Series.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import keras as k\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import preprocessing\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dropout,Input, Dense,merge, Add\n",
    "from keras import optimizers\n",
    "from keras.models import Model\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_ = np.loadtxt(\"./series/random.txt\")\n",
    "green_ = np.loadtxt(\"./series/green.txt\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = pd.DataFrame(columns=[\"amu\",\"beta\"])\n",
    "X[\"amu\"] = random_[:,0]\n",
    "X[\"beta\"] = random_[:,1]\n",
    "Y = pd.DataFrame(columns=[\"G\",\"damu\",\"dbeta\"])\n",
    "Y[\"G\"] = green_[:,0]\n",
    "Y[\"damu\"] = green_[:,1]\n",
    "Y[\"dbeta\"] = green_[:,2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "G       -0.001727\n",
       "damu     0.013334\n",
       "dbeta    0.010205\n",
       "dtype: float64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y.median(axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.30, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# scaler = Pre()\n",
    "random_scale = preprocessing.MinMaxScaler([0,5])\n",
    "green_scale = preprocessing.MinMaxScaler([0,5])\n",
    "\n",
    "random_scale.fit(X_train)\n",
    "green_scale.fit(y_train)\n",
    "X_train = random_scale.transform(X_train)\n",
    "X_test = random_scale.transform(X_test)\n",
    "\n",
    "y_train = green_scale.transform(y_train)\n",
    "y_test = green_scale.transform(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y1,Y2,Y3 =y_train[:,0],y_train[:,1],y_train[:,2]\n",
    "Y1_,Y2_,Y3_ =y_test[:,0],y_test[:,1],y_test[:,2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y1,Y2,Y3 = Y1.reshape(Y1.size,1),Y2.reshape(Y2.size,1),Y3.reshape(Y3.size,1)\n",
    "Y1_,Y2_,Y3_ = Y1_.reshape(Y1_.size,1),Y2_.reshape(Y2_.size,1),Y3_.reshape(Y3_.size,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = Input(shape=(2,))\n",
    "x = Dense(8, activation='tanh')(inputs)\n",
    "x = Dense(64, activation='tanh')(x)\n",
    "x = Dense(64, activation='tanh')(x)\n",
    "x = Dense(64, activation='tanh')(x)\n",
    "x_1 = Dense(64, activation='tanh')(x)\n",
    "x = Dense(64, activation='tanh')(x_1)\n",
    "# x = Dense(64, activation='relu')(x)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "x1_ = Dense(64, activation='tanh')(x)\n",
    "x1 = Dense(64, activation='tanh')(x1_)\n",
    "x1 = Dense(64, activation='tanh')(x1)\n",
    "x1 = Dense(64, activation='tanh')(x1)\n",
    "x1 = Dense(64, activation='tanh')(x1)\n",
    "# x = Dense(64, activation='relu')(x)\n",
    "x1 = Add()([x1, x1_])\n",
    "x1 = Dense(64, activation='tanh')(x)\n",
    "y1 = Dense(1, activation=None)(x1)\n",
    "\n",
    "x2_ = Dense(64, activation='tanh')(x)\n",
    "x2 = Dense(64, activation='tanh')(x2_)\n",
    "x2 = Dense(64, activation='tanh')(x2)\n",
    "x2 = Dense(64, activation='tanh')(x2)\n",
    "x2 = Dense(64, activation='tanh')(x2)\n",
    "x2 = Dense(64, activation='tanh')(x2)\n",
    "# x = Dense(64, activation='relu')(x)\n",
    "x2 = Add()([x2, x2_])\n",
    "y2 = Dense(1, activation=None)(x2)\n",
    "\n",
    "x3 = Dense(32, activation='tanh')(x)\n",
    "x3 = Dense(10, activation='tanh')(x3)\n",
    "y3 = Dense(1, activation=None)(x3)\n",
    "\n",
    "model = Model(inputs=inputs, outputs=[y1,y2,y3])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "sgd = optimizers.adam(lr=0.0001)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "def time_decay(epoch, initial_lrate):\n",
    "    decay_rate = 0.001\n",
    "    new_lrate = initial_lrate/(1+decay_rate*epoch)\n",
    "    return new_lrate\n",
    "from keras.callbacks import LearningRateScheduler\n",
    "lrate = LearningRateScheduler(time_decay,verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss=['mean_squared_error','mean_squared_error','mean_squared_error'], optimizer=sgd, metrics=['mean_absolute_percentage_error'],)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 49749 samples, validate on 21322 samples\n",
      "Epoch 1/40\n",
      "\n",
      "Epoch 00001: LearningRateScheduler setting learning rate to 9.999999747378752e-05.\n",
      "49749/49749 [==============================] - 15s 294us/step - loss: 0.3575 - dense_36_loss: 0.2070 - dense_43_loss: 0.0804 - dense_46_loss: 0.0701 - dense_36_mean_absolute_percentage_error: 48121.4052 - dense_43_mean_absolute_percentage_error: 16557.0958 - dense_46_mean_absolute_percentage_error: 46586.9814 - val_loss: 0.1624 - val_dense_36_loss: 0.0902 - val_dense_43_loss: 0.0409 - val_dense_46_loss: 0.0313 - val_dense_36_mean_absolute_percentage_error: 13.4531 - val_dense_43_mean_absolute_percentage_error: 13.1375 - val_dense_46_mean_absolute_percentage_error: 12.0512\n",
