIkTO7ZBg7T2XJ86dUo5OTlKTEyU0+nUtddey38/LkNrz/PSpUs1cOBA9ejRQ8nJycrNzdXZs2c7qdvQtGPHDk2YMEFJSUlyOBxat27dJT+zbds23XjjjXI6nfrhD3+olStXdmyT9ryh6sq0Zs0aKzw83Fq+fLn12WefWTNnzrSio6OtqqqqZsd/+OGHVpcuXazFixdbn3/+uTVv3jyrW7du1r59+zq589DS2vM8efJkq6ioyPr000+t8vJy66GHHrKioqKsr7/+upM7Dy2tPc/nHTlyxPrBD35g3XHHHdZPf/rTzmk2xLX2XHu9XmvEiBHW+PHjrQ8++MA6cuSItW3bNqusrKyTOw8trT3Pq1atspxOp7Vq1SrryJEj1rvvvmslJiZaubm5ndx5aNm4caP19NNPW2vXrrUkWW+99dZFx3/55ZdWz549rby8POvzzz+3XnrpJatLly7Wpk2bOqxHwk473HzzzVZOTo5v/dy5c1ZSUpJVWFjY7Pj77rvPuueee/y2paenWw8//HCH9hnqWnuev6+xsdGKiIiwXn/99Y5q0QhtOc+NjY3Wrbfeav3nf/6nlZ2dTdi5TK0918uWLbP69+9v1dfXd1aLRmjtec7JybHuvvtuv215eXnWbbfd1qF9muRyws6TTz5pXX/99X7b7r//fmvs2LEd1hdfY7VRfX29SktLNXr0aN+2sLAwjR49WiUlJc1+pqSkxG+8JI0dO7bF8Wjbef6+2tpaNTQ0KCYmpqPaDHltPc8LFy5UXFycZsyY0RltGqEt53r9+vXKyMhQTk6O4uPjdcMNN+jXv/61zp0711lth5y2nOdbb71VpaWlvq+6vvzyS23cuFHjx4/vlJ6vFIH4XWjEE5QDwe1269y5cxc8pTk+Pl4HDhxo9jOVlZXNjq+srOywPkNdW87z982dO1dJSUkX/MuF/68t5/mDDz7Qa6+9prKysk7o0BxtOddffvmltm7dqilTpmjjxo06dOiQfvnLX6qhoUEFBQWd0XbIact5njx5stxut26//XZZlqXGxkY98sgj+rd/+7fOaPmK0dLvQo/Ho7q6OvXo0cP2n8mVHRht0aJFWrNmjd566y1179490O0Y4/Tp05o2bZp+//vfKzY2NtDtGK+pqUlxcXF69dVXNXz4cN1///16+umn9corrwS6NaNs27ZNv/71r/Xyyy9rz549Wrt2rd555x0999xzgW4N7cSVnTaKjY1Vly5dVFVV5be9qqpKCQkJzX4mISGhVePRtvN83vPPP69FixZpy5YtGjJkSEe2GfJae54PHz6so0ePasKECb5tTU1NkqSuXbvq4MGDGjBgQMc2HaLa8nc6MTFR3bp1U5cuXXzbBg0apMrKStXX1ys8PLxDew5FbTnPzzzzjKZNm6Zf/OIXkqTBgwerpqZGs2bN0tNPP62wMK4P2KGl34WRkZEdclVH4spOm4WHh2v48OEqLi72bWtqalJxcbEyMjKa/UxGRobfeEnavHlzi+PRtvMsSYsXL9Zzzz2nTZs2acSIEZ3Rakhr7XlOS0vTvn37VFZW5lt+8pOfKDMzU2VlZUpOTu7M9kNKW/5O33bbbTp06JAvUErS//3f/ykxMZGg04K2nOfa2toLAs35gGnxGknbBOR3YYdNfb4CrFmzxnI6ndbKlSutzz//3Jo1a5YVHR1tVVZWWpZlWdOmTbOeeuop3/gPP/zQ6tq1q/X8889b5eXlVkFBAbeeX4bWnudFixZZ4eHh1n//939bFRUVvuX06dOBOoSQ0Nrz/H3cjXX5WnuuXS6XFRERYc2ePds6ePCgtWHDBisuLs761a9+FahDCAmtPc8FBQVWRESE9cYbb1hffvml9Ze//MUaMGCAdd999wXqEELC6dOnrU8//dT69NNPLUnWkiVLrE8//dT66quvLMuyrKeeesqaNm2ab/z5W8+feOIJq7y83CoqKuLW82D30ksvWSkpKVZ4eLh18803W7t27fLtu+uuu6zs7Gy/8X/605+sa6+91goPD7euv/5665133unkjkNTa85z3759LUkXLAUFBZ3feIhp7d/n7yLstE5rz/XOnTut9PR0y+l0Wv3797f+/d//3WpsbOzkrkNPa85zQ0ODNX/+fGvAgAFW9+7dreTkZOuXv/yl9be//a3zGw8h7733XrP/zT1/brOzs6277rrrgs8MGzbMCg8Pt/r372+tWLGiQ3t0WBbX5gAAgLmYswMAAIxG2AEAAEYj7AAAAKMRdgAAgNEIOwAAwGiEHQAAYDTCDgAAMBphBwAAGI2wAwAAjEbYAWCkHTt2aMKECUpKSpLD4dC6desC3RKAACHsADBSTU2Nhg4dqqKiokC3AiDAuga6AQDoCFlZWcrKygp0GwCCAFd2AACA0Qg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbuxABjpzJkzOnTokG/9yJEjKisrU0xMjFJSUgLYGYDO5rAsywp0EwBgt23btikzM/OC7dnZ2Vq5cmXnNwQgYAg7AADAaMzZAQAARiPsAAAAoxF2AACA0Qg7AADAaIQdAABgNMIOAAAwGmEHAAAYjbADAACMRtgBAABGI+wAAACjEXYAAIDR/h/nsvUJqYeKegAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(scaled_distances,x='1')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "36269bdc",
   "metadata": {},
   "source": [
    "# Some important functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6625f6eb",
   "metadata": {},
   "outputs": [],
   "source": [
    "#get distance between two coordinates\n",
    "def distance_bw_coordinates(x1,y1,x2,y2):\n",
    "    point1 = biomass_history_scaled.loc[(biomass_history_scaled['Latitude']==x1) & (biomass_history_scaled['Longitude']==y1)].index[0]\n",
    "    point2 = biomass_history_scaled.loc[(biomass_history_scaled['Latitude']==x2) & (biomass_history_scaled['Longitude']==y2)].index[0]\n",
    "    return scaled_distances.iloc[point1,point2]\n",
    "# distance_bw_coordinates(24.671875,71.3125,24.671875,71.4375)\n",
    "# 0.01617031498066282"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "2393d61b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#get the biomass history at a particular coordinate\n",
    "def get_biomass_hist_scaled(x,y,year=None,data=biomass_history_scaled):\n",
    "    if(year!=None):\n",
    "        return data.loc[(data['Latitude']==x) & (data['Longitude']==y), str(year)][0]\n",
    "    else:\n",
    "        return data.loc[(data['Latitude']==x) & (data['Longitude']==y)]\n",
    "# get_biomass_hist_scaled(24.671875,71.3125,2010)\n",
    "# 0.009597778"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa5cb5f9",
   "metadata": {},
   "source": [
    "# Splitting into train and test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "b40fb834",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_df,train_df = biomass_history_scaled,biomass_history_scaled.drop(['2017'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "56fbb2c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "#for a particular coordinate this gives the train and the test data\n",
    "#x and y are the latitude and longitude and n is the number of dataset for train set\n",
    "#this will create batches of dataset\n",
    "def data_batch_generator(x,y,n=1,data=train_df):\n",
    "    data_x=[]\n",
    "    data_y=[]\n",
    "    data=np.array(get_biomass_hist_scaled(x,y,data=data))[0][3:]\n",
    "    xi=[]\n",
    "    yi=[]\n",
    "    test = test_df.iloc[:,-n-1:]\n",
    "    test_df_x = test.iloc[:,:-1]\n",
    "    test_df_y = test.iloc[:,-1]\n",
    "    idx = biomass_history_scaled.loc[(biomass_history_scaled['Latitude']==x) & (biomass_history_scaled['Longitude']==y)].index[0]\n",
    "    test_df_x = np.array(test_df_x.iloc[idx])\n",
    "    test_df_y = np.array(test_df_y.iloc[idx])\n",
    "    for i in range(len(data)-n+1):\n",
    "        try:\n",
    "            xi = data[i:i+n]\n",
    "            yi = data[i+n]\n",
    "            data_x.append(xi)\n",
    "            data_y.append(yi)\n",
    "            \n",
    "        except:\n",
    "            print(\"Data generation Successful\")\n",
    "    return np.array(data_x),np.array(data_y),np.array([test_df_x]),test_df_y.reshape(-1)\n",
    "\n",
    "# data_batch_generator(24.671875,71.3125,3)\n",
    "# ([array([0.00959778, 0.0114975 , 0.01055145], dtype=float32),\n",
    "#   array([0.0114975 , 0.01055145, 0.00488663], dtype=float32),\n",
    "#   array([0.01055145, 0.00488663, 0.01236725], dtype=float32),\n",
    "#   array([0.00488663, 0.01236725, 0.00862122], dtype=float32)],\n",
    "#  [array(0.00488663, dtype=float32),\n",
    "#   array(0.01236725, dtype=float32),\n",
    "#   array(0.00862122, dtype=float32),\n",
    "#   array(0.01437378, dtype=float32)],\n",
    "#  array([0.01236725, 0.00862122, 0.01437378], dtype=float32),\n",
    "#  array(0.00586319, dtype=float32))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "fb8601f1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data generation Successful\n"
     ]
    }
   ],
   "source": [
    "#for coordinates 24.671875,71.3125\n",
    "train_x,train_y,test_x,test_y = data_batch_generator(24.671875,71.3125,4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "00867a75",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3, 4)\n",
      "(3,)\n",
      "(1, 4)\n",
      "(1,)\n"
     ]
    }
   ],
   "source": [
    "print(train_x.shape)\n",
    "print(train_y.shape)\n",
    "print(test_x.shape)\n",
    "print(test_y.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ed6dab1",
   "metadata": {},
   "source": [
    "# DNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "227848a0",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-22 19:57:24.601624: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.668925: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.669530: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.672568: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.673021: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.673400: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.785501: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.785883: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.786054: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-08-22 19:57:24.786186: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1639] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 1166 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3060 Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 8.6\n"
     ]
    }
   ],
   "source": [
    "input_layer = Input(shape=(4,),name='input')\n",
    "dense_layer = Dense(units=32,activation='relu',name='d1')(input_layer)\n",
    "dense_layer = Dense(units=64, activation='relu',name='d2')(input_layer)\n",
    "dense_layer = Dense(units=128, activation='relu',name='d3')(dense_layer)\n",
    "dense_layer = Dense(units=256, activation='relu',name='d4')(dense_layer)\n",
    "dense_layer = Dense(units=512, activation='relu',name='d5')(dense_layer)\n",
    "dense_layer = Dense(units=1024, activation='relu',name='d6')(dense_layer)\n",
    "dense_layer = Dense(units=256, activation='relu',name='d7')(dense_layer)\n",
    "output_layer = Dense(units=1, activation='linear',name='output')(dense_layer)\n",
    "model = Model(inputs=input_layer, outputs=output_layer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "b7c9b6d2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input (InputLayer)          [(None, 4)]               0         \n",
      "                                                                 \n",
      " d2 (Dense)                  (None, 64)                320       \n",
      "                                                                 \n",
      " d3 (Dense)                  (None, 128)               8320      \n",
      "                                                                 \n",
      " d4 (Dense)                  (None, 256)               33024     \n",
      "                                                                 \n",
      " d5 (Dense)                  (None, 512)               131584    \n",
      "                                                                 \n",
      " d6 (Dense)                  (None, 1024)              525312    \n",
      "                                                                 \n",
      " d7 (Dense)                  (None, 256)               262400    \n",
      "                                                                 \n",
      " output (Dense)              (None, 1)                 257       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 961217 (3.67 MB)\n",
      "Trainable params: 961217 (3.67 MB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.compile(optimizer='adam', loss='mean_squared_error',metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "cdd18e6c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-22 19:57:26.075815: I tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:606] TensorFloat-32 will be used for the matrix multiplication. This will only be logged once.\n",
