LSTM2.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://medium.com/@mrconnor/forecasting-the-stock-market-using-lstm-will-it-rise-tomorrow-94ff6b6a34b6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 351,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 352,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.arange(0,100,0.5) \n",
    "y = np.sin(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 353,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x245070570a0>]"
      ]
     },
     "execution_count": 353,
     "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": [
    "plt.plot(x, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert an array of values into a dataset matrix\n",
    "def create_dataset(dataset, look_back=15):\n",
    "    dataX, dataY = [], []\n",
    "    for i in range(len(dataset)-look_back-1):\n",
    "        a = dataset[i:(i+look_back), 0]\n",
    "        dataX.append(a)\n",
    "        dataY.append(dataset[i + look_back, 0])\n",
    "    return np.array(dataX), np.array(dataY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, y_train = create_dataset(train, look_back=15)\n",
    "x_test, y_test = create_dataset(test, look_back=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 354,
   "metadata": {},
   "outputs": [],
   "source": [
    "trainX = np.array(x)\n",
    "trainY = np.array(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 355,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 200, 1)\n",
      "(1, 200)\n"
     ]
    }
   ],
   "source": [
    "trainX = np.reshape(trainX, (1, trainX.shape[0], 1))\n",
    "trainY = np.reshape(trainY, (1, trainY.shape[0]))\n",
    "print(trainX.shape)\n",
    "print(trainY.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 356,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.LSTM(4, batch_input_shape=(1, 200, 1)),\n",
    "    tf.keras.layers.Dense(1)\n",
    "])\n",
    "\n",
    "model.compile(optimizer='adam', loss='mse')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 357,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/200\n",
      "1/1 [==============================] - 3s 3s/step - loss: 0.8127\n",
      "Epoch 2/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.8099\n",
      "Epoch 3/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.8071\n",
      "Epoch 4/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.8044\n",
      "Epoch 5/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.8016\n",
      "Epoch 6/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.7988\n",
      "Epoch 7/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.7961\n",
      "Epoch 8/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.7934\n",
      "Epoch 9/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.7906\n",
      "Epoch 10/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.7879\n",
      "Epoch 11/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.7853\n",
      "Epoch 12/200\n",
      "1/1 [==============================] - 0s 58ms/step - loss: 0.7826\n",
      "Epoch 13/200\n",
      "1/1 [==============================] - 0s 50ms/step - loss: 0.7799\n",
      "Epoch 14/200\n",
      "1/1 [==============================] - 0s 44ms/step - loss: 0.7773\n",
      "Epoch 15/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.7747\n",
      "Epoch 16/200\n",
      "1/1 [==============================] - 0s 45ms/step - loss: 0.7720\n",
      "Epoch 17/200\n",
      "1/1 [==============================] - 0s 67ms/step - loss: 0.7695\n",
      "Epoch 18/200\n",
      "1/1 [==============================] - 0s 48ms/step - loss: 0.7669\n",
      "Epoch 19/200\n",
      "1/1 [==============================] - 0s 44ms/step - loss: 0.7643\n",
      "Epoch 20/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.7618\n",
      "Epoch 21/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.7592\n",
      "Epoch 22/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.7567\n",
      "Epoch 23/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.7542\n",
      "Epoch 24/200\n",
      "1/1 [==============================] - 0s 42ms/step - loss: 0.7517\n",
      "Epoch 25/200\n",
      "1/1 [==============================] - 0s 47ms/step - loss: 0.7492\n",
      "Epoch 26/200\n",
      "1/1 [==============================] - 0s 59ms/step - loss: 0.7468\n",
      "Epoch 27/200\n",
      "1/1 [==============================] - 0s 45ms/step - loss: 0.7443\n",
      "Epoch 28/200\n",
      "1/1 [==============================] - 0s 42ms/step - loss: 0.7419\n",
      "Epoch 29/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.7395\n",
      "Epoch 30/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.7371\n",
      "Epoch 31/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.7347\n",
      "Epoch 32/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.7323\n",
      "Epoch 33/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.7300\n",
      "Epoch 34/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.7276\n",
      "Epoch 35/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.7253\n",
      "Epoch 36/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.7229\n",
      "Epoch 37/200\n",
      "1/1 [==============================] - 0s 31ms/step - loss: 0.7206\n",
      "Epoch 38/200\n",
      "1/1 [==============================] - 0s 31ms/step - loss: 0.7183\n",
      "Epoch 39/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.7160\n",
      "Epoch 40/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.7137\n",
      "Epoch 41/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.7114\n",
      "Epoch 42/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.7091\n",
      "Epoch 43/200\n",
      "1/1 [==============================] - 0s 43ms/step - loss: 0.7068\n",
      "Epoch 44/200\n",
      "1/1 [==============================] - 0s 46ms/step - loss: 0.7045\n",
      "Epoch 45/200\n",
      "1/1 [==============================] - 0s 47ms/step - loss: 0.7022\n",
      "Epoch 46/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.6998\n",
      "Epoch 47/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.6975\n",
      "Epoch 48/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.6951\n",
      "Epoch 49/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.6926\n",
      "Epoch 50/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.6902\n",
      "Epoch 51/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.6876\n",
      "Epoch 52/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.6850\n",
      "Epoch 53/200\n",
      "1/1 [==============================] - 0s 48ms/step - loss: 0.6823\n",
      "Epoch 54/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.6796\n",
      "Epoch 55/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.6767\n",
      "Epoch 56/200\n",
      "1/1 [==============================] - 0s 65ms/step - loss: 0.6737\n",
      "Epoch 57/200\n",
      "1/1 [==============================] - 0s 43ms/step - loss: 0.6706\n",
      "Epoch 58/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.6673\n",
      "Epoch 59/200\n",
      "1/1 [==============================] - 0s 64ms/step - loss: 0.6639\n",
      "Epoch 60/200\n",
      "1/1 [==============================] - 0s 50ms/step - loss: 0.6604\n",
      "Epoch 61/200\n",
      "1/1 [==============================] - 0s 49ms/step - loss: 0.6568\n",
      "Epoch 62/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.6531\n",
      "Epoch 63/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.6494\n",
      "Epoch 64/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.6457\n",
      "Epoch 65/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.6422\n",
      "Epoch 66/200\n",
      "1/1 [==============================] - 0s 30ms/step - loss: 0.6390\n",
      "Epoch 67/200\n",
      "1/1 [==============================] - 0s 31ms/step - loss: 0.6361\n",
      "Epoch 68/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.6334\n",
      "Epoch 69/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.6311\n",
      "Epoch 70/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.6290\n",
      "Epoch 71/200\n",
      "1/1 [==============================] - 0s 46ms/step - loss: 0.6271\n",
      "Epoch 72/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.6253\n",
      "Epoch 73/200\n",
      "1/1 [==============================] - 0s 47ms/step - loss: 0.6235\n",
      "Epoch 74/200\n",
      "1/1 [==============================] - 0s 46ms/step - loss: 0.6218\n",
      "Epoch 75/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.6200\n",
      "Epoch 76/200\n",
      "1/1 [==============================] - 0s 42ms/step - loss: 0.6183\n",
      "Epoch 77/200\n",
      "1/1 [==============================] - 0s 57ms/step - loss: 0.6167\n",
      "Epoch 78/200\n",
      "1/1 [==============================] - 0s 45ms/step - loss: 0.6150\n",
      "Epoch 79/200\n",
      "1/1 [==============================] - 0s 46ms/step - loss: 0.6133\n",
      "Epoch 80/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.6116\n",
      "Epoch 81/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.6099\n",
      "Epoch 82/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.6082\n",
      "Epoch 83/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.6064\n",
      "Epoch 84/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.6047\n",
      "Epoch 85/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.6028\n",
      "Epoch 86/200\n",
      "1/1 [==============================] - 0s 44ms/step - loss: 0.6010\n",
      "Epoch 87/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.5990\n",
      "Epoch 88/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.5970\n",
      "Epoch 89/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5948\n",
      "Epoch 90/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.5925\n",
      "Epoch 91/200\n",
      "1/1 [==============================] - 0s 30ms/step - loss: 0.5900\n",
      "Epoch 92/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.5873\n",
      "Epoch 93/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5843\n",
      "Epoch 94/200\n",
      "1/1 [==============================] - 0s 29ms/step - loss: 0.5810\n",
      "Epoch 95/200\n",
      "1/1 [==============================] - 0s 43ms/step - loss: 0.5773\n",
      "Epoch 96/200\n",
      "1/1 [==============================] - 0s 67ms/step - loss: 0.5731\n",
      "Epoch 97/200\n",
      "1/1 [==============================] - 0s 55ms/step - loss: 0.5684\n",
      "Epoch 98/200\n",
      "1/1 [==============================] - 0s 52ms/step - loss: 0.5631\n",
      "Epoch 99/200\n",
      "1/1 [==============================] - 0s 46ms/step - loss: 0.5572\n",
      "Epoch 100/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.5507\n",
      "Epoch 101/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.5435\n",
      "Epoch 102/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.5358\n",
      "Epoch 103/200\n",
      "1/1 [==============================] - 0s 30ms/step - loss: 0.5279\n",
      "Epoch 104/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5201\n",
      "Epoch 105/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.5131\n",
      "Epoch 106/200\n",
      "1/1 [==============================] - 0s 31ms/step - loss: 0.5073\n",
      "Epoch 107/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.5033\n",
      "Epoch 108/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.5015\n",
      "Epoch 109/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.5016\n",
      "Epoch 110/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5032\n",
      "Epoch 111/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.5054\n",
      "Epoch 112/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5075\n",
      "Epoch 113/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5091\n",
      "Epoch 114/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.5101\n",
      "Epoch 115/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5104\n",
      "Epoch 116/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.5101\n",
      "Epoch 117/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.5092\n",
      "Epoch 118/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.5080\n",
      "Epoch 119/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.5065\n",
      "Epoch 120/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.5051\n",
      "Epoch 121/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.5037\n",
      "Epoch 122/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.5026\n",
      "Epoch 123/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.5018\n",
      "Epoch 124/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5014\n",
      "Epoch 125/200\n",
      "1/1 [==============================] - 0s 31ms/step - loss: 0.5014\n",
      "Epoch 126/200\n",
      "1/1 [==============================] - 0s 31ms/step - loss: 0.5015\n",
      "Epoch 127/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5019\n",
      "Epoch 128/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.5023\n",
      "Epoch 129/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.5026\n",
      "Epoch 130/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5029\n",
      "Epoch 131/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.5030\n",
      "Epoch 132/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5030\n",
      "Epoch 133/200\n",
      "1/1 [==============================] - 0s 47ms/step - loss: 0.5028\n",
      "Epoch 134/200\n",
      "1/1 [==============================] - 0s 67ms/step - loss: 0.5026\n",
      "Epoch 135/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.5023\n",
      "Epoch 136/200\n",
      "1/1 [==============================] - 0s 47ms/step - loss: 0.5020\n",
      "Epoch 137/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.5018\n",
      "Epoch 138/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5016\n",
      "Epoch 139/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.5014\n",
      "Epoch 140/200\n",
      "1/1 [==============================] - 0s 44ms/step - loss: 0.5014\n",
      "Epoch 141/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.5014\n",
      "Epoch 142/200\n",
      "1/1 [==============================] - 0s 42ms/step - loss: 0.5014\n",
      "Epoch 143/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.5015\n",
      "Epoch 144/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5015\n",
      "Epoch 145/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5016\n",
      "Epoch 146/200\n",
      "1/1 [==============================] - 0s 47ms/step - loss: 0.5017\n",
      "Epoch 147/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.5017\n",
      "Epoch 148/200\n",
      "1/1 [==============================] - 0s 42ms/step - loss: 0.5017\n",
      "Epoch 149/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.5016\n",
      "Epoch 150/200\n",
      "1/1 [==============================] - 0s 30ms/step - loss: 0.5016\n",
      "Epoch 151/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5015\n",
      "Epoch 152/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5015\n",
      "Epoch 153/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.5014\n",
      "Epoch 154/200\n",
      "1/1 [==============================] - 0s 31ms/step - loss: 0.5014\n",
      "Epoch 155/200\n",
      "1/1 [==============================] - 0s 44ms/step - loss: 0.5014\n",
      "Epoch 156/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.5014\n",
      "Epoch 157/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5014\n",
      "Epoch 158/200\n",
      "1/1 [==============================] - 0s 46ms/step - loss: 0.5014\n",
      "Epoch 159/200\n",
      "1/1 [==============================] - 0s 43ms/step - loss: 0.5014\n",
      "Epoch 160/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.5014\n",
      "Epoch 161/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.5014\n",
      "Epoch 162/200\n",
      "1/1 [==============================] - 0s 43ms/step - loss: 0.5014\n",
      "Epoch 163/200\n",
      "1/1 [==============================] - 0s 47ms/step - loss: 0.5014\n",
      "Epoch 164/200\n",
      "1/1 [==============================] - 0s 49ms/step - loss: 0.5014\n",
      "Epoch 165/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.5014\n",
      "Epoch 166/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5014\n",
      "Epoch 167/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.5014\n",
      "Epoch 168/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.5014\n",
      "Epoch 169/200\n",
      "1/1 [==============================] - 0s 67ms/step - loss: 0.5014\n",
      "Epoch 170/200\n",
      "1/1 [==============================] - 0s 55ms/step - loss: 0.5014\n",
      "Epoch 171/200\n",
      "1/1 [==============================] - 0s 46ms/step - loss: 0.5014\n",
      "Epoch 172/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.5014\n",
      "Epoch 173/200\n",
      "1/1 [==============================] - 0s 40ms/step - loss: 0.5014\n",
      "Epoch 174/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.5014\n",
      "Epoch 175/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.5014\n",
      "Epoch 176/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.5014\n",
      "Epoch 177/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.5014\n",
      "Epoch 178/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.5014\n",
      "Epoch 179/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5014\n",
      "Epoch 180/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5014\n",
      "Epoch 181/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5014\n",
      "Epoch 182/200\n",
      "1/1 [==============================] - 0s 51ms/step - loss: 0.5014\n",
      "Epoch 183/200\n",
      "1/1 [==============================] - 0s 43ms/step - loss: 0.5014\n",
      "Epoch 184/200\n",
      "1/1 [==============================] - 0s 37ms/step - loss: 0.5014\n",
      "Epoch 185/200\n",
      "1/1 [==============================] - 0s 45ms/step - loss: 0.5014\n",
      "Epoch 186/200\n",
      "1/1 [==============================] - 0s 43ms/step - loss: 0.5014\n",
      "Epoch 187/200\n",
      "1/1 [==============================] - 0s 41ms/step - loss: 0.5014\n",
      "Epoch 188/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.5014\n",
      "Epoch 189/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.5014\n",
      "Epoch 190/200\n",
      "1/1 [==============================] - 0s 33ms/step - loss: 0.5014\n",
      "Epoch 191/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.5014\n",
      "Epoch 192/200\n",
      "1/1 [==============================] - 0s 29ms/step - loss: 0.5014\n",
      "Epoch 193/200\n",
      "1/1 [==============================] - 0s 39ms/step - loss: 0.5014\n",
      "Epoch 194/200\n",
      "1/1 [==============================] - 0s 38ms/step - loss: 0.5014\n",
      "Epoch 195/200\n",
      "1/1 [==============================] - 0s 36ms/step - loss: 0.5014\n",
      "Epoch 196/200\n",
      "1/1 [==============================] - 0s 32ms/step - loss: 0.5014\n",
      "Epoch 197/200\n",
      "1/1 [==============================] - 0s 31ms/step - loss: 0.5014\n",
      "Epoch 198/200\n",
      "1/1 [==============================] - 0s 34ms/step - loss: 0.5014\n",
      "Epoch 199/200\n",
      "1/1 [==============================] - 0s 35ms/step - loss: 0.5014\n",
      "Epoch 200/200\n",
      "1/1 [==============================] - 0s 47ms/step - loss: 0.5014\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x24505ee3700>"
      ]
     },
     "execution_count": 357,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(trainX, trainY, epochs=200, batch_size=1, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 360,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 200, 1)\n",
      "1/1 [==============================] - 0s 43ms/step\n",
      "(1, 1)\n"
     ]
    }
   ],
   "source": [
    "print(trainX.shape)\n",
    "trainPredict = model.predict(trainX)\n",
    "print(trainPredict.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 359,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x2454e63f1c0>]"
      ]
     },
     "execution_count": 359,
     "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": [
    "# print(trainPredict[:,0])\n",
    "plt.plot(trainX[:,0,0], trainPredict[:,0])"
   ]
  }
 ],
 "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": "3.9.1"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "e8e86f11f8de651d2c5fce83c6190e2e2c0f42ae90d60e5c5b35a501d45892ca"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}