      "Epoch 2/40\n",
      "\n",
      "Epoch 00002: LearningRateScheduler setting learning rate to 9.990009737641112e-05.\n",
      "49749/49749 [==============================] - 11s 222us/step - loss: 0.1596 - dense_36_loss: 0.0892 - dense_43_loss: 0.0394 - dense_46_loss: 0.0310 - dense_36_mean_absolute_percentage_error: 51247.4256 - dense_43_mean_absolute_percentage_error: 47892.1068 - dense_46_mean_absolute_percentage_error: 46206.3997 - val_loss: 0.1606 - val_dense_36_loss: 0.0882 - val_dense_43_loss: 0.0412 - val_dense_46_loss: 0.0311 - val_dense_36_mean_absolute_percentage_error: 13.1851 - val_dense_43_mean_absolute_percentage_error: 13.4895 - val_dense_46_mean_absolute_percentage_error: 11.8020\n",
      "Epoch 3/40\n",
      "\n",
      "Epoch 00003: LearningRateScheduler setting learning rate to 9.970069718138201e-05.\n",
      "49749/49749 [==============================] - 11s 222us/step - loss: 0.1580 - dense_36_loss: 0.0876 - dense_43_loss: 0.0394 - dense_46_loss: 0.0310 - dense_36_mean_absolute_percentage_error: 52097.0027 - dense_43_mean_absolute_percentage_error: 46937.8363 - dense_46_mean_absolute_percentage_error: 45660.5139 - val_loss: 0.1595 - val_dense_36_loss: 0.0860 - val_dense_43_loss: 0.0417 - val_dense_46_loss: 0.0318 - val_dense_36_mean_absolute_percentage_error: 13.0848 - val_dense_43_mean_absolute_percentage_error: 13.7603 - val_dense_46_mean_absolute_percentage_error: 12.3983\n",
      "Epoch 4/40\n",
      "\n",
      "Epoch 00004: LearningRateScheduler setting learning rate to 9.940248622270237e-05.\n",
      "49749/49749 [==============================] - 11s 223us/step - loss: 0.1572 - dense_36_loss: 0.0867 - dense_43_loss: 0.0395 - dense_46_loss: 0.0309 - dense_36_mean_absolute_percentage_error: 50464.0015 - dense_43_mean_absolute_percentage_error: 46668.9805 - dense_46_mean_absolute_percentage_error: 46735.2963 - val_loss: 0.1599 - val_dense_36_loss: 0.0867 - val_dense_43_loss: 0.0405 - val_dense_46_loss: 0.0328 - val_dense_36_mean_absolute_percentage_error: 13.0201 - val_dense_43_mean_absolute_percentage_error: 13.1580 - val_dense_46_mean_absolute_percentage_error: 12.6142\n",
      "Epoch 5/40\n",
      "\n",
      "Epoch 00005: LearningRateScheduler setting learning rate to 9.900646270772227e-05.\n",
      "49749/49749 [==============================] - 11s 225us/step - loss: 0.1563 - dense_36_loss: 0.0863 - dense_43_loss: 0.0391 - dense_46_loss: 0.0309 - dense_36_mean_absolute_percentage_error: 50912.8423 - dense_43_mean_absolute_percentage_error: 46113.6805 - dense_46_mean_absolute_percentage_error: 45966.1937 - val_loss: 0.1629 - val_dense_36_loss: 0.0872 - val_dense_43_loss: 0.0434 - val_dense_46_loss: 0.0323 - val_dense_36_mean_absolute_percentage_error: 13.2332 - val_dense_43_mean_absolute_percentage_error: 13.7569 - val_dense_46_mean_absolute_percentage_error: 12.7242\n",
      "Epoch 6/40\n",
      "\n",
      "Epoch 00006: LearningRateScheduler setting learning rate to 9.851389598166246e-05.\n",
      "49749/49749 [==============================] - 11s 228us/step - loss: 0.1557 - dense_36_loss: 0.0859 - dense_43_loss: 0.0392 - dense_46_loss: 0.0307 - dense_36_mean_absolute_percentage_error: 50599.3479 - dense_43_mean_absolute_percentage_error: 48023.1826 - dense_46_mean_absolute_percentage_error: 45377.2575 - val_loss: 0.1561 - val_dense_36_loss: 0.0845 - val_dense_43_loss: 0.0405 - val_dense_46_loss: 0.0311 - val_dense_36_mean_absolute_percentage_error: 12.5453 - val_dense_43_mean_absolute_percentage_error: 13.1017 - val_dense_46_mean_absolute_percentage_error: 11.9500\n",
      "Epoch 7/40\n",
      "\n",
      "Epoch 00007: LearningRateScheduler setting learning rate to 9.792633964513497e-05.\n",
      "49749/49749 [==============================] - 12s 235us/step - loss: 0.1546 - dense_36_loss: 0.0851 - dense_43_loss: 0.0389 - dense_46_loss: 0.0306 - dense_36_mean_absolute_percentage_error: 54536.6205 - dense_43_mean_absolute_percentage_error: 46916.6347 - dense_46_mean_absolute_percentage_error: 46022.5169 - val_loss: 0.1581 - val_dense_36_loss: 0.0835 - val_dense_43_loss: 0.0437 - val_dense_46_loss: 0.0309 - val_dense_36_mean_absolute_percentage_error: 12.7257 - val_dense_43_mean_absolute_percentage_error: 14.0519 - val_dense_46_mean_absolute_percentage_error: 11.5609\n",
      "Epoch 8/40\n",
      "\n",
