material/deep_learning/template_numeric_sequence_data_prediction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "template_numeric_sequence_data_prediction.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mZAFnuGfwqy-"
      },
      "source": [
        "# 숫자 순차열 데이터 예측 Template"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hb0EOx6OyNIw"
      },
      "source": [
        "# 데이터 준비"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VBfCpbwh0h9R"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B5AtaEi6yPhX"
      },
      "source": [
        "## 기반 숫자열 생성\n",
        "\n",
        "숫자열이 1개라도 2차원 배열에 답는다. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V6wKtOwFwqXM"
      },
      "source": [
        "numbers = np.zeros((2000,1))\n",
        "numbers[:,0] = np.sin(np.arange(0,1000,0.5))"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "k4hO2CThTFIV",
        "outputId": "1ee07347-d8a3-4158-a0e8-1bb008fc55c7"
      },
      "source": [
        "plt.plot(numbers[:100,0])\n",
        "plt.show()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6RAkHiJbxNfE"
      },
      "source": [
        "SEQUENCE_LENGTH = 5\n",
        "OFFSET = 1"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f-z-YNfNDBn5"
      },
      "source": [
        "## 입출력 데이터 생성\n",
        "\n",
        "12345678901234567890의 숫자열로 다음의 입출력 데이터를 생성한다.\n",
        "```\n",
        "12345 -> 6\n",
        "23456 -> 7\n",
        "34567 -> 8\n",
        "...\n",
        "56789 -> 0\n",
        "```"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yYxpmUQtwqTY"
      },
      "source": [
        "raw_x = []\n",
        "raw_y = []\n",
        "\n",
        "MAX_SAMPLE_COUNT = len(numbers)-(SEQUENCE_LENGTH+OFFSET)+1\n",
        "SAMPLING_COUNT = MAX_SAMPLE_COUNT\n",
        "for j in range(SAMPLING_COUNT):\n",
        "  i = np.random.randint(MAX_SAMPLE_COUNT)\n",
        "  raw_x.append(numbers[i:i+SEQUENCE_LENGTH,:])\n",
        "  raw_y.append(numbers[i+SEQUENCE_LENGTH+OFFSET-1,:])\n",
        "#              0         1              \n",
        "#              01234567890123456789\n",
        "# numbers[0] = 12345678901234567890\n",
        "#                 i = 3\n",
        "#                 <-x->   = [3:8] = 45678\n",
        "#                      y  = [8]   = 9\n",
        "\n"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "15RBiNi9wqQU",
        "outputId": "191e52fc-02f2-4aa0-c0cf-96e8260dd8a5"
      },
      "source": [
        "print(len(raw_x))\n",
        "# len(raw_x) = 1994 = len(numbers) - SEQUENCE_LENGTH - OFFSET = 2000 - 5 - 1"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1994\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "S7eFX6jIyrnx"
      },
      "source": [
        "## 데이터를 2차원 벡터화 하기\n",
        "\n",
        "RNN은 2차원 입력을 전제로 한다. 다중 숫자열은 그 자체로 2차원 데이터이다. 따로 해줄 것은 없다.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z-RCPdshTuFf",
        "outputId": "ea74a076-1155-49fb-d42f-be6e997c3a2d"
      },
      "source": [
        "print(raw_x[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[0.        ]\n",
            " [0.47942554]\n",
            " [0.84147098]\n",
            " [0.99749499]\n",
            " [0.90929743]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4mg54AkZ3boa",
        "outputId": "fed07712-f3ed-488d-80e2-dde75660c4e1"
      },
      "source": [
        "x = np.array(raw_x)\n",
        "y = np.array(raw_y)\n",
        "print(x.shape)\n",
        "print(y.shape)\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "train_x, test_x, train_y, test_y = train_test_split(x, y, test_size=0.1, shuffle=True)\n",
        "\n",
        "\n",
        "print(\"Training Data:\")\n",
        "print(train_x.shape)\n",
        "print(train_y.shape)\n",
        "\n",
        "print(\"Test Data:\")\n",
        "print(test_x.shape)\n",
        "print(test_y.shape)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(1994, 5, 1)\n",
            "(1994, 1)\n",
            "Training Data:\n",
            "(1794, 5, 1)\n",
            "(1794, 1)\n",
            "Test Data:\n",
            "(200, 5, 1)\n",
            "(200, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lMq2Om3w8qNy"
      },
      "source": [
        "준비된 데이터는 다음과 같습니다."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "41UcKyn-q5Ez",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 421
        },
        "outputId": "70260a1e-6e67-4b8d-f2d6-4bca09d44634"
      },
