Model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Machine Learning Model \n",
    "# LSTM\n",
    "# Train Data\n",
    "# x_train lat,lon,distance,speed,time diff\n",
    "# y_train pred_lat,pred_lon\n",
    "\n",
    "# !pip install pandas\n",
    "# !pip install scikit-learn\n",
    "# !pip install matplotlib\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for kaggle!!--------------------------------\n",
    "# import os\n",
    "# for dirname, _, filenames in os.walk('/kaggle/input'):\n",
    "#     for filename in filenames:\n",
    "#         print(os.path.join(dirname, filename))\n",
    "# end for kaggle!!------------------------------\n",
    "\n",
    "\n",
    "df = pd.read_csv('Dataset.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_x = df[['Longitude','Latitude','Time_diff','Haversine_Distance','Calculated_Speed']]\n",
    "train_y=df[['Pred_Lat','Pred_Lon']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        Longitude   Latitude  Time_diff  Haversine_Distance  Calculated_Speed\n",
      "19825 -105.434017  41.233209        8.0            3.209280          0.401160\n",
      "19826 -105.434054  41.233212        5.0            3.111959          0.622392\n",
      "19827 -105.434079  41.233234       27.0            3.217884          0.119181\n",
      "        Pred_Lat    Pred_Lon\n",
      "19825  41.233212 -105.434054\n",
      "19826  41.233234 -105.434079\n",
      "19827  41.233234 -105.434079\n"
     ]
    }
   ],
   "source": [
    "\n",
    "print(train_x.tail(3))\n",
    "\n",
    "\n",
    "print(train_y.tail(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(19828, 5)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "# convert pandas dataframe to numpy array\n",
    "\n",
    "x_data = train_x.values\n",
    "x_data = np.array(x_data)\n",
    "# x_data = x_data.reshape(19828,5,1)\n",
    "x_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(19828, 2)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_data = train_y.values\n",
    "y_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_data shape : (19828, 5)\n",
      "new X_data shape : (19828, 5, 1)\n",
      "y_data shape : (19828, 2)\n",
      "new y_data shape : (19828, 2)\n"
     ]
    }
   ],
   "source": [
    "# Preporcessing Normalizing Valuse\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "sc = MinMaxScaler(feature_range=(-1,1))\n",
    "# new_df= x_data[].reshape(x_data.shape[0],5)\n",
    "new_x_data = sc.fit_transform(x_data)\n",
    "new_x_data = new_x_data.reshape(19828,5,1)\n",
    "\n",
    "print('X_data shape :', x_data.shape)\n",
    "print('new X_data shape :', new_x_data.shape)\n",
    "\n",
    "\n",
    "\n",
    "new_y_data = sc.fit_transform(y_data)\n",
    "\n",
    "print('y_data shape :', y_data.shape)\n",
    "print('new y_data shape :', new_y_data.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# new_x_data.reshape(new_x_data.shape[0],5,1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for inversing transform\n",
    "\n",
    "\n",
    "new_df.reshape(19828,5)\n",
    "new_df = sc.inverse_transform(new_df)\n",
    "new_df.reshape(new_df.shape[0],5,1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "\n",
    "\n",
    "from tensorflow.python.keras.models import Sequential\n",
    "from tensorflow.python.keras.layers.recurrent import LSTM,RNN\n",
    "from tensorflow.python.keras.layers.core import Dense, Activation, Dropout\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.utils import shuffle\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, x_test, y_train, y_test = train_test_split(new_x_data, new_y_data, test_size=0.33, random_state=42)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(13284, 5, 1)\n",
      "(13284, 2)\n",
      "(6544, 5, 1)\n",
      "(6544, 2)\n"
     ]
    }
   ],
   "source": [
    "print(x_train.shape)\n",
    "print(y_train.shape)\n",
    "print(x_test.shape)\n",
    "print(y_test.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "lstm_6 (LSTM)                (None, 5, 50)             10400     \n",
      "_________________________________________________________________\n",
      "dropout_6 (Dropout)          (None, 5, 50)             0         \n",
      "_________________________________________________________________\n",
      "lstm_7 (LSTM)                (None, 5, 100)            60400     \n",
      "_________________________________________________________________\n",
      "dropout_7 (Dropout)          (None, 5, 100)            0         \n",
      "_________________________________________________________________\n",
      "lstm_8 (LSTM)                (None, 50)                30200     \n",
      "_________________________________________________________________\n",
      "dropout_8 (Dropout)          (None, 50)                0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 2)                 102       \n",
      "=================================================================\n",
      "Total params: 101,102\n",
      "Trainable params: 101,102\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "\n",
    "# first layer\n",
    "model.add(LSTM(units=50, batch_input_shape=(None,5,1),return_sequences=True))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "# second layer\n",
    "# model.add(Dense(128, 1))\n",
    "model.add(LSTM(units=100,return_sequences=True))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "# third layer\n",
    "model.add(LSTM(units=50,return_sequences=False))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "# fourth dense layer\n",
    "model.add(Dense(units=2))\n",
    "\n",
    "\n",
    "model.compile(loss='mean_absolute_error',optimizer='adam',metrics=['accuracy'])\n",
    "model.summary()\n",
    "          \n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 13284 samples, validate on 6544 samples\n",
      "Epoch 1/50\n",
      "13284/13284 [==============================] - 18s 1ms/sample - loss: 0.1606 - accuracy: 0.8656 - val_loss: 0.0634 - val_accuracy: 0.9418\n",
      "Epoch 2/50\n",
      "13284/13284 [==============================] - 8s 637us/sample - loss: 0.0834 - accuracy: 0.9452 - val_loss: 0.0418 - val_accuracy: 0.9690\n",
      "Epoch 3/50\n",
      "13284/13284 [==============================] - 9s 658us/sample - loss: 0.0741 - accuracy: 0.9556 - val_loss: 0.0551 - val_accuracy: 0.9623\n",
      "Epoch 4/50\n",
      "13284/13284 [==============================] - 8s 626us/sample - loss: 0.0681 - accuracy: 0.9611 - val_loss: 0.0369 - val_accuracy: 0.9804\n",
      "Epoch 5/50\n",
      "13284/13284 [==============================] - 9s 664us/sample - loss: 0.0646 - accuracy: 0.9663 - val_loss: 0.0274 - val_accuracy: 0.9840\n",
      "Epoch 6/50\n",
      "13284/13284 [==============================] - 8s 628us/sample - loss: 0.0605 - accuracy: 0.9694 - val_loss: 0.0277 - val_accuracy: 0.9759\n",
      "Epoch 7/50\n",
      "13284/13284 [==============================] - 9s 680us/sample - loss: 0.0603 - accuracy: 0.9725 - val_loss: 0.0204 - val_accuracy: 0.9921\n",
      "Epoch 8/50\n",
      "13284/13284 [==============================] - 8s 638us/sample - loss: 0.0581 - accuracy: 0.9761 - val_loss: 0.0253 - val_accuracy: 0.9913\n",
      "Epoch 9/50\n",
      "13284/13284 [==============================] - 11s 810us/sample - loss: 0.0563 - accuracy: 0.9764 - val_loss: 0.0329 - val_accuracy: 0.9873\n",
      "Epoch 10/50\n",
      "13284/13284 [==============================] - 10s 731us/sample - loss: 0.0557 - accuracy: 0.9778 - val_loss: 0.0244 - val_accuracy: 0.9884\n",
      "Epoch 11/50\n",
      "13284/13284 [==============================] - 10s 716us/sample - loss: 0.0549 - accuracy: 0.9773 - val_loss: 0.0185 - val_accuracy: 0.9892\n",
      "Epoch 12/50\n",
      "13284/13284 [==============================] - 13s 978us/sample - loss: 0.0540 - accuracy: 0.9760 - val_loss: 0.0340 - val_accuracy: 0.9830\n",
      "Epoch 13/50\n",
      "13284/13284 [==============================] - 12s 920us/sample - loss: 0.0527 - accuracy: 0.9807 - val_loss: 0.0215 - val_accuracy: 0.9899\n",
      "Epoch 14/50\n",