      "2023-08-22 19:57:26.135452: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x7ff40c50fa60 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
      "2023-08-22 19:57:26.135565: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
      "2023-08-22 19:57:26.143953: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:255] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
      "2023-08-22 19:57:26.276128: I tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:432] Loaded cuDNN version 8904\n",
      "2023-08-22 19:57:26.446354: I ./tensorflow/compiler/jit/device_compiler.h:186] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 - 4s - loss: 1.4705e-04 - accuracy: 0.0000e+00 - val_loss: 6.9491e-05 - val_accuracy: 0.0000e+00 - 4s/epoch - 4s/step\n",
      "Epoch 2/100\n",
      "1/1 - 0s - loss: 1.2205e-05 - accuracy: 0.0000e+00 - val_loss: 4.3861e-05 - val_accuracy: 0.0000e+00 - 37ms/epoch - 37ms/step\n",
      "Epoch 3/100\n",
      "1/1 - 0s - loss: 6.1647e-06 - accuracy: 0.0000e+00 - val_loss: 1.5398e-05 - val_accuracy: 0.0000e+00 - 45ms/epoch - 45ms/step\n",
      "Epoch 4/100\n",
      "1/1 - 0s - loss: 9.4640e-06 - accuracy: 0.0000e+00 - val_loss: 7.3991e-05 - val_accuracy: 0.0000e+00 - 42ms/epoch - 42ms/step\n",
      "Epoch 5/100\n",
      "1/1 - 0s - loss: 1.2936e-05 - accuracy: 0.0000e+00 - val_loss: 3.4637e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 6/100\n",
      "1/1 - 0s - loss: 5.3633e-06 - accuracy: 0.0000e+00 - val_loss: 1.2243e-05 - val_accuracy: 0.0000e+00 - 57ms/epoch - 57ms/step\n",
      "Epoch 7/100\n",
      "1/1 - 0s - loss: 1.1025e-05 - accuracy: 0.0000e+00 - val_loss: 5.1443e-05 - val_accuracy: 0.0000e+00 - 50ms/epoch - 50ms/step\n",
      "Epoch 8/100\n",
      "1/1 - 0s - loss: 7.1350e-06 - accuracy: 0.0000e+00 - val_loss: 5.4000e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 9/100\n",
      "1/1 - 0s - loss: 7.5921e-06 - accuracy: 0.0000e+00 - val_loss: 1.7743e-05 - val_accuracy: 0.0000e+00 - 40ms/epoch - 40ms/step\n",
      "Epoch 10/100\n",
      "1/1 - 0s - loss: 8.1110e-06 - accuracy: 0.0000e+00 - val_loss: 2.6893e-05 - val_accuracy: 0.0000e+00 - 46ms/epoch - 46ms/step\n",
      "Epoch 11/100\n",
      "1/1 - 0s - loss: 5.8113e-06 - accuracy: 0.0000e+00 - val_loss: 5.7629e-05 - val_accuracy: 0.0000e+00 - 48ms/epoch - 48ms/step\n",
      "Epoch 12/100\n",
      "1/1 - 0s - loss: 8.1935e-06 - accuracy: 0.0000e+00 - val_loss: 3.3823e-05 - val_accuracy: 0.0000e+00 - 45ms/epoch - 45ms/step\n",
      "Epoch 13/100\n",
      "1/1 - 0s - loss: 5.1806e-06 - accuracy: 0.0000e+00 - val_loss: 1.6881e-05 - val_accuracy: 0.0000e+00 - 47ms/epoch - 47ms/step\n",
      "Epoch 14/100\n",
      "1/1 - 0s - loss: 8.2542e-06 - accuracy: 0.0000e+00 - val_loss: 4.8308e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 15/100\n",
      "1/1 - 0s - loss: 6.3259e-06 - accuracy: 0.0000e+00 - val_loss: 4.7446e-05 - val_accuracy: 0.0000e+00 - 48ms/epoch - 48ms/step\n",
      "Epoch 16/100\n",
      "1/1 - 0s - loss: 6.1691e-06 - accuracy: 0.0000e+00 - val_loss: 1.8592e-05 - val_accuracy: 0.0000e+00 - 41ms/epoch - 41ms/step\n",
      "Epoch 17/100\n",
      "1/1 - 0s - loss: 7.4584e-06 - accuracy: 0.0000e+00 - val_loss: 3.3056e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 18/100\n",
      "1/1 - 0s - loss: 5.0564e-06 - accuracy: 0.0000e+00 - val_loss: 5.4908e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 19/100\n",
      "1/1 - 0s - loss: 7.4942e-06 - accuracy: 0.0000e+00 - val_loss: 2.4403e-05 - val_accuracy: 0.0000e+00 - 41ms/epoch - 41ms/step\n",
      "Epoch 20/100\n",
      "1/1 - 0s - loss: 5.8185e-06 - accuracy: 0.0000e+00 - val_loss: 2.4588e-05 - val_accuracy: 0.0000e+00 - 51ms/epoch - 51ms/step\n",
      "Epoch 21/100\n",
      "1/1 - 0s - loss: 5.7522e-06 - accuracy: 0.0000e+00 - val_loss: 5.1557e-05 - val_accuracy: 0.0000e+00 - 42ms/epoch - 42ms/step\n",
      "Epoch 22/100\n",
      "1/1 - 0s - loss: 6.8166e-06 - accuracy: 0.0000e+00 - val_loss: 3.3490e-05 - val_accuracy: 0.0000e+00 - 45ms/epoch - 45ms/step\n",
      "Epoch 23/100\n",
      "1/1 - 0s - loss: 4.9522e-06 - accuracy: 0.0000e+00 - val_loss: 2.1396e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 24/100\n",
      "1/1 - 0s - loss: 6.3953e-06 - accuracy: 0.0000e+00 - val_loss: 4.3297e-05 - val_accuracy: 0.0000e+00 - 51ms/epoch - 51ms/step\n",
      "Epoch 25/100\n",
      "1/1 - 0s - loss: 5.5103e-06 - accuracy: 0.0000e+00 - val_loss: 4.1403e-05 - val_accuracy: 0.0000e+00 - 37ms/epoch - 37ms/step\n",
      "Epoch 26/100\n",
      "1/1 - 0s - loss: 5.2928e-06 - accuracy: 0.0000e+00 - val_loss: 2.2210e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 27/100\n",
      "1/1 - 0s - loss: 6.0892e-06 - accuracy: 0.0000e+00 - val_loss: 3.5538e-05 - val_accuracy: 0.0000e+00 - 47ms/epoch - 47ms/step\n",
      "Epoch 28/100\n",
      "1/1 - 0s - loss: 4.8824e-06 - accuracy: 0.0000e+00 - val_loss: 4.4761e-05 - val_accuracy: 0.0000e+00 - 39ms/epoch - 39ms/step\n",
      "Epoch 29/100\n",
      "1/1 - 0s - loss: 5.6348e-06 - accuracy: 0.0000e+00 - val_loss: 2.5170e-05 - val_accuracy: 0.0000e+00 - 39ms/epoch - 39ms/step\n",
      "Epoch 30/100\n",
      "1/1 - 0s - loss: 5.4154e-06 - accuracy: 0.0000e+00 - val_loss: 3.0799e-05 - val_accuracy: 0.0000e+00 - 38ms/epoch - 38ms/step\n",