      "Epoch 00008: LearningRateScheduler setting learning rate to 9.724562291736607e-05.\n",
      "49749/49749 [==============================] - 12s 239us/step - loss: 0.1540 - dense_36_loss: 0.0846 - dense_43_loss: 0.0388 - dense_46_loss: 0.0306 - dense_36_mean_absolute_percentage_error: 49918.9780 - dense_43_mean_absolute_percentage_error: 46796.7334 - dense_46_mean_absolute_percentage_error: 45367.4828 - val_loss: 0.1551 - val_dense_36_loss: 0.0833 - val_dense_43_loss: 0.0408 - val_dense_46_loss: 0.0310 - val_dense_36_mean_absolute_percentage_error: 12.7698 - val_dense_43_mean_absolute_percentage_error: 13.4029 - val_dense_46_mean_absolute_percentage_error: 11.6468\n",
      "Epoch 9/40\n",
      "\n",
      "Epoch 00009: LearningRateScheduler setting learning rate to 9.647383483261654e-05.\n",
      "49749/49749 [==============================] - 12s 246us/step - loss: 0.1538 - dense_36_loss: 0.0844 - dense_43_loss: 0.0388 - dense_46_loss: 0.0306 - dense_36_mean_absolute_percentage_error: 52085.6856 - dense_43_mean_absolute_percentage_error: 47203.1921 - dense_46_mean_absolute_percentage_error: 45703.6660 - val_loss: 0.1570 - val_dense_36_loss: 0.0857 - val_dense_43_loss: 0.0403 - val_dense_46_loss: 0.0311 - val_dense_36_mean_absolute_percentage_error: 13.1450 - val_dense_43_mean_absolute_percentage_error: 13.0591 - val_dense_46_mean_absolute_percentage_error: 11.5793\n",
      "Epoch 10/40\n",
      "\n",
      "Epoch 00010: LearningRateScheduler setting learning rate to 9.561331574163586e-05.\n",
      "49749/49749 [==============================] - 12s 250us/step - loss: 0.1526 - dense_36_loss: 0.0836 - dense_43_loss: 0.0385 - dense_46_loss: 0.0306 - dense_36_mean_absolute_percentage_error: 51922.1186 - dense_43_mean_absolute_percentage_error: 47985.4996 - dense_46_mean_absolute_percentage_error: 45798.9498 - val_loss: 0.1535 - val_dense_36_loss: 0.0822 - val_dense_43_loss: 0.0402 - val_dense_46_loss: 0.0310 - val_dense_36_mean_absolute_percentage_error: 12.6681 - val_dense_43_mean_absolute_percentage_error: 13.0511 - val_dense_46_mean_absolute_percentage_error: 11.6415\n",
      "Epoch 11/40\n",
      "\n",
      "Epoch 00011: LearningRateScheduler setting learning rate to 9.466664886360269e-05.\n",
      "49749/49749 [==============================] - 13s 252us/step - loss: 0.1511 - dense_36_loss: 0.0822 - dense_43_loss: 0.0385 - dense_46_loss: 0.0305 - dense_36_mean_absolute_percentage_error: 50790.0365 - dense_43_mean_absolute_percentage_error: 47595.3384 - dense_46_mean_absolute_percentage_error: 45402.9131 - val_loss: 0.1521 - val_dense_36_loss: 0.0800 - val_dense_43_loss: 0.0413 - val_dense_46_loss: 0.0308 - val_dense_36_mean_absolute_percentage_error: 12.0405 - val_dense_43_mean_absolute_percentage_error: 13.6856 - val_dense_46_mean_absolute_percentage_error: 11.6763\n",
      "Epoch 12/40\n",
      "\n",
      "Epoch 00012: LearningRateScheduler setting learning rate to 9.363664469140942e-05.\n",
      "49749/49749 [==============================] - 13s 257us/step - loss: 0.1497 - dense_36_loss: 0.0809 - dense_43_loss: 0.0383 - dense_46_loss: 0.0305 - dense_36_mean_absolute_percentage_error: 50612.2764 - dense_43_mean_absolute_percentage_error: 48084.9474 - dense_46_mean_absolute_percentage_error: 46031.9031 - val_loss: 0.1505 - val_dense_36_loss: 0.0796 - val_dense_43_loss: 0.0400 - val_dense_46_loss: 0.0310 - val_dense_36_mean_absolute_percentage_error: 12.3538 - val_dense_43_mean_absolute_percentage_error: 12.9487 - val_dense_46_mean_absolute_percentage_error: 11.5153\n",
      "Epoch 13/40\n",
      "\n",
      "Epoch 00013: LearningRateScheduler setting learning rate to 9.252632548940541e-05.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "49749/49749 [==============================] - 12s 246us/step - loss: 0.1467 - dense_36_loss: 0.0782 - dense_43_loss: 0.0380 - dense_46_loss: 0.0305 - dense_36_mean_absolute_percentage_error: 49431.0735 - dense_43_mean_absolute_percentage_error: 46961.5791 - dense_46_mean_absolute_percentage_error: 45821.2071 - val_loss: 0.1470 - val_dense_36_loss: 0.0760 - val_dense_43_loss: 0.0402 - val_dense_46_loss: 0.0308 - val_dense_36_mean_absolute_percentage_error: 11.7501 - val_dense_43_mean_absolute_percentage_error: 12.7899 - val_dense_46_mean_absolute_percentage_error: 11.5909\n",
      "Epoch 14/40\n",
      "\n",
      "Epoch 00014: LearningRateScheduler setting learning rate to 9.13389170655442e-05.\n",