      "13284/13284 [==============================] - 14s 1ms/sample - loss: 0.0517 - accuracy: 0.9781 - val_loss: 0.0244 - val_accuracy: 0.9902\n",
      "Epoch 15/50\n",
      "13284/13284 [==============================] - 14s 1ms/sample - loss: 0.0511 - accuracy: 0.9794 - val_loss: 0.0172 - val_accuracy: 0.9933loss: 0.0512 - accuracy: \n",
      "Epoch 16/50\n",
      "13284/13284 [==============================] - 13s 1ms/sample - loss: 0.0509 - accuracy: 0.9798 - val_loss: 0.0157 - val_accuracy: 0.9960\n",
      "Epoch 17/50\n",
      "13284/13284 [==============================] - 11s 805us/sample - loss: 0.0510 - accuracy: 0.9815 - val_loss: 0.0174 - val_accuracy: 0.9890\n",
      "Epoch 18/50\n",
      "13284/13284 [==============================] - 13s 957us/sample - loss: 0.0505 - accuracy: 0.9798 - val_loss: 0.0189 - val_accuracy: 0.9895\n",
      "Epoch 19/50\n",
      "13284/13284 [==============================] - 12s 925us/sample - loss: 0.0498 - accuracy: 0.9782 - val_loss: 0.0229 - val_accuracy: 0.9919\n",
      "Epoch 20/50\n",
      "13284/13284 [==============================] - 10s 766us/sample - loss: 0.0498 - accuracy: 0.9788 - val_loss: 0.0114 - val_accuracy: 0.9940\n",
      "Epoch 21/50\n",
      "13284/13284 [==============================] - 10s 789us/sample - loss: 0.0487 - accuracy: 0.9810 - val_loss: 0.0254 - val_accuracy: 0.9879\n",
      "Epoch 22/50\n",
      "13284/13284 [==============================] - 14s 1ms/sample - loss: 0.0491 - accuracy: 0.9809 - val_loss: 0.0192 - val_accuracy: 0.9919\n",
      "Epoch 23/50\n",
      "13284/13284 [==============================] - 13s 1ms/sample - loss: 0.0487 - accuracy: 0.9804 - val_loss: 0.0171 - val_accuracy: 0.9930\n",
      "Epoch 24/50\n",
      "13284/13284 [==============================] - 13s 965us/sample - loss: 0.0486 - accuracy: 0.9798 - val_loss: 0.0334 - val_accuracy: 0.9815\n",
      "Epoch 25/50\n",
      "13284/13284 [==============================] - 10s 779us/sample - loss: 0.0483 - accuracy: 0.9811 - val_loss: 0.0141 - val_accuracy: 0.9957\n",
      "Epoch 26/50\n",
      "13284/13284 [==============================] - 13s 999us/sample - loss: 0.0477 - accuracy: 0.9826 - val_loss: 0.0136 - val_accuracy: 0.9960\n",
      "Epoch 27/50\n",
      "13284/13284 [==============================] - 11s 865us/sample - loss: 0.0479 - accuracy: 0.9792 - val_loss: 0.0161 - val_accuracy: 0.9969\n",
      "Epoch 28/50\n",
      "13284/13284 [==============================] - 10s 753us/sample - loss: 0.0469 - accuracy: 0.9813 - val_loss: 0.0243 - val_accuracy: 0.9899\n",
      "Epoch 29/50\n",
      "13284/13284 [==============================] - 13s 961us/sample - loss: 0.0475 - accuracy: 0.9827 - val_loss: 0.0261 - val_accuracy: 0.9893\n",
      "Epoch 30/50\n",
      "13284/13284 [==============================] - 13s 971us/sample - loss: 0.0474 - accuracy: 0.9798 - val_loss: 0.0149 - val_accuracy: 0.9924\n",
      "Epoch 31/50\n",
      "13284/13284 [==============================] - 13s 969us/sample - loss: 0.0471 - accuracy: 0.9811 - val_loss: 0.0120 - val_accuracy: 0.9950\n",
      "Epoch 32/50\n",
      "13284/13284 [==============================] - 15s 1ms/sample - loss: 0.0474 - accuracy: 0.9788 - val_loss: 0.0206 - val_accuracy: 0.9962\n",
      "Epoch 33/50\n",
      "13284/13284 [==============================] - 13s 971us/sample - loss: 0.0470 - accuracy: 0.9813 - val_loss: 0.0129 - val_accuracy: 0.9974\n",
      "Epoch 34/50\n",
      "13284/13284 [==============================] - 12s 913us/sample - loss: 0.0462 - accuracy: 0.9813 - val_loss: 0.0157 - val_accuracy: 0.9905\n",
      "Epoch 35/50\n",
      "13284/13284 [==============================] - 13s 999us/sample - loss: 0.0463 - accuracy: 0.9798 - val_loss: 0.0189 - val_accuracy: 0.9945\n",
      "Epoch 36/50\n",
      "13284/13284 [==============================] - 12s 903us/sample - loss: 0.0459 - accuracy: 0.9818 - val_loss: 0.0219 - val_accuracy: 0.9969\n",
      "Epoch 37/50\n",
      "13284/13284 [==============================] - 11s 841us/sample - loss: 0.0461 - accuracy: 0.9811 - val_loss: 0.0155 - val_accuracy: 0.9956\n",