      "Epoch 31/100\n",
      "1/1 - 0s - loss: 4.8376e-06 - accuracy: 0.0000e+00 - val_loss: 4.4319e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 32/100\n",
      "1/1 - 0s - loss: 5.5401e-06 - accuracy: 0.0000e+00 - val_loss: 2.7826e-05 - val_accuracy: 0.0000e+00 - 42ms/epoch - 42ms/step\n",
      "Epoch 33/100\n",
      "1/1 - 0s - loss: 4.9742e-06 - accuracy: 0.0000e+00 - val_loss: 2.8924e-05 - val_accuracy: 0.0000e+00 - 45ms/epoch - 45ms/step\n",
      "Epoch 34/100\n",
      "1/1 - 0s - loss: 4.8446e-06 - accuracy: 0.0000e+00 - val_loss: 4.2962e-05 - val_accuracy: 0.0000e+00 - 46ms/epoch - 46ms/step\n",
      "Epoch 35/100\n",
      "1/1 - 0s - loss: 5.3082e-06 - accuracy: 0.0000e+00 - val_loss: 2.8776e-05 - val_accuracy: 0.0000e+00 - 50ms/epoch - 50ms/step\n",
      "Epoch 36/100\n",
      "1/1 - 0s - loss: 4.7794e-06 - accuracy: 0.0000e+00 - val_loss: 2.8927e-05 - val_accuracy: 0.0000e+00 - 59ms/epoch - 59ms/step\n",
      "Epoch 37/100\n",
      "1/1 - 0s - loss: 4.7303e-06 - accuracy: 0.0000e+00 - val_loss: 4.1532e-05 - val_accuracy: 0.0000e+00 - 51ms/epoch - 51ms/step\n",
      "Epoch 38/100\n",
      "1/1 - 0s - loss: 5.0770e-06 - accuracy: 0.0000e+00 - val_loss: 2.8119e-05 - val_accuracy: 0.0000e+00 - 60ms/epoch - 60ms/step\n",
      "Epoch 39/100\n",
      "1/1 - 0s - loss: 4.7038e-06 - accuracy: 0.0000e+00 - val_loss: 3.0441e-05 - val_accuracy: 0.0000e+00 - 47ms/epoch - 47ms/step\n",
      "Epoch 40/100\n",
      "1/1 - 0s - loss: 4.5205e-06 - accuracy: 0.0000e+00 - val_loss: 3.9789e-05 - val_accuracy: 0.0000e+00 - 58ms/epoch - 58ms/step\n",
      "Epoch 41/100\n",
      "1/1 - 0s - loss: 4.8117e-06 - accuracy: 0.0000e+00 - val_loss: 2.6637e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 42/100\n",
      "1/1 - 0s - loss: 4.6913e-06 - accuracy: 0.0000e+00 - val_loss: 3.3502e-05 - val_accuracy: 0.0000e+00 - 42ms/epoch - 42ms/step\n",
      "Epoch 43/100\n",
      "1/1 - 0s - loss: 4.3718e-06 - accuracy: 0.0000e+00 - val_loss: 3.6353e-05 - val_accuracy: 0.0000e+00 - 34ms/epoch - 34ms/step\n",
      "Epoch 44/100\n",
      "1/1 - 0s - loss: 4.4477e-06 - accuracy: 0.0000e+00 - val_loss: 2.5976e-05 - val_accuracy: 0.0000e+00 - 36ms/epoch - 36ms/step\n",
      "Epoch 45/100\n",
      "1/1 - 0s - loss: 4.6064e-06 - accuracy: 0.0000e+00 - val_loss: 3.7387e-05 - val_accuracy: 0.0000e+00 - 38ms/epoch - 38ms/step\n",
      "Epoch 46/100\n",
      "1/1 - 0s - loss: 4.4496e-06 - accuracy: 0.0000e+00 - val_loss: 3.0196e-05 - val_accuracy: 0.0000e+00 - 38ms/epoch - 38ms/step\n",
      "Epoch 47/100\n",
      "1/1 - 0s - loss: 4.2056e-06 - accuracy: 0.0000e+00 - val_loss: 2.9329e-05 - val_accuracy: 0.0000e+00 - 37ms/epoch - 37ms/step\n",
      "Epoch 48/100\n",
      "1/1 - 0s - loss: 4.1854e-06 - accuracy: 0.0000e+00 - val_loss: 3.6858e-05 - val_accuracy: 0.0000e+00 - 36ms/epoch - 36ms/step\n",
      "Epoch 49/100\n",
      "1/1 - 0s - loss: 4.3023e-06 - accuracy: 0.0000e+00 - val_loss: 2.5500e-05 - val_accuracy: 0.0000e+00 - 38ms/epoch - 38ms/step\n",
      "Epoch 50/100\n",
      "1/1 - 0s - loss: 4.3387e-06 - accuracy: 0.0000e+00 - val_loss: 3.6852e-05 - val_accuracy: 0.0000e+00 - 40ms/epoch - 40ms/step\n",
      "Epoch 51/100\n",
      "1/1 - 0s - loss: 4.2316e-06 - accuracy: 0.0000e+00 - val_loss: 2.7113e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 52/100\n",
      "1/1 - 0s - loss: 4.0564e-06 - accuracy: 0.0000e+00 - val_loss: 3.2804e-05 - val_accuracy: 0.0000e+00 - 35ms/epoch - 35ms/step\n",
      "Epoch 53/100\n",
      "1/1 - 0s - loss: 3.9066e-06 - accuracy: 0.0000e+00 - val_loss: 3.0245e-05 - val_accuracy: 0.0000e+00 - 37ms/epoch - 37ms/step\n",
      "Epoch 54/100\n",
      "1/1 - 0s - loss: 3.8121e-06 - accuracy: 0.0000e+00 - val_loss: 2.8958e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 55/100\n",
      "1/1 - 0s - loss: 3.7651e-06 - accuracy: 0.0000e+00 - val_loss: 3.3233e-05 - val_accuracy: 0.0000e+00 - 35ms/epoch - 35ms/step\n",
      "Epoch 56/100\n",
      "1/1 - 0s - loss: 3.7623e-06 - accuracy: 0.0000e+00 - val_loss: 2.4965e-05 - val_accuracy: 0.0000e+00 - 45ms/epoch - 45ms/step\n",
      "Epoch 57/100\n",
      "1/1 - 0s - loss: 3.8478e-06 - accuracy: 0.0000e+00 - val_loss: 3.9063e-05 - val_accuracy: 0.0000e+00 - 45ms/epoch - 45ms/step\n",
      "Epoch 58/100\n",
      "1/1 - 0s - loss: 4.2065e-06 - accuracy: 0.0000e+00 - val_loss: 1.6577e-05 - val_accuracy: 0.0000e+00 - 39ms/epoch - 39ms/step\n",
      "Epoch 59/100\n",
      "1/1 - 0s - loss: 5.3604e-06 - accuracy: 0.0000e+00 - val_loss: 6.0794e-05 - val_accuracy: 0.0000e+00 - 49ms/epoch - 49ms/step\n",
      "Epoch 60/100\n",
      "1/1 - 0s - loss: 9.1781e-06 - accuracy: 0.0000e+00 - val_loss: 2.6987e-06 - val_accuracy: 0.0000e+00 - 45ms/epoch - 45ms/step\n",
      "Epoch 61/100\n",
      "1/1 - 0s - loss: 1.7892e-05 - accuracy: 0.0000e+00 - val_loss: 9.5652e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 62/100\n",
      "1/1 - 0s - loss: 2.2786e-05 - accuracy: 0.0000e+00 - val_loss: 8.1486e-06 - val_accuracy: 0.0000e+00 - 42ms/epoch - 42ms/step\n",
      "Epoch 63/100\n",
      "1/1 - 0s - loss: 9.8680e-06 - accuracy: 0.0000e+00 - val_loss: 1.8924e-05 - val_accuracy: 0.0000e+00 - 46ms/epoch - 46ms/step\n",
      "Epoch 64/100\n",
      "1/1 - 0s - loss: 4.4085e-06 - accuracy: 0.0000e+00 - val_loss: 7.5561e-05 - val_accuracy: 0.0000e+00 - 47ms/epoch - 47ms/step\n",
      "Epoch 65/100\n",
      "1/1 - 0s - loss: 1.4365e-05 - accuracy: 0.0000e+00 - val_loss: 1.1416e-05 - val_accuracy: 0.0000e+00 - 45ms/epoch - 45ms/step\n",
      "Epoch 66/100\n",
      "1/1 - 0s - loss: 7.4278e-06 - accuracy: 0.0000e+00 - val_loss: 1.6513e-05 - val_accuracy: 0.0000e+00 - 64ms/epoch - 64ms/step\n",
      "Epoch 67/100\n",
      "1/1 - 0s - loss: 5.1131e-06 - accuracy: 0.0000e+00 - val_loss: 6.6110e-05 - val_accuracy: 0.0000e+00 - 41ms/epoch - 41ms/step\n",
      "Epoch 68/100\n",
      "1/1 - 0s - loss: 1.0947e-05 - accuracy: 0.0000e+00 - val_loss: 2.2260e-05 - val_accuracy: 0.0000e+00 - 40ms/epoch - 40ms/step\n",
      "Epoch 69/100\n",
      "1/1 - 0s - loss: 3.7260e-06 - accuracy: 0.0000e+00 - val_loss: 1.0964e-05 - val_accuracy: 0.0000e+00 - 40ms/epoch - 40ms/step\n",
      "Epoch 70/100\n",
      "1/1 - 0s - loss: 7.5766e-06 - accuracy: 0.0000e+00 - val_loss: 5.2467e-05 - val_accuracy: 0.0000e+00 - 47ms/epoch - 47ms/step\n",
      "Epoch 71/100\n",
      "1/1 - 0s - loss: 6.7335e-06 - accuracy: 0.0000e+00 - val_loss: 3.7894e-05 - val_accuracy: 0.0000e+00 - 41ms/epoch - 41ms/step\n",
      "Epoch 72/100\n",
      "1/1 - 0s - loss: 3.8374e-06 - accuracy: 0.0000e+00 - val_loss: 1.0782e-05 - val_accuracy: 0.0000e+00 - 41ms/epoch - 41ms/step\n",
      "Epoch 73/100\n",
      "1/1 - 0s - loss: 7.6059e-06 - accuracy: 0.0000e+00 - val_loss: 3.3139e-05 - val_accuracy: 0.0000e+00 - 38ms/epoch - 38ms/step\n",
      "Epoch 74/100\n",
      "1/1 - 0s - loss: 3.3326e-06 - accuracy: 0.0000e+00 - val_loss: 4.9641e-05 - val_accuracy: 0.0000e+00 - 40ms/epoch - 40ms/step\n",
      "Epoch 75/100\n",
      "1/1 - 0s - loss: 6.0124e-06 - accuracy: 0.0000e+00 - val_loss: 1.7850e-05 - val_accuracy: 0.0000e+00 - 41ms/epoch - 41ms/step\n",
      "Epoch 76/100\n",
      "1/1 - 0s - loss: 4.4419e-06 - accuracy: 0.0000e+00 - val_loss: 1.9143e-05 - val_accuracy: 0.0000e+00 - 39ms/epoch - 39ms/step\n",
      "Epoch 77/100\n",
      "1/1 - 0s - loss: 4.0447e-06 - accuracy: 0.0000e+00 - val_loss: 4.6365e-05 - val_accuracy: 0.0000e+00 - 40ms/epoch - 40ms/step\n",
      "Epoch 78/100\n",
      "1/1 - 0s - loss: 5.2370e-06 - accuracy: 0.0000e+00 - val_loss: 3.0250e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 79/100\n",
      "1/1 - 0s - loss: 3.0388e-06 - accuracy: 0.0000e+00 - val_loss: 1.5232e-05 - val_accuracy: 0.0000e+00 - 38ms/epoch - 38ms/step\n",
      "Epoch 80/100\n",
      "1/1 - 0s - loss: 5.0055e-06 - accuracy: 0.0000e+00 - val_loss: 3.4351e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 81/100\n",
      "1/1 - 0s - loss: 3.2371e-06 - accuracy: 0.0000e+00 - val_loss: 3.9589e-05 - val_accuracy: 0.0000e+00 - 39ms/epoch - 39ms/step\n",
      "Epoch 82/100\n",
      "1/1 - 0s - loss: 3.8913e-06 - accuracy: 0.0000e+00 - val_loss: 1.8265e-05 - val_accuracy: 0.0000e+00 - 40ms/epoch - 40ms/step\n",
      "Epoch 83/100\n",
      "1/1 - 0s - loss: 3.8640e-06 - accuracy: 0.0000e+00 - val_loss: 2.3918e-05 - val_accuracy: 0.0000e+00 - 36ms/epoch - 36ms/step\n",
      "Epoch 84/100\n",
      "1/1 - 0s - loss: 2.9213e-06 - accuracy: 0.0000e+00 - val_loss: 4.0451e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 85/100\n",
      "1/1 - 0s - loss: 3.9903e-06 - accuracy: 0.0000e+00 - val_loss: 2.4433e-05 - val_accuracy: 0.0000e+00 - 37ms/epoch - 37ms/step\n",
      "Epoch 86/100\n",
      "1/1 - 0s - loss: 2.7420e-06 - accuracy: 0.0000e+00 - val_loss: 1.8805e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 87/100\n",
      "1/1 - 0s - loss: 3.3601e-06 - accuracy: 0.0000e+00 - val_loss: 3.5868e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 88/100\n",
      "1/1 - 0s - loss: 3.2090e-06 - accuracy: 0.0000e+00 - val_loss: 2.9295e-05 - val_accuracy: 0.0000e+00 - 42ms/epoch - 42ms/step\n",
      "Epoch 89/100\n",
      "1/1 - 0s - loss: 2.5399e-06 - accuracy: 0.0000e+00 - val_loss: 1.7822e-05 - val_accuracy: 0.0000e+00 - 41ms/epoch - 41ms/step\n",
      "Epoch 90/100\n",
      "1/1 - 0s - loss: 3.2666e-06 - accuracy: 0.0000e+00 - val_loss: 3.0914e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 91/100\n",
      "1/1 - 0s - loss: 2.5365e-06 - accuracy: 0.0000e+00 - val_loss: 3.1302e-05 - val_accuracy: 0.0000e+00 - 49ms/epoch - 49ms/step\n",
      "Epoch 92/100\n",
      "1/1 - 0s - loss: 2.5192e-06 - accuracy: 0.0000e+00 - val_loss: 1.8187e-05 - val_accuracy: 0.0000e+00 - 55ms/epoch - 55ms/step\n",
      "Epoch 93/100\n",
      "1/1 - 0s - loss: 2.8523e-06 - accuracy: 0.0000e+00 - val_loss: 2.7803e-05 - val_accuracy: 0.0000e+00 - 46ms/epoch - 46ms/step\n",
      "Epoch 94/100\n",
      "1/1 - 0s - loss: 2.1641e-06 - accuracy: 0.0000e+00 - val_loss: 3.0873e-05 - val_accuracy: 0.0000e+00 - 48ms/epoch - 48ms/step\n",
      "Epoch 95/100\n",
      "1/1 - 0s - loss: 2.3261e-06 - accuracy: 0.0000e+00 - val_loss: 1.8322e-05 - val_accuracy: 0.0000e+00 - 52ms/epoch - 52ms/step\n",
      "Epoch 96/100\n",
      "1/1 - 0s - loss: 2.4701e-06 - accuracy: 0.0000e+00 - val_loss: 2.6903e-05 - val_accuracy: 0.0000e+00 - 44ms/epoch - 44ms/step\n",
      "Epoch 97/100\n",
      "1/1 - 0s - loss: 1.9183e-06 - accuracy: 0.0000e+00 - val_loss: 2.8665e-05 - val_accuracy: 0.0000e+00 - 41ms/epoch - 41ms/step\n",
      "Epoch 98/100\n",
      "1/1 - 0s - loss: 1.9633e-06 - accuracy: 0.0000e+00 - val_loss: 1.7890e-05 - val_accuracy: 0.0000e+00 - 46ms/epoch - 46ms/step\n",
      "Epoch 99/100\n",
      "1/1 - 0s - loss: 2.1601e-06 - accuracy: 0.0000e+00 - val_loss: 2.7848e-05 - val_accuracy: 0.0000e+00 - 43ms/epoch - 43ms/step\n",
      "Epoch 100/100\n",
      "1/1 - 0s - loss: 1.7814e-06 - accuracy: 0.0000e+00 - val_loss: 2.4471e-05 - val_accuracy: 0.0000e+00 - 36ms/epoch - 36ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.src.callbacks.History at 0x7ff543ed59a0>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(train_x,train_y,validation_data=(test_x,test_y),verbose=2,epochs=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "9e0ff26e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 101ms/step\n"
     ]
    }
   ],
   "source": [
    "prediction = model.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "af32d6eb",
   "metadata": {},
   "outputs": [],
   "source": [
    "inverse_transformed_prediction = ct.named_transformers_['std'].inverse_transform(\n",
    "     np.hstack([get_biomass_hist_scaled(24.671875,71.3125).iloc[:,3:-1],prediction])\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "f73ab401",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 8.478675 ,  8.86512  ,  9.204971 ,  6.0252056, 10.788841 ,\n",
       "         6.6471505,  7.3883886,  9.549628 ]], dtype=float32)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inverse_transformed_prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "133f35c2",
   "metadata": {},
   "source": [
    "# LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "f2316d10",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " lstm (LSTM)                 (None, 64)                17664     \n",
      "                                                                 \n",
      " dense (Dense)               (None, 32)                2080      \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 1)                 33        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 19777 (77.25 KB)\n",
      "Trainable params: 19777 (77.25 KB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# input_layer = Input(shape=(4,),name='input')\n",
    "# dense_layer = LSTM(64,name='lstm',input_shape=(None,4))(input_layer)\n",
    "# dense_layer = Dense(32,name='d1')(dese_layer)\n",
    "# output_layer = Dense(1,name='output')(dese_layer)\n",
    "# model = Model(inputs=input_layer, outputs=output_layer)\n",
    "model=Sequential()\n",
    "model.add(LSTM(64,input_shape=(None,4)))\n",
    "model.add(Dense(32))\n",
    "# model.add(Dense(256))\n",
    "# model.add(Dense(16))\n",
    "# model.add(Dense(64))\n",
    "# model.add(Dense(8))\n",
    "model.add(Dense(1))\n",
    "model.compile(loss='mean_squared_error',optimizer='adam')\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "46a41a98",
   "metadata": {},
   "outputs": [],
   "source": [
    "trainX = np.reshape(train_x,(train_x.shape[0],1,train_x.shape[1]))    \n",
    "testX = np.reshape(test_x,(test_x.shape[0],1,test_x.shape[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "ec3c3157",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 - 3s - loss: 1.4970e-04 - val_loss: 1.8122e-05 - 3s/epoch - 3s/step\n",
      "Epoch 2/100\n",
      "1/1 - 0s - loss: 1.0307e-05 - val_loss: 1.4699e-04 - 54ms/epoch - 54ms/step\n",
      "Epoch 3/100\n",
      "1/1 - 0s - loss: 4.2543e-05 - val_loss: 2.1388e-04 - 44ms/epoch - 44ms/step\n",
      "Epoch 4/100\n",
      "1/1 - 0s - loss: 7.8702e-05 - val_loss: 1.7385e-04 - 44ms/epoch - 44ms/step\n",
      "Epoch 5/100\n",
      "1/1 - 0s - loss: 5.6413e-05 - val_loss: 9.7224e-05 - 44ms/epoch - 44ms/step\n",
      "Epoch 6/100\n",
      "1/1 - 0s - loss: 2.0619e-05 - val_loss: 3.6086e-05 - 44ms/epoch - 44ms/step\n",
      "Epoch 7/100\n",
      "1/1 - 0s - loss: 6.6623e-06 - val_loss: 7.3533e-06 - 50ms/epoch - 50ms/step\n",
      "Epoch 8/100\n",
      "1/1 - 0s - loss: 1.8159e-05 - val_loss: 6.0483e-07 - 47ms/epoch - 47ms/step\n",
      "Epoch 9/100\n",
      "1/1 - 0s - loss: 3.4932e-05 - val_loss: 2.0812e-07 - 48ms/epoch - 48ms/step\n",
      "Epoch 10/100\n",
      "1/1 - 0s - loss: 3.8371e-05 - val_loss: 2.2535e-06 - 46ms/epoch - 46ms/step\n",
      "Epoch 11/100\n",
      "1/1 - 0s - loss: 2.7617e-05 - val_loss: 1.1890e-05 - 43ms/epoch - 43ms/step\n",
      "Epoch 12/100\n",
      "1/1 - 0s - loss: 1.3494e-05 - val_loss: 3.3284e-05 - 54ms/epoch - 54ms/step\n",
      "Epoch 13/100\n",
      "1/1 - 0s - loss: 6.5972e-06 - val_loss: 6.2821e-05 - 49ms/epoch - 49ms/step\n",
      "Epoch 14/100\n",
      "1/1 - 0s - loss: 9.8652e-06 - val_loss: 8.9264e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 15/100\n",
      "1/1 - 0s - loss: 1.7788e-05 - val_loss: 1.0108e-04 - 46ms/epoch - 46ms/step\n",
      "Epoch 16/100\n",
      "1/1 - 0s - loss: 2.2250e-05 - val_loss: 9.4435e-05 - 66ms/epoch - 66ms/step\n",
      "Epoch 17/100\n",
      "1/1 - 0s - loss: 1.9708e-05 - val_loss: 7.4428e-05 - 52ms/epoch - 52ms/step\n",
      "Epoch 18/100\n",
      "1/1 - 0s - loss: 1.2950e-05 - val_loss: 5.0347e-05 - 50ms/epoch - 50ms/step\n",
      "Epoch 19/100\n",
      "1/1 - 0s - loss: 7.4292e-06 - val_loss: 3.0122e-05 - 45ms/epoch - 45ms/step\n",
      "Epoch 20/100\n",
      "1/1 - 0s - loss: 6.6306e-06 - val_loss: 1.7266e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 21/100\n",
      "1/1 - 0s - loss: 9.8407e-06 - val_loss: 1.1285e-05 - 57ms/epoch - 57ms/step\n",
      "Epoch 22/100\n",
      "1/1 - 0s - loss: 1.3409e-05 - val_loss: 1.0368e-05 - 65ms/epoch - 65ms/step\n",
      "Epoch 23/100\n",
      "1/1 - 0s - loss: 1.4113e-05 - val_loss: 1.3700e-05 - 55ms/epoch - 55ms/step\n",
      "Epoch 24/100\n",
      "1/1 - 0s - loss: 1.1593e-05 - val_loss: 2.1537e-05 - 44ms/epoch - 44ms/step\n",
      "Epoch 25/100\n",
      "1/1 - 0s - loss: 8.0807e-06 - val_loss: 3.3588e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 26/100\n",
      "1/1 - 0s - loss: 6.2127e-06 - val_loss: 4.7521e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 27/100\n",
      "1/1 - 0s - loss: 6.9293e-06 - val_loss: 5.9161e-05 - 42ms/epoch - 42ms/step\n",
      "Epoch 28/100\n",
      "1/1 - 0s - loss: 8.9758e-06 - val_loss: 6.4487e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 29/100\n",
      "1/1 - 0s - loss: 1.0235e-05 - val_loss: 6.1940e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 30/100\n",
      "1/1 - 0s - loss: 9.6081e-06 - val_loss: 5.3148e-05 - 51ms/epoch - 51ms/step\n",
      "Epoch 31/100\n",
      "1/1 - 0s - loss: 7.7543e-06 - val_loss: 4.1722e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 32/100\n",
      "1/1 - 0s - loss: 6.2496e-06 - val_loss: 3.1247e-05 - 45ms/epoch - 45ms/step\n",
      "Epoch 33/100\n",
      "1/1 - 0s - loss: 6.1277e-06 - val_loss: 2.3890e-05 - 47ms/epoch - 47ms/step\n",
      "Epoch 34/100\n",
      "1/1 - 0s - loss: 7.1050e-06 - val_loss: 2.0287e-05 - 54ms/epoch - 54ms/step\n",
      "Epoch 35/100\n",
      "1/1 - 0s - loss: 8.0255e-06 - val_loss: 2.0270e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 36/100\n",
      "1/1 - 0s - loss: 7.9850e-06 - val_loss: 2.3511e-05 - 53ms/epoch - 53ms/step\n",
      "Epoch 37/100\n",
      "1/1 - 0s - loss: 7.0658e-06 - val_loss: 2.9481e-05 - 65ms/epoch - 65ms/step\n",
      "Epoch 38/100\n",
      "1/1 - 0s - loss: 6.1068e-06 - val_loss: 3.6987e-05 - 50ms/epoch - 50ms/step\n",
      "Epoch 39/100\n",
      "1/1 - 0s - loss: 5.8544e-06 - val_loss: 4.4026e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 40/100\n",
      "1/1 - 0s - loss: 6.3178e-06 - val_loss: 4.8385e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 41/100\n",
      "1/1 - 0s - loss: 6.8685e-06 - val_loss: 4.8702e-05 - 65ms/epoch - 65ms/step\n",
      "Epoch 42/100\n",
      "1/1 - 0s - loss: 6.9054e-06 - val_loss: 4.5162e-05 - 52ms/epoch - 52ms/step\n",
      "Epoch 43/100\n",
      "1/1 - 0s - loss: 6.4022e-06 - val_loss: 3.9304e-05 - 65ms/epoch - 65ms/step\n",
      "Epoch 44/100\n",
      "1/1 - 0s - loss: 5.8409e-06 - val_loss: 3.3155e-05 - 51ms/epoch - 51ms/step\n",
      "Epoch 45/100\n",
      "1/1 - 0s - loss: 5.6829e-06 - val_loss: 2.8355e-05 - 44ms/epoch - 44ms/step\n",
      "Epoch 46/100\n",
      "1/1 - 0s - loss: 5.9394e-06 - val_loss: 2.5780e-05 - 39ms/epoch - 39ms/step\n",
      "Epoch 47/100\n",
      "1/1 - 0s - loss: 6.2312e-06 - val_loss: 2.5653e-05 - 43ms/epoch - 43ms/step\n",
      "Epoch 48/100\n",
      "1/1 - 0s - loss: 6.2111e-06 - val_loss: 2.7766e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 49/100\n",
      "1/1 - 0s - loss: 5.8929e-06 - val_loss: 3.1525e-05 - 50ms/epoch - 50ms/step\n",
      "Epoch 50/100\n",
      "1/1 - 0s - loss: 5.5811e-06 - val_loss: 3.5897e-05 - 45ms/epoch - 45ms/step\n",
      "Epoch 51/100\n",
      "1/1 - 0s - loss: 5.5270e-06 - val_loss: 3.9545e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 52/100\n",
      "1/1 - 0s - loss: 5.6882e-06 - val_loss: 4.1286e-05 - 45ms/epoch - 45ms/step\n",
      "Epoch 53/100\n",
      "1/1 - 0s - loss: 5.8150e-06 - val_loss: 4.0620e-05 - 43ms/epoch - 43ms/step\n",
      "Epoch 54/100\n",
      "1/1 - 0s - loss: 5.7353e-06 - val_loss: 3.7954e-05 - 42ms/epoch - 42ms/step\n",
      "Epoch 55/100\n",
      "1/1 - 0s - loss: 5.5182e-06 - val_loss: 3.4348e-05 - 42ms/epoch - 42ms/step\n",
      "Epoch 56/100\n",
      "1/1 - 0s - loss: 5.3647e-06 - val_loss: 3.0990e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 57/100\n",
      "1/1 - 0s - loss: 5.3782e-06 - val_loss: 2.8773e-05 - 41ms/epoch - 41ms/step\n",
      "Epoch 58/100\n",
      "1/1 - 0s - loss: 5.4710e-06 - val_loss: 2.8125e-05 - 55ms/epoch - 55ms/step\n",
      "Epoch 59/100\n",
      "1/1 - 0s - loss: 5.4876e-06 - val_loss: 2.9039e-05 - 62ms/epoch - 62ms/step\n",
      "Epoch 60/100\n",
      "1/1 - 0s - loss: 5.3802e-06 - val_loss: 3.1132e-05 - 55ms/epoch - 55ms/step\n",
      "Epoch 61/100\n",
      "1/1 - 0s - loss: 5.2436e-06 - val_loss: 3.3700e-05 - 41ms/epoch - 41ms/step\n",
      "Epoch 62/100\n",
      "1/1 - 0s - loss: 5.1895e-06 - val_loss: 3.5869e-05 - 47ms/epoch - 47ms/step\n",
      "Epoch 63/100\n",
      "1/1 - 0s - loss: 5.2218e-06 - val_loss: 3.6889e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 64/100\n",
      "1/1 - 0s - loss: 5.2487e-06 - val_loss: 3.6445e-05 - 47ms/epoch - 47ms/step\n",
      "Epoch 65/100\n",
      "1/1 - 0s - loss: 5.2003e-06 - val_loss: 3.4788e-05 - 42ms/epoch - 42ms/step\n",
      "Epoch 66/100\n",
      "1/1 - 0s - loss: 5.1036e-06 - val_loss: 3.2581e-05 - 43ms/epoch - 43ms/step\n",