      "49749/49749 [==============================] - 12s 248us/step - loss: 0.1433 - dense_36_loss: 0.0751 - dense_43_loss: 0.0377 - dense_46_loss: 0.0304 - dense_36_mean_absolute_percentage_error: 51184.5140 - dense_43_mean_absolute_percentage_error: 47445.8947 - dense_46_mean_absolute_percentage_error: 46432.5868 - val_loss: 0.1438 - val_dense_36_loss: 0.0731 - val_dense_43_loss: 0.0397 - val_dense_46_loss: 0.0310 - val_dense_36_mean_absolute_percentage_error: 11.6312 - val_dense_43_mean_absolute_percentage_error: 13.1132 - val_dense_46_mean_absolute_percentage_error: 11.4390\n",
      "Epoch 15/40\n",
      "\n",
      "Epoch 00015: LearningRateScheduler setting learning rate to 9.007782624121498e-05.\n",
      "49749/49749 [==============================] - 13s 253us/step - loss: 0.1382 - dense_36_loss: 0.0707 - dense_43_loss: 0.0371 - dense_46_loss: 0.0304 - dense_36_mean_absolute_percentage_error: 52427.9906 - dense_43_mean_absolute_percentage_error: 47563.3187 - dense_46_mean_absolute_percentage_error: 45670.8842 - val_loss: 0.1363 - val_dense_36_loss: 0.0671 - val_dense_43_loss: 0.0384 - val_dense_46_loss: 0.0308 - val_dense_36_mean_absolute_percentage_error: 10.9861 - val_dense_43_mean_absolute_percentage_error: 12.6641 - val_dense_46_mean_absolute_percentage_error: 11.5124\n",
      "Epoch 16/40\n",
      "\n",
      "Epoch 00016: LearningRateScheduler setting learning rate to 8.874662562268856e-05.\n",
      "49749/49749 [==============================] - 12s 250us/step - loss: 0.1303 - dense_36_loss: 0.0647 - dense_43_loss: 0.0353 - dense_46_loss: 0.0303 - dense_36_mean_absolute_percentage_error: 51044.1125 - dense_43_mean_absolute_percentage_error: 46858.0439 - dense_46_mean_absolute_percentage_error: 45266.2544 - val_loss: 0.1302 - val_dense_36_loss: 0.0629 - val_dense_43_loss: 0.0366 - val_dense_46_loss: 0.0307 - val_dense_36_mean_absolute_percentage_error: 10.0669 - val_dense_43_mean_absolute_percentage_error: 12.6973 - val_dense_46_mean_absolute_percentage_error: 11.8943\n",
      "Epoch 17/40\n",
      "\n",
      "Epoch 00017: LearningRateScheduler setting learning rate to 8.734903846249557e-05.\n",
      "49749/49749 [==============================] - 13s 253us/step - loss: 0.1237 - dense_36_loss: 0.0604 - dense_43_loss: 0.0333 - dense_46_loss: 0.0300 - dense_36_mean_absolute_percentage_error: 50781.2678 - dense_43_mean_absolute_percentage_error: 47082.3492 - dense_46_mean_absolute_percentage_error: 46650.8452 - val_loss: 0.1240 - val_dense_36_loss: 0.0580 - val_dense_43_loss: 0.0355 - val_dense_46_loss: 0.0305 - val_dense_36_mean_absolute_percentage_error: 10.1194 - val_dense_43_mean_absolute_percentage_error: 11.3700 - val_dense_46_mean_absolute_percentage_error: 11.9400\n",
      "Epoch 18/40\n",
      "\n",
      "Epoch 00018: LearningRateScheduler setting learning rate to 8.588892361011829e-05.\n",
      "49749/49749 [==============================] - 13s 253us/step - loss: 0.1195 - dense_36_loss: 0.0579 - dense_43_loss: 0.0318 - dense_46_loss: 0.0297 - dense_36_mean_absolute_percentage_error: 48455.1894 - dense_43_mean_absolute_percentage_error: 47176.2210 - dense_46_mean_absolute_percentage_error: 46357.4479 - val_loss: 0.1210 - val_dense_36_loss: 0.0577 - val_dense_43_loss: 0.0331 - val_dense_46_loss: 0.0302 - val_dense_36_mean_absolute_percentage_error: 10.3281 - val_dense_43_mean_absolute_percentage_error: 12.4553 - val_dense_46_mean_absolute_percentage_error: 11.6250\n",
      "Epoch 19/40\n",
      "\n",
      "Epoch 00019: LearningRateScheduler setting learning rate to 8.437026055138156e-05.\n",
      "49749/49749 [==============================] - 13s 257us/step - loss: 0.1153 - dense_36_loss: 0.0559 - dense_43_loss: 0.0300 - dense_46_loss: 0.0294 - dense_36_mean_absolute_percentage_error: 46494.4875 - dense_43_mean_absolute_percentage_error: 46232.8753 - dense_46_mean_absolute_percentage_error: 46796.0677 - val_loss: 0.1178 - val_dense_36_loss: 0.0556 - val_dense_43_loss: 0.0325 - val_dense_46_loss: 0.0297 - val_dense_36_mean_absolute_percentage_error: 9.6747 - val_dense_43_mean_absolute_percentage_error: 11.6361 - val_dense_46_mean_absolute_percentage_error: 11.4426\n",
      "Epoch 20/40\n",
      "\n",
      "Epoch 00020: LearningRateScheduler setting learning rate to 8.279711311505752e-05.\n",
      "49749/49749 [==============================] - 13s 258us/step - loss: 0.1122 - dense_36_loss: 0.0542 - dense_43_loss: 0.0287 - dense_46_loss: 0.0293 - dense_36_mean_absolute_percentage_error: 49730.4910 - dense_43_mean_absolute_percentage_error: 46526.8707 - dense_46_mean_absolute_percentage_error: 47838.5776 - val_loss: 0.1112 - val_dense_36_loss: 0.0525 - val_dense_43_loss: 0.0294 - val_dense_46_loss: 0.0293 - val_dense_36_mean_absolute_percentage_error: 9.2901 - val_dense_43_mean_absolute_percentage_error: 11.1950 - val_dense_46_mean_absolute_percentage_error: 11.4039\n",
      "Epoch 21/40\n",
      "\n",
      "Epoch 00021: LearningRateScheduler setting learning rate to 8.117364332600333e-05.\n",
      "49749/49749 [==============================] - 13s 259us/step - loss: 0.1094 - dense_36_loss: 0.0525 - dense_43_loss: 0.0278 - dense_46_loss: 0.0291 - dense_36_mean_absolute_percentage_error: 49270.5068 - dense_43_mean_absolute_percentage_error: 45213.2414 - dense_46_mean_absolute_percentage_error: 46029.3425 - val_loss: 0.1099 - val_dense_36_loss: 0.0513 - val_dense_43_loss: 0.0293 - val_dense_46_loss: 0.0293 - val_dense_36_mean_absolute_percentage_error: 9.2251 - val_dense_43_mean_absolute_percentage_error: 11.5428 - val_dense_46_mean_absolute_percentage_error: 11.8655\n",
      "Epoch 22/40\n",
      "\n",
      "Epoch 00022: LearningRateScheduler setting learning rate to 7.950406104014287e-05.\n",
      "49749/49749 [==============================] - 13s 259us/step - loss: 0.1065 - dense_36_loss: 0.0507 - dense_43_loss: 0.0269 - dense_46_loss: 0.0290 - dense_36_mean_absolute_percentage_error: 48712.5583 - dense_43_mean_absolute_percentage_error: 43629.8312 - dense_46_mean_absolute_percentage_error: 47933.3544 - val_loss: 0.1065 - val_dense_36_loss: 0.0496 - val_dense_43_loss: 0.0278 - val_dense_46_loss: 0.0291 - val_dense_36_mean_absolute_percentage_error: 9.4185 - val_dense_43_mean_absolute_percentage_error: 11.1917 - val_dense_46_mean_absolute_percentage_error: 11.4678\n",
      "Epoch 23/40\n",
      "\n",
      "Epoch 00023: LearningRateScheduler setting learning rate to 7.779262371045954e-05.\n",
      "49749/49749 [==============================] - 13s 260us/step - loss: 0.1037 - dense_36_loss: 0.0487 - dense_43_loss: 0.0263 - dense_46_loss: 0.0287 - dense_36_mean_absolute_percentage_error: 48734.4879 - dense_43_mean_absolute_percentage_error: 45979.4809 - dense_46_mean_absolute_percentage_error: 47269.6601 - val_loss: 0.1058 - val_dense_36_loss: 0.0485 - val_dense_43_loss: 0.0283 - val_dense_46_loss: 0.0290 - val_dense_36_mean_absolute_percentage_error: 9.3856 - val_dense_43_mean_absolute_percentage_error: 11.7787 - val_dense_46_mean_absolute_percentage_error: 11.3714\n",
      "Epoch 24/40\n",
      "\n",
      "Epoch 00024: LearningRateScheduler setting learning rate to 7.60436219296155e-05.\n",
      "49749/49749 [==============================] - 13s 262us/step - loss: 0.1003 - dense_36_loss: 0.0469 - dense_43_loss: 0.0249 - dense_46_loss: 0.0284 - dense_36_mean_absolute_percentage_error: 50449.0290 - dense_43_mean_absolute_percentage_error: 45292.9250 - dense_46_mean_absolute_percentage_error: 47723.3414 - val_loss: 0.1011 - val_dense_36_loss: 0.0459 - val_dense_43_loss: 0.0266 - val_dense_46_loss: 0.0285 - val_dense_36_mean_absolute_percentage_error: 8.7664 - val_dense_43_mean_absolute_percentage_error: 11.0964 - val_dense_46_mean_absolute_percentage_error: 11.5295\n",
      "Epoch 25/40\n",
      "\n",
      "Epoch 00025: LearningRateScheduler setting learning rate to 7.426135084642738e-05.\n",
      "49749/49749 [==============================] - 13s 260us/step - loss: 0.0968 - dense_36_loss: 0.0451 - dense_43_loss: 0.0235 - dense_46_loss: 0.0281 - dense_36_mean_absolute_percentage_error: 49821.2566 - dense_43_mean_absolute_percentage_error: 50507.6198 - dense_46_mean_absolute_percentage_error: 46126.4773 - val_loss: 0.0994 - val_dense_36_loss: 0.0443 - val_dense_43_loss: 0.0266 - val_dense_46_loss: 0.0285 - val_dense_36_mean_absolute_percentage_error: 9.1769 - val_dense_43_mean_absolute_percentage_error: 10.4584 - val_dense_46_mean_absolute_percentage_error: 11.3047\n",
      "Epoch 26/40\n",
      "\n",
      "Epoch 00026: LearningRateScheduler setting learning rate to 7.245009594665069e-05.\n",
      "49749/49749 [==============================] - 13s 266us/step - loss: 0.0930 - dense_36_loss: 0.0427 - dense_43_loss: 0.0223 - dense_46_loss: 0.0279 - dense_36_mean_absolute_percentage_error: 50946.5172 - dense_43_mean_absolute_percentage_error: 44359.6363 - dense_46_mean_absolute_percentage_error: 48643.2788 - val_loss: 0.0935 - val_dense_36_loss: 0.0414 - val_dense_43_loss: 0.0240 - val_dense_46_loss: 0.0282 - val_dense_36_mean_absolute_percentage_error: 8.7231 - val_dense_43_mean_absolute_percentage_error: 10.8793 - val_dense_46_mean_absolute_percentage_error: 11.1075\n",
      "Epoch 27/40\n",
      "\n",
      "Epoch 00027: LearningRateScheduler setting learning rate to 7.061412600849412e-05.\n",
      "49749/49749 [==============================] - 13s 263us/step - loss: 0.0890 - dense_36_loss: 0.0397 - dense_43_loss: 0.0219 - dense_46_loss: 0.0275 - dense_36_mean_absolute_percentage_error: 45863.6770 - dense_43_mean_absolute_percentage_error: 45853.7877 - dense_46_mean_absolute_percentage_error: 45187.8718 - val_loss: 0.0908 - val_dense_36_loss: 0.0399 - val_dense_43_loss: 0.0236 - val_dense_46_loss: 0.0273 - val_dense_36_mean_absolute_percentage_error: 9.1720 - val_dense_43_mean_absolute_percentage_error: 11.6363 - val_dense_46_mean_absolute_percentage_error: 11.6873\n",
      "Epoch 28/40\n",
      "\n",
      "Epoch 00028: LearningRateScheduler setting learning rate to 6.875767192994667e-05.\n",
      "49749/49749 [==============================] - 13s 264us/step - loss: 0.0846 - dense_36_loss: 0.0371 - dense_43_loss: 0.0210 - dense_46_loss: 0.0265 - dense_36_mean_absolute_percentage_error: 46961.1233 - dense_43_mean_absolute_percentage_error: 46962.4089 - dense_46_mean_absolute_percentage_error: 49474.9049 - val_loss: 0.0830 - val_dense_36_loss: 0.0355 - val_dense_43_loss: 0.0217 - val_dense_46_loss: 0.0258 - val_dense_36_mean_absolute_percentage_error: 7.7003 - val_dense_43_mean_absolute_percentage_error: 11.5808 - val_dense_46_mean_absolute_percentage_error: 11.8220\n",
      "Epoch 29/40\n",
      "\n",
      "Epoch 00029: LearningRateScheduler setting learning rate to 6.688489152412093e-05.\n",
      "49749/49749 [==============================] - 13s 264us/step - loss: 0.0799 - dense_36_loss: 0.0351 - dense_43_loss: 0.0195 - dense_46_loss: 0.0253 - dense_36_mean_absolute_percentage_error: 43285.8491 - dense_43_mean_absolute_percentage_error: 44774.1328 - dense_46_mean_absolute_percentage_error: 48813.7314 - val_loss: 0.0788 - val_dense_36_loss: 0.0339 - val_dense_43_loss: 0.0199 - val_dense_46_loss: 0.0250 - val_dense_36_mean_absolute_percentage_error: 7.2934 - val_dense_43_mean_absolute_percentage_error: 11.4305 - val_dense_46_mean_absolute_percentage_error: 12.4671\n",
      "Epoch 30/40\n",
      "\n",
      "Epoch 00030: LearningRateScheduler setting learning rate to 6.499989108708339e-05.\n",
      "49749/49749 [==============================] - 13s 269us/step - loss: 0.0762 - dense_36_loss: 0.0338 - dense_43_loss: 0.0180 - dense_46_loss: 0.0244 - dense_36_mean_absolute_percentage_error: 41588.6888 - dense_43_mean_absolute_percentage_error: 52691.4964 - dense_46_mean_absolute_percentage_error: 48889.4410 - val_loss: 0.0766 - val_dense_36_loss: 0.0338 - val_dense_43_loss: 0.0179 - val_dense_46_loss: 0.0249 - val_dense_36_mean_absolute_percentage_error: 7.8437 - val_dense_43_mean_absolute_percentage_error: 10.1809 - val_dense_46_mean_absolute_percentage_error: 10.9412\n",
      "Epoch 31/40\n",
      "\n",
      "Epoch 00031: LearningRateScheduler setting learning rate to 6.310669032775445e-05.\n",
      "49749/49749 [==============================] - 13s 266us/step - loss: 0.0740 - dense_36_loss: 0.0329 - dense_43_loss: 0.0173 - dense_46_loss: 0.0238 - dense_36_mean_absolute_percentage_error: 42311.3509 - dense_43_mean_absolute_percentage_error: 46743.8050 - dense_46_mean_absolute_percentage_error: 50237.1747 - val_loss: 0.0713 - val_dense_36_loss: 0.0318 - val_dense_43_loss: 0.0162 - val_dense_46_loss: 0.0232 - val_dense_36_mean_absolute_percentage_error: 7.3674 - val_dense_43_mean_absolute_percentage_error: 11.1926 - val_dense_46_mean_absolute_percentage_error: 11.5766\n",
      "Epoch 32/40\n",
      "\n",
      "Epoch 00032: LearningRateScheduler setting learning rate to 6.120920161627212e-05.\n",
      "49749/49749 [==============================] - 13s 267us/step - loss: 0.0706 - dense_36_loss: 0.0318 - dense_43_loss: 0.0155 - dense_46_loss: 0.0232 - dense_36_mean_absolute_percentage_error: 41454.8616 - dense_43_mean_absolute_percentage_error: 41913.2288 - dense_46_mean_absolute_percentage_error: 47883.5820 - val_loss: 0.0695 - val_dense_36_loss: 0.0310 - val_dense_43_loss: 0.0161 - val_dense_46_loss: 0.0225 - val_dense_36_mean_absolute_percentage_error: 6.9524 - val_dense_43_mean_absolute_percentage_error: 9.4894 - val_dense_46_mean_absolute_percentage_error: 12.3232\n",
      "Epoch 33/40\n",
      "\n",
      "Epoch 00033: LearningRateScheduler setting learning rate to 5.9311244604737285e-05.\n",
      "49749/49749 [==============================] - 13s 268us/step - loss: 0.0686 - dense_36_loss: 0.0312 - dense_43_loss: 0.0145 - dense_46_loss: 0.0229 - dense_36_mean_absolute_percentage_error: 37876.8435 - dense_43_mean_absolute_percentage_error: 40336.9398 - dense_46_mean_absolute_percentage_error: 49925.5565 - val_loss: 0.0655 - val_dense_36_loss: 0.0303 - val_dense_43_loss: 0.0125 - val_dense_46_loss: 0.0226 - val_dense_36_mean_absolute_percentage_error: 7.1892 - val_dense_43_mean_absolute_percentage_error: 9.0005 - val_dense_46_mean_absolute_percentage_error: 11.4496\n",
      "Epoch 34/40\n",
      "\n",
      "Epoch 00034: LearningRateScheduler setting learning rate to 5.741650089310502e-05.\n",
      "49749/49749 [==============================] - 14s 271us/step - loss: 0.0647 - dense_36_loss: 0.0306 - dense_43_loss: 0.0116 - dense_46_loss: 0.0225 - dense_36_mean_absolute_percentage_error: 42046.3071 - dense_43_mean_absolute_percentage_error: 40309.0243 - dense_46_mean_absolute_percentage_error: 49883.2071 - val_loss: 0.0652 - val_dense_36_loss: 0.0297 - val_dense_43_loss: 0.0138 - val_dense_46_loss: 0.0218 - val_dense_36_mean_absolute_percentage_error: 6.9713 - val_dense_43_mean_absolute_percentage_error: 9.8641 - val_dense_46_mean_absolute_percentage_error: 12.0360\n",
      "Epoch 35/40\n",
      "\n",
      "Epoch 00035: LearningRateScheduler setting learning rate to 5.5528532288521814e-05.\n",
      "49749/49749 [==============================] - 14s 280us/step - loss: 0.0609 - dense_36_loss: 0.0299 - dense_43_loss: 0.0089 - dense_46_loss: 0.0221 - dense_36_mean_absolute_percentage_error: 42980.1227 - dense_43_mean_absolute_percentage_error: 33538.7253 - dense_46_mean_absolute_percentage_error: 51513.8161 - val_loss: 0.0581 - val_dense_36_loss: 0.0291 - val_dense_43_loss: 0.0076 - val_dense_46_loss: 0.0214 - val_dense_36_mean_absolute_percentage_error: 6.7384 - val_dense_43_mean_absolute_percentage_error: 8.0279 - val_dense_46_mean_absolute_percentage_error: 12.4352\n",
      "Epoch 36/40\n",
      "\n",
      "Epoch 00036: LearningRateScheduler setting learning rate to 5.365075677934729e-05.\n",
      "49749/49749 [==============================] - 13s 271us/step - loss: 0.0582 - dense_36_loss: 0.0294 - dense_43_loss: 0.0070 - dense_46_loss: 0.0218 - dense_36_mean_absolute_percentage_error: 40548.9975 - dense_43_mean_absolute_percentage_error: 28947.3857 - dense_46_mean_absolute_percentage_error: 52309.5817 - val_loss: 0.0552 - val_dense_36_loss: 0.0297 - val_dense_43_loss: 0.0042 - val_dense_46_loss: 0.0213 - val_dense_36_mean_absolute_percentage_error: 7.5334 - val_dense_43_mean_absolute_percentage_error: 5.0273 - val_dense_46_mean_absolute_percentage_error: 11.8840\n",
      "Epoch 37/40\n",
      "\n",
      "Epoch 00037: LearningRateScheduler setting learning rate to 5.178644571769767e-05.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "49749/49749 [==============================] - 13s 269us/step - loss: 0.0546 - dense_36_loss: 0.0286 - dense_43_loss: 0.0047 - dense_46_loss: 0.0213 - dense_36_mean_absolute_percentage_error: 38336.9047 - dense_43_mean_absolute_percentage_error: 39010.4090 - dense_46_mean_absolute_percentage_error: 56650.0648 - val_loss: 0.0527 - val_dense_36_loss: 0.0280 - val_dense_43_loss: 0.0043 - val_dense_46_loss: 0.0204 - val_dense_36_mean_absolute_percentage_error: 6.7114 - val_dense_43_mean_absolute_percentage_error: 5.0403 - val_dense_46_mean_absolute_percentage_error: 12.3929\n",
      "Epoch 38/40\n",
      "\n",
      "Epoch 00038: LearningRateScheduler setting learning rate to 4.993871400193846e-05.\n",
      "49749/49749 [==============================] - 13s 270us/step - loss: 0.0522 - dense_36_loss: 0.0279 - dense_43_loss: 0.0037 - dense_46_loss: 0.0206 - dense_36_mean_absolute_percentage_error: 39475.0744 - dense_43_mean_absolute_percentage_error: 17863.0592 - dense_46_mean_absolute_percentage_error: 56033.3056 - val_loss: 0.0508 - val_dense_36_loss: 0.0277 - val_dense_43_loss: 0.0021 - val_dense_46_loss: 0.0209 - val_dense_36_mean_absolute_percentage_error: 7.3128 - val_dense_43_mean_absolute_percentage_error: 3.7709 - val_dense_46_mean_absolute_percentage_error: 11.3116\n",
      "Epoch 39/40\n",
      "\n",
      "Epoch 00039: LearningRateScheduler setting learning rate to 4.811051382072205e-05.\n",
      "49749/49749 [==============================] - 13s 270us/step - loss: 0.0535 - dense_36_loss: 0.0275 - dense_43_loss: 0.0059 - dense_46_loss: 0.0201 - dense_36_mean_absolute_percentage_error: 40247.7431 - dense_43_mean_absolute_percentage_error: 17682.3946 - dense_46_mean_absolute_percentage_error: 56128.6635 - val_loss: 0.0472 - val_dense_36_loss: 0.0267 - val_dense_43_loss: 0.0016 - val_dense_46_loss: 0.0190 - val_dense_36_mean_absolute_percentage_error: 6.7975 - val_dense_43_mean_absolute_percentage_error: 3.5655 - val_dense_46_mean_absolute_percentage_error: 11.9997\n",