      "Epoch 38/50\n",
      "13284/13284 [==============================] - 11s 805us/sample - loss: 0.0458 - accuracy: 0.9820 - val_loss: 0.0112 - val_accuracy: 0.9922\n",
      "Epoch 39/50\n",
      "13284/13284 [==============================] - 10s 788us/sample - loss: 0.0456 - accuracy: 0.9816 - val_loss: 0.0157 - val_accuracy: 0.9922\n",
      "Epoch 40/50\n",
      "13284/13284 [==============================] - 11s 796us/sample - loss: 0.0457 - accuracy: 0.9813 - val_loss: 0.0151 - val_accuracy: 0.9919\n",
      "Epoch 41/50\n",
      "13284/13284 [==============================] - 11s 805us/sample - loss: 0.0460 - accuracy: 0.9811 - val_loss: 0.0140 - val_accuracy: 0.9885\n",
      "Epoch 42/50\n",
      "13284/13284 [==============================] - 11s 831us/sample - loss: 0.0455 - accuracy: 0.9812 - val_loss: 0.0192 - val_accuracy: 0.9836\n",
      "Epoch 43/50\n",
      "13284/13284 [==============================] - 11s 797us/sample - loss: 0.0456 - accuracy: 0.9807 - val_loss: 0.0173 - val_accuracy: 0.9942\n",
      "Epoch 44/50\n",
      "13284/13284 [==============================] - 14s 1ms/sample - loss: 0.0448 - accuracy: 0.9817 - val_loss: 0.0248 - val_accuracy: 0.9931\n",
      "Epoch 45/50\n",
      "13284/13284 [==============================] - 10s 780us/sample - loss: 0.0452 - accuracy: 0.9821 - val_loss: 0.0137 - val_accuracy: 0.9945\n",
      "Epoch 46/50\n",
      "13284/13284 [==============================] - 10s 788us/sample - loss: 0.0449 - accuracy: 0.9825 - val_loss: 0.0157 - val_accuracy: 0.9960\n",
      "Epoch 47/50\n",
      "13284/13284 [==============================] - 10s 736us/sample - loss: 0.0453 - accuracy: 0.9808 - val_loss: 0.0181 - val_accuracy: 0.9879\n",
      "Epoch 48/50\n",
      "13284/13284 [==============================] - 10s 771us/sample - loss: 0.0443 - accuracy: 0.9822 - val_loss: 0.0196 - val_accuracy: 0.9965\n",
      "Epoch 49/50\n",
      "13284/13284 [==============================] - 12s 880us/sample - loss: 0.0449 - accuracy: 0.9799 - val_loss: 0.0130 - val_accuracy: 0.9956\n",
      "Epoch 50/50\n",
      "13284/13284 [==============================] - 10s 750us/sample - loss: 0.0446 - accuracy: 0.9825 - val_loss: 0.0238 - val_accuracy: 0.9899\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(x_train,y_train,epochs=50,validation_data=(x_test,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('C:/Users/User-0005/Desktop/ML/GPS_pr/model/gps/model.h5')\n",
    "# model.save(args[\"model\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.python.keras.models import load_model\n",
    "model1 = load_model('model/gps/model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.2449942 , -0.6566762 ],\n",
       "       [-0.28479466, -0.1603694 ],\n",
       "       [-0.8354998 ,  0.3866222 ],\n",
       "       ...,\n",
       "       [ 0.3392489 ,  0.42517418],\n",
       "       [ 0.38600916, -0.79164934],\n",
       "       [ 0.4464331 ,  0.890225  ]], dtype=float32)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = model1.predict(x_test)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  41.252075, -105.42498 ],\n",
       "       [  41.218777, -105.38482 ],\n",
       "       [  41.184166, -105.34056 ],\n",
       "       ...,\n",
       "       [  41.258003, -105.33744 ],\n",
       "       [  41.26094 , -105.435905],\n",
       "       [  41.26474 , -105.29982 ]], dtype=float32)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "res_df = sc.inverse_transform(result)\n",
    "# res_df.reshape(new_df.shape[0],5,1)\n",
    "res_df\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  41.25146  -105.426549]\n",
      " [  41.216984 -105.38606 ]\n",
      " [  41.180747 -105.338721]\n",
      " ...\n",
      " [  41.256388 -105.334362]\n",
      " [  41.261073 -105.43826 ]\n",
      " [  41.261798 -105.295944]]\n"
     ]
    }
   ],
   "source": [
    "y_test_actual=sc.inverse_transform(y_test)\n",
    "print(y_test_actual)\n",
    "# for i in range(len(result)):\n",
    "#     plt.scatter(result[0][i],result[1][i],c='r')\n",
    "#     plt.scatter(y_test[0][i],y_test[1][i],c='g')\n",
    "\n",
    "# plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting actual points with GPS points\n",
    "\n",
    "predicted_df = pd.DataFrame(data=res_df[0:,0:],index=[i for i in range(res_df.shape[0])],columns=['f'+str(i) for i in range(res_df.shape[1])])\n",
    "\n",
    "actual_df = pd.DataFrame(data=y_test_actual[0:,0:],index=[i for i in range(y_test_actual.shape[0])],columns=['f'+str(i) for i in range(y_test_actual.shape[1])])\n",
    "\n",
    "# for i in range(len(res_df)):\n",
    "# #   print(i)\n",
    "plt.scatter(x=predicted_df['f0'],y=predicted_df['f1'],c='r',s=0.1,alpha=0.5)\n",
    "plt.scatter(x=actual_df['f0'],y=actual_df['f1'],c='g',s=0.1,alpha=0.5)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import pyproj\n",
    "\n",
    "# # Example position data, should be somewhere in Germany\n",
    "# x = 784413.4586438217\n",
    "# y = 5735894.791428658\n",
    "# z = -2667746.972596679\n",
    "\n",
    "\n",
    "# # ecef = pyproj.Proj(proj='geocent', ellps='WGS84', datum='WGS84')\n",
    "# # lla = pyproj.Proj(proj='latlong', ellps='WGS84', datum='WGS84')\n",
    "# # lon, lat, alt = pyproj.transform(ecef, lla, x, y, z, radians=True)\n",
    "\n",
    "# # print(lat, lon, alt)\n",
    "\n",
    "# lon, lat, alt = pyproj.transform(ecef, lla, x, y, z, radians=False)\n",
    "\n",
    "# print(lat, lon, alt)\n",
    "\n",
    "# lon, lat, alt = -24.887220848803036, 82.2128095673836, 0\n",
    "\n",
    "# x,y,z = pyproj.transform(lla,ecef,lat,lon,alt)\n",
    "# print(x,y,z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import pyproj\n",
    "\n",
    "# def convert_to_ecef(lat,lon,alt=0):\n",
    "#     return pyproj.transform(lla,ecef,lat,lon,alt)\n",
    "\n",
    "# def convert_to_lla(x,y,z):\n",
    "#     return pyproj.transform(ecef, lla, x, y, z, radians=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x25a70f52908>"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "data = np.random.random(size=(16, 16))\n",
    "\n",
    "plt.imshow(data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image, ImageDraw\n",
    "img = Image.new('RGB', (20, 20), color = (255, 255, 255))\n",
    "d = ImageDraw.Draw(img)\n",
    "d.text((5,5), \"O\", fill=(0,0,0))\n",
    "img.save('pil_text.png')\n",
    "\n",
    "# plt.imshow(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(20, 20, 3)\n",
      "(20, 20)\n",
      "[[1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 0. 0. 1. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 0. 0. 1. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 0. 0. 1. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 0. 0. 1. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]\n",
      " [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]]\n",
      "(40, 40)\n"
     ]
    }
   ],
   "source": [
    "x=plt.imread('pil_text.png', format=None)\n",
    "# plt.imshow(x)\n",
    "\n",
    "# print(x)\n",
    "# x.reshape(x.shape[0],x.shape[0],1)\n",
    "print(x.shape)\n",
    "lum_img = x[:,:,0]\n",
    "# plt.imshow(x)\n",
    "print(lum_img.shape)\n",
    "# plt.imshow(lum_img)\n",
    "# lum_img = lum_img.reshape(20,20)\n",
    "print(lum_img)\n",
    "# plt.imshow(lum_img,cmap='gray')\n",
    "\n",
    "# xx = np.append(xx,values=[lum_img,lum_img],axis=1)\n",
    "xx= np.append(lum_img,lum_img,axis=0)\n",
    "yy= np.append(lum_img,lum_img,axis=0)\n",
    "\n",
    "zz = np.append(xx,yy,axis=1)\n",
    "print(zz.shape)\n",
    "\n",
    "# np.concatenate((arr,[[lumimg,lum_img , lum_img]] ) , axis=0)\n",
    "# plt.imshow(zz)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x25a71b5a898>"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y=plt.imread('pil_text1.png', format=None)\n",
    "plt.imshow(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x25a78cc8518>"
      ]
     },
     "execution_count": 140,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "z= np.multiply(x,y)\n",
    "plt.imshow(z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}