Simple_RNN_sine.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyO2EpeWHUGar89sYLB0vZP0",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/tusharbansal22/Tensorflow-Notebooks/blob/main/Simple_RNN_sine.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Eq6FEdXP6zs_",
        "outputId": "47e61f34-d0f3-4193-bfd5-5031fb64fe26"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.9.2\n"
          ]
        }
      ],
      "source": [
        "import tensorflow as tf\n",
        "print(tf.__version__)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.layers import Input, SimpleRNN, Dense,Flatten\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.optimizers import SGD, Adam\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "m2GD_NI066e1"
      },
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "series = np.sin(0.1*np.arange(200)) + np.random.randn(200)*0.1\n",
        "\n",
        "plt.plot(series)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "8uQIqBZg7TEk",
        "outputId": "51d0f9a3-004e-45a0-d9f1-c057ead81399"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "T=10\n",
        "D=1\n",
        "X=[]\n",
        "Y=[]\n",
        "for t in range(len(series)-T):\n",
        "  x=series[t:t+T]\n",
        "  X.append(x)\n",
        "  y=series[t+T]\n",
        "  Y.append(y)\n",
        "\n",
        "X=np.array(X).reshape(-1,T,1)\n",
        "Y=np.array(Y)\n",
        "N=len(X)\n",
        "print(\"X.shape: \",X.shape, \" Y.shape \",Y.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nSb3ozUA7xUZ",
        "outputId": "3dc67362-8d7f-4f07-efa1-94a7f2e5f4f7"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X.shape:  (190, 10, 1)  Y.shape  (190,)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "i=Input(shape=(T,1))\n",
        "x=SimpleRNN(5, activation='relu')(i)\n",
        "x=Dense(1)(x)\n",
        "model=Model(i,x)\n",
        "model.compile(loss='mse',optimizer=Adam(lr=0.1))\n",
        "\n",
        "r=model.fit(X[:N//2],Y[:-N//2],epochs=80,validation_data=(X[-N//2:],Y[-N//2:]))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RkR4xOdt8kf8",
        "outputId": "343d2e62-d372-4a40-d8cc-aa75d4947299"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/80\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.8/dist-packages/keras/optimizers/optimizer_v2/adam.py:110: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "3/3 [==============================] - 1s 129ms/step - loss: 0.4566 - val_loss: 0.1898\n",
            "Epoch 2/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.1330 - val_loss: 0.0658\n",
            "Epoch 3/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0779 - val_loss: 0.1038\n",
            "Epoch 4/80\n",
            "3/3 [==============================] - 0s 29ms/step - loss: 0.1114 - val_loss: 0.0967\n",
            "Epoch 5/80\n",
            "3/3 [==============================] - 0s 19ms/step - loss: 0.0956 - val_loss: 0.0539\n",
            "Epoch 6/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0502 - val_loss: 0.0336\n",
            "Epoch 7/80\n",
            "3/3 [==============================] - 0s 16ms/step - loss: 0.0375 - val_loss: 0.0561\n",
            "Epoch 8/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0543 - val_loss: 0.0475\n",
            "Epoch 9/80\n",
            "3/3 [==============================] - 0s 24ms/step - loss: 0.0388 - val_loss: 0.0346\n",
            "Epoch 10/80\n",
            "3/3 [==============================] - 0s 19ms/step - loss: 0.0313 - val_loss: 0.0358\n",
            "Epoch 11/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0359 - val_loss: 0.0342\n",
            "Epoch 12/80\n",
            "3/3 [==============================] - 0s 24ms/step - loss: 0.0333 - val_loss: 0.0300\n",
            "Epoch 13/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0282 - val_loss: 0.0282\n",
            "Epoch 14/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0291 - val_loss: 0.0295\n",
            "Epoch 15/80\n",
            "3/3 [==============================] - 0s 19ms/step - loss: 0.0288 - val_loss: 0.0324\n",
            "Epoch 16/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0274 - val_loss: 0.0289\n",
            "Epoch 17/80\n",
            "3/3 [==============================] - 0s 19ms/step - loss: 0.0253 - val_loss: 0.0266\n",
            "Epoch 18/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0259 - val_loss: 0.0264\n",
            "Epoch 19/80\n",
            "3/3 [==============================] - 0s 27ms/step - loss: 0.0245 - val_loss: 0.0270\n",
            "Epoch 20/80\n",
            "3/3 [==============================] - 0s 29ms/step - loss: 0.0243 - val_loss: 0.0292\n",
            "Epoch 21/80\n",
            "3/3 [==============================] - 0s 20ms/step - loss: 0.0240 - val_loss: 0.0271\n",
            "Epoch 22/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0232 - val_loss: 0.0255\n",
            "Epoch 23/80\n",
            "3/3 [==============================] - 0s 19ms/step - loss: 0.0229 - val_loss: 0.0257\n",
            "Epoch 24/80\n",
            "3/3 [==============================] - 0s 26ms/step - loss: 0.0228 - val_loss: 0.0259\n",
            "Epoch 25/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0225 - val_loss: 0.0255\n",
            "Epoch 26/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0225 - val_loss: 0.0255\n",