      "Epoch 67/100\n",
      "1/1 - 0s - loss: 5.0366e-06 - val_loss: 3.0586e-05 - 47ms/epoch - 47ms/step\n",
      "Epoch 68/100\n",
      "1/1 - 0s - loss: 5.0308e-06 - val_loss: 2.9381e-05 - 47ms/epoch - 47ms/step\n",
      "Epoch 69/100\n",
      "1/1 - 0s - loss: 5.0421e-06 - val_loss: 2.9227e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 70/100\n",
      "1/1 - 0s - loss: 5.0143e-06 - val_loss: 3.0047e-05 - 51ms/epoch - 51ms/step\n",
      "Epoch 71/100\n",
      "1/1 - 0s - loss: 4.9467e-06 - val_loss: 3.1472e-05 - 47ms/epoch - 47ms/step\n",
      "Epoch 72/100\n",
      "1/1 - 0s - loss: 4.8861e-06 - val_loss: 3.2944e-05 - 53ms/epoch - 53ms/step\n",
      "Epoch 73/100\n",
      "1/1 - 0s - loss: 4.8636e-06 - val_loss: 3.3895e-05 - 47ms/epoch - 47ms/step\n",
      "Epoch 74/100\n",
      "1/1 - 0s - loss: 4.8593e-06 - val_loss: 3.3968e-05 - 48ms/epoch - 48ms/step\n",
      "Epoch 75/100\n",
      "1/1 - 0s - loss: 4.8347e-06 - val_loss: 3.3168e-05 - 50ms/epoch - 50ms/step\n",
      "Epoch 76/100\n",
      "1/1 - 0s - loss: 4.7826e-06 - val_loss: 3.1837e-05 - 53ms/epoch - 53ms/step\n",
      "Epoch 77/100\n",
      "1/1 - 0s - loss: 4.7313e-06 - val_loss: 3.0478e-05 - 49ms/epoch - 49ms/step\n",
      "Epoch 78/100\n",
      "1/1 - 0s - loss: 4.7035e-06 - val_loss: 2.9543e-05 - 45ms/epoch - 45ms/step\n",
      "Epoch 79/100\n",
      "1/1 - 0s - loss: 4.6886e-06 - val_loss: 2.9283e-05 - 50ms/epoch - 50ms/step\n",
      "Epoch 80/100\n",
      "1/1 - 0s - loss: 4.6614e-06 - val_loss: 2.9689e-05 - 54ms/epoch - 54ms/step\n",
      "Epoch 81/100\n",
      "1/1 - 0s - loss: 4.6168e-06 - val_loss: 3.0523e-05 - 54ms/epoch - 54ms/step\n",
      "Epoch 82/100\n",
      "1/1 - 0s - loss: 4.5729e-06 - val_loss: 3.1398e-05 - 53ms/epoch - 53ms/step\n",
      "Epoch 83/100\n",
      "1/1 - 0s - loss: 4.5440e-06 - val_loss: 3.1926e-05 - 50ms/epoch - 50ms/step\n",
      "Epoch 84/100\n",
      "1/1 - 0s - loss: 4.5223e-06 - val_loss: 3.1878e-05 - 54ms/epoch - 54ms/step\n",
      "Epoch 85/100\n",
      "1/1 - 0s - loss: 4.4917e-06 - val_loss: 3.1276e-05 - 49ms/epoch - 49ms/step\n",
      "Epoch 86/100\n",
      "1/1 - 0s - loss: 4.4506e-06 - val_loss: 3.0369e-05 - 57ms/epoch - 57ms/step\n",
      "Epoch 87/100\n",
      "1/1 - 0s - loss: 4.4119e-06 - val_loss: 2.9503e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 88/100\n",
      "1/1 - 0s - loss: 4.3830e-06 - val_loss: 2.8973e-05 - 52ms/epoch - 52ms/step\n",
      "Epoch 89/100\n",
      "1/1 - 0s - loss: 4.3565e-06 - val_loss: 2.8912e-05 - 49ms/epoch - 49ms/step\n",
      "Epoch 90/100\n",
      "1/1 - 0s - loss: 4.3230e-06 - val_loss: 2.9256e-05 - 44ms/epoch - 44ms/step\n",
      "Epoch 91/100\n",
      "1/1 - 0s - loss: 4.2843e-06 - val_loss: 2.9788e-05 - 42ms/epoch - 42ms/step\n",
      "Epoch 92/100\n",
      "1/1 - 0s - loss: 4.2491e-06 - val_loss: 3.0220e-05 - 51ms/epoch - 51ms/step\n",
      "Epoch 93/100\n",
      "1/1 - 0s - loss: 4.2197e-06 - val_loss: 3.0324e-05 - 65ms/epoch - 65ms/step\n",
      "Epoch 94/100\n",
      "1/1 - 0s - loss: 4.1895e-06 - val_loss: 3.0031e-05 - 49ms/epoch - 49ms/step\n",
      "Epoch 95/100\n",
      "1/1 - 0s - loss: 4.1542e-06 - val_loss: 2.9452e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 96/100\n",
      "1/1 - 0s - loss: 4.1174e-06 - val_loss: 2.8815e-05 - 56ms/epoch - 56ms/step\n",
      "Epoch 97/100\n",
      "1/1 - 0s - loss: 4.0843e-06 - val_loss: 2.8352e-05 - 46ms/epoch - 46ms/step\n",
      "Epoch 98/100\n",
      "1/1 - 0s - loss: 4.0536e-06 - val_loss: 2.8197e-05 - 43ms/epoch - 43ms/step\n",
      "Epoch 99/100\n",
      "1/1 - 0s - loss: 4.0206e-06 - val_loss: 2.8338e-05 - 49ms/epoch - 49ms/step\n",
      "Epoch 100/100\n",
      "1/1 - 0s - loss: 3.9846e-06 - val_loss: 2.8634e-05 - 43ms/epoch - 43ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.src.callbacks.History at 0x7ff53dce4400>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(trainX,train_y,validation_data=(testX,test_y),verbose=2,epochs=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "17202def",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 297ms/step\n"
     ]
    }
   ],
   "source": [
    "prediction = model.predict(testX)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "7bd84dfb",
   "metadata": {},
   "outputs": [],
   "source": [
    "inverse_transformed_prediction = ct.named_transformers_['std'].inverse_transform(\n",
    "     np.hstack([get_biomass_hist_scaled(24.671875,71.3125).iloc[:,3:-1],prediction])\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "0c626d42",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 8.478675 ,  8.86512  ,  9.204971 ,  6.0252056, 10.788841 ,\n",
       "         6.6471505,  7.3883886,  9.906676 ]], dtype=float32)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inverse_transformed_prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9307e3d3",
   "metadata": {},
   "source": [
    "## Exporting scaled data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "1e17dc48",
   "metadata": {},
   "outputs": [],
   "source": [
    "transformed_df.to_csv('scaled.csv')"
   ]
  }
 ],
 "metadata": {
  "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": "3.9.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}