      "Epoch 40/40\n",
      "\n",
      "Epoch 00040: LearningRateScheduler setting learning rate to 4.6304631934575634e-05.\n",
      "49749/49749 [==============================] - 15s 297us/step - loss: 0.0486 - dense_36_loss: 0.0264 - dense_43_loss: 0.0029 - dense_46_loss: 0.0192 - dense_36_mean_absolute_percentage_error: 40334.2111 - dense_43_mean_absolute_percentage_error: 18041.3354 - dense_46_mean_absolute_percentage_error: 57282.3890 - val_loss: 0.0465 - val_dense_36_loss: 0.0265 - val_dense_43_loss: 0.0016 - val_dense_46_loss: 0.0183 - val_dense_36_mean_absolute_percentage_error: 7.1923 - val_dense_43_mean_absolute_percentage_error: 2.8838 - val_dense_46_mean_absolute_percentage_error: 11.6216\n"
     ]
    }
   ],
   "source": [
    "history = model.fit([X_train], [Y1,Y2,Y3], epochs=40, batch_size=32,validation_data=([X_test],[Y1_,Y2_,Y3_]),callbacks=[lrate])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Model loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Test'], loc='upper left')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_ = random_scale.transform(X)\n",
    "output_ = model.predict(input_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "output_ = np.transpose(np.array(output_))\n",
    "output_ = green_scale.inverse_transform(output_.reshape(output_.shape[1],3))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f5a3421cf98>"
      ]
     },
     "execution_count": 54,
     "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": [
    "plt.scatter(output_[:,0],green_[:,0],alpha=0.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "G         12.772139\n",
       "damu      69.942667\n",
       "dbeta    325.809673\n",
       "dtype: float64"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y.max(axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f5a29514588>"
      ]
     },
     "execution_count": 76,
     "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": [
    "plt.scatter(X['beta'][0:71071],Y[\"G\"][0:71071])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_filon = np.loadtxt(\"./Filon/random.txt\")\n",
    "green_filon = np.loadtxt(\"./Filon/green.txt\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = pd.DataFrame(columns=[\"amu\",\"beta\"])\n",
    "X[\"amu\"] = random_filon[:,0]\n",
    "X[\"beta\"] = random_filon[:,1]\n",
    "Y = pd.DataFrame(columns=[\"G\",\"damu\",\"dbeta\"])\n",
    "Y[\"G\"] = green_filon[:,0]\n",
    "Y[\"damu\"] = green_filon[:,1]\n",
    "Y[\"dbeta\"] = green_filon[:,2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f479d944550>"
      ]
     },
     "execution_count": 10,
     "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",
    "plt.scatter(X['beta'][0:40629],Y[\"G\"][0:40629])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_filon = np.loadtxt(\"./asy/random.txt\")\n",
    "green_filon = np.loadtxt(\"./asy/green.txt\")\n",
    "X = pd.DataFrame(columns=[\"amu\",\"beta\"])\n",
    "X[\"amu\"] = random_filon[:,0]\n",
    "X[\"beta\"] = random_filon[:,1]\n",
    "Y = pd.DataFrame(columns=[\"G\",\"damu\",\"dbeta\"])\n",
    "Y[\"G\"] = green_filon[:,0]\n",
    "Y[\"damu\"] = green_filon[:,1]\n",
    "Y[\"dbeta\"] = green_filon[:,2]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f479998c438>"
      ]
     },
     "execution_count": 25,
     "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": [
    "plt.scatter(X['beta'][0:851499],Y[\"dbeta\"][0:851499])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_filon = np.loadtxt(\"./Bessel/random.txt\")\n",
    "green_filon = np.loadtxt(\"./Bessel/green.txt\")\n",
    "X = pd.DataFrame(columns=[\"amu\",\"beta\"])\n",
    "X[\"amu\"] = random_filon[:,0]\n",
    "X[\"beta\"] = random_filon[:,1]\n",
    "Y = pd.DataFrame(columns=[\"G\",\"damu\",\"dbeta\"])\n",
    "Y[\"G\"] = green_filon[:,0]\n",
    "Y[\"damu\"] = green_filon[:,1]\n",
    "Y[\"dbeta\"] = green_filon[:,2]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f4799117400>"
      ]
     },
     "execution_count": 20,
     "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": [
    "plt.scatter(X['beta'][0:30003],Y[\"damu\"][0:30003])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f4799c71dd8>"
      ]
     },
     "execution_count": 22,
     "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": [
    "plt.scatter(X['beta'][0:30003],Y[\"dbeta\"][0:30003])"
   ]
  },
  {
   "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
}