            "Epoch 27/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0221 - val_loss: 0.0260\n",
            "Epoch 28/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0221 - val_loss: 0.0254\n",
            "Epoch 29/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0222 - val_loss: 0.0242\n",
            "Epoch 30/80\n",
            "3/3 [==============================] - 0s 24ms/step - loss: 0.0219 - val_loss: 0.0239\n",
            "Epoch 31/80\n",
            "3/3 [==============================] - 0s 24ms/step - loss: 0.0219 - val_loss: 0.0265\n",
            "Epoch 32/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0220 - val_loss: 0.0252\n",
            "Epoch 33/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0227 - val_loss: 0.0238\n",
            "Epoch 34/80\n",
            "3/3 [==============================] - 0s 20ms/step - loss: 0.0216 - val_loss: 0.0254\n",
            "Epoch 35/80\n",
            "3/3 [==============================] - 0s 16ms/step - loss: 0.0219 - val_loss: 0.0253\n",
            "Epoch 36/80\n",
            "3/3 [==============================] - 0s 22ms/step - loss: 0.0212 - val_loss: 0.0233\n",
            "Epoch 37/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0217 - val_loss: 0.0235\n",
            "Epoch 38/80\n",
            "3/3 [==============================] - 0s 27ms/step - loss: 0.0214 - val_loss: 0.0248\n",
            "Epoch 39/80\n",
            "3/3 [==============================] - 0s 21ms/step - loss: 0.0215 - val_loss: 0.0234\n",
            "Epoch 40/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0231 - val_loss: 0.0233\n",
            "Epoch 41/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0213 - val_loss: 0.0268\n",
            "Epoch 42/80\n",
            "3/3 [==============================] - 0s 28ms/step - loss: 0.0222 - val_loss: 0.0233\n",
            "Epoch 43/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0209 - val_loss: 0.0232\n",
            "Epoch 44/80\n",
            "3/3 [==============================] - 0s 26ms/step - loss: 0.0212 - val_loss: 0.0236\n",
            "Epoch 45/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0210 - val_loss: 0.0240\n",
            "Epoch 46/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0209 - val_loss: 0.0234\n",
            "Epoch 47/80\n",
            "3/3 [==============================] - 0s 26ms/step - loss: 0.0209 - val_loss: 0.0233\n",
            "Epoch 48/80\n",
            "3/3 [==============================] - 0s 21ms/step - loss: 0.0211 - val_loss: 0.0255\n",
            "Epoch 49/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0217 - val_loss: 0.0234\n",
            "Epoch 50/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0209 - val_loss: 0.0230\n",
            "Epoch 51/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0209 - val_loss: 0.0240\n",
            "Epoch 52/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0214 - val_loss: 0.0241\n",
            "Epoch 53/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0214 - val_loss: 0.0239\n",
            "Epoch 54/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0210 - val_loss: 0.0227\n",
            "Epoch 55/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0210 - val_loss: 0.0237\n",
            "Epoch 56/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0231 - val_loss: 0.0252\n",
            "Epoch 57/80\n",
            "3/3 [==============================] - 0s 30ms/step - loss: 0.0222 - val_loss: 0.0271\n",
            "Epoch 58/80\n",
            "3/3 [==============================] - 0s 19ms/step - loss: 0.0240 - val_loss: 0.0234\n",
            "Epoch 59/80\n",
            "3/3 [==============================] - 0s 22ms/step - loss: 0.0219 - val_loss: 0.0278\n",
            "Epoch 60/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0241 - val_loss: 0.0276\n",
            "Epoch 61/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0226 - val_loss: 0.0240\n",
            "Epoch 62/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0218 - val_loss: 0.0229\n",
            "Epoch 63/80\n",
            "3/3 [==============================] - 0s 24ms/step - loss: 0.0226 - val_loss: 0.0275\n",
            "Epoch 64/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0224 - val_loss: 0.0264\n",
            "Epoch 65/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0235 - val_loss: 0.0237\n",
            "Epoch 66/80\n",
            "3/3 [==============================] - 0s 25ms/step - loss: 0.0222 - val_loss: 0.0251\n",
            "Epoch 67/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0236 - val_loss: 0.0238\n",
            "Epoch 68/80\n",
            "3/3 [==============================] - 0s 19ms/step - loss: 0.0206 - val_loss: 0.0240\n",
            "Epoch 69/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0219 - val_loss: 0.0243\n",
            "Epoch 70/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0207 - val_loss: 0.0241\n",
            "Epoch 71/80\n",
            "3/3 [==============================] - 0s 16ms/step - loss: 0.0218 - val_loss: 0.0230\n",
            "Epoch 72/80\n",
            "3/3 [==============================] - 0s 16ms/step - loss: 0.0211 - val_loss: 0.0250\n",
            "Epoch 73/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0218 - val_loss: 0.0253\n",
            "Epoch 74/80\n",
            "3/3 [==============================] - 0s 18ms/step - loss: 0.0215 - val_loss: 0.0238\n",
            "Epoch 75/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0207 - val_loss: 0.0236\n",
            "Epoch 76/80\n",
            "3/3 [==============================] - 0s 23ms/step - loss: 0.0213 - val_loss: 0.0236\n",
            "Epoch 77/80\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0220 - val_loss: 0.0235\n",
            "Epoch 78/80\n",
            "3/3 [==============================] - 0s 26ms/step - loss: 0.0208 - val_loss: 0.0261\n",
            "Epoch 79/80\n",
            "3/3 [==============================] - 0s 19ms/step - loss: 0.0212 - val_loss: 0.0235\n",
            "Epoch 80/80\n",
            "3/3 [==============================] - 0s 24ms/step - loss: 0.0226 - val_loss: 0.0247\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(r.history['loss'],label='loss')\n",
        "plt.plot(r.history['val_loss'],label='val_loss')\n",
        "plt.legend()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "YAZ4VmqR9R3d",
        "outputId": "6eabec77-f874-4a57-fbf0-ed3fe55bd386"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f44f9645a60>"
            ]
          },
          "metadata": {},
          "execution_count": 49
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "validation_target= Y[-N//2:]\n",
        "validation_predictions=[]\n",
        "\n",
        "i=-N//2\n",
        "\n",
        "while(len(validation_predictions)<len(validation_target)):\n",
        "  p=model.predict(X[i].reshape(1,-1,1))[0,0]\n",
        "  i+=1\n",
        "  validation_predictions.append(p)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rhX_qeIT95wn",
        "outputId": "008ec312-2f80-4c66-f922-2d60ce3e7c50"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 151ms/step\n",
            "1/1 [==============================] - 0s 17ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 24ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 28ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 24ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 53ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 24ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 25ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 26ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 27ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 26ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 24ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 24ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 28ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 29ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(validation_target,label='forecast_target')\n",
        "plt.plot(validation_predictions,label='forecast_predictions')\n",
        "plt.legend()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "7K8mmViU9-k4",
        "outputId": "0a02c3f8-e794-4b86-cc58-0b283551dae0"
      },
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f44f9a8cc10>"
            ]
          },
          "metadata": {},
          "execution_count": 51
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "validation_target= Y[-N//2:]\n",
        "validation_predictions=[]\n",
        "\n",
        "last_x=X[-N//2]\n",
        "\n",
        "while(len(validation_predictions)<len(validation_target)):\n",
        "  p=model.predict(last_x.reshape(1,-1,1))[0,0]\n",
        "  validation_predictions.append(p)\n",
        "\n",
        "  last_x=np.roll(last_x,-1)\n",
        "  last_x[-1]=p"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v_PijrMj-C2S",
        "outputId": "c1e87932-5796-48d3-f4de-688253f79a53"
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 24ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 30ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 26ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 25ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 25ms/step\n",
            "1/1 [==============================] - 0s 29ms/step\n",
            "1/1 [==============================] - 0s 25ms/step\n",
            "1/1 [==============================] - 0s 26ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 24ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 24ms/step\n",
            "1/1 [==============================] - 0s 22ms/step\n",
            "1/1 [==============================] - 0s 26ms/step\n",
            "1/1 [==============================] - 0s 28ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 21ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 20ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 18ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n",
            "1/1 [==============================] - 0s 27ms/step\n",
            "1/1 [==============================] - 0s 23ms/step\n",
            "1/1 [==============================] - 0s 19ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(validation_target,label='forecast_target')\n",
        "plt.plot(validation_predictions,label='forecast_predictions')\n",
        "plt.legend()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "1r8gvp4d-QEY",
        "outputId": "c327a80c-8bc9-4e7f-862f-7631724eb073"
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f44f1a7e280>"
            ]
          },
          "metadata": {},
          "execution_count": 53
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}