      "source": [
        "print(\"train_x[0]\\n\", train_x[0])\n",
        "print()\n",
        "print(\"train_y[0]\\n\", train_y[0])\n",
        "\n",
        "plt.plot(train_x[0], '.')\n",
        "plt.plot([len(train_x[0])], train_y[0], 'x')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "train_x[0]\n",
            " [[ 0.98360406]\n",
            " [ 0.94965409]\n",
            " [ 0.68319568]\n",
            " [ 0.24946714]\n",
            " [-0.24533966]]\n",
            "\n",
            "train_y[0]\n",
            " [-0.68007875]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# DNN 사용"
      ],
      "metadata": {
        "id": "jIlJNH5ssEB5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten, Input\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "# input_shape = (5,1)\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mae\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=1, validation_split=0.1, batch_size=128)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mae = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mae=\", mae)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x)\n",
        "print(test_y.shape)\n",
        "print(y_.shape)\n",
        "plt.scatter(test_y.squeeze(), y_)\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "WjOTs0jxq1lO",
        "outputId": "912e9936-a04b-4174-b264-a280ed2e3f40",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 860
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " flatten (Flatten)           (None, 5)                 0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 10)                60        \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 1)                 11        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 71\n",
            "Trainable params: 71\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 0s 3ms/step - loss: 0.0209 - mae: 0.1176\n",
            "loss= 0.020861124619841576\n",
            "mae= 0.11763329803943634\n",
            "(200, 1)\n",
            "(200, 1)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0-mHtIAk_qU9"
      },
      "source": [
        "# Template"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "k3CG02lDAI73"
      },
      "source": [
        "## LSTM"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 860
        },
        "id": "GEffW4RR3bgJ",
        "outputId": "19c48f80-ef5c-445d-f412-86ad3d088fd9"
      },
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten, Input\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "# input_shape = (5,1)\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(LSTM(128))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mae\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=1, validation_split=0.1, batch_size=128)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mae = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mae=\", mae)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x)\n",
        "plt.scatter(test_y, y_)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " lstm (LSTM)                 (None, 128)               66560     \n",
            "                                                                 \n",
            " flatten_1 (Flatten)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 10)                1290      \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 1)                 11        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 67,861\n",
            "Trainable params: 67,861\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de3zU9Z3v8ddnZjIJCSHcgoEEDBCQRlESAoKg9dJ2deuNVFdstfWC2tZLdXcfu909++juo+fRPae75/TUeqlaWlur1XpBS63ai4pWFCVcBLka7gRCAkhIgFzne/6YiSQYIGR+yW9m8n4+HvOYmd/88vl9HtPynq+/+c33a845REQk+QX8bkBERLyhQBcRSREKdBGRFKFAFxFJEQp0EZEUEfLrwMOHD3eFhYV+HV5EJCktW7Zsr3Mut6vXfAv0wsJCKioq/Dq8iEhSMrNtx3tNp1xERFKEAl1EJEUo0EVEUoQCXUQkRSjQRURShAJdRCRFKNBFRFJE0gX6R1V1/PC19WjaXxGRzpIu0Jdt+4SfLtrEks37/W5FRCShJF2gXzdtNCOy0/nJ6x/73YqISEJJukDPSAty+wXjeG/zPiq2apQuItIu6QId4Gvnns6wrDA/eaPS71ZERBJGUgb6gHCQeeeP4+2NtazcccDvdkREEkJSBjrAjTNPZ3BmGg/oXLqICJDEgT4wPcSts8by+voaPqqq87sdERHfJW2gA3xjViHZGSEe1Ll0EZHkDvRBGWncPGssr62pZkN1vd/tiIj4KqkDHeCWWYVkhYM88IbOpYtI/5b0gT44M8zXzyvkD6t3U1nT4Hc7IiK+SfpAB5g3eywZoSAPv6lz6SLSf6VEoA8bmM4NM8bwuw93sXXvIb/bERHxRUoEOsBtF4wjFDAeXqRRuoj0TykT6COyM7h++hgWLK9ix/7DfrcjItLnUibQAe74/DgCZjzy1ia/WxER6XMpFegjcwZwbVkBz1XsZHfdEb/bERHpUykV6ADfunA8Eed49K3NfrciItKnUi7QC4ZkUl6az9MfbKemvtHvdkRE+kzKBTrAty8soqUtws/e1ihdRPqPlAz0wuFZXD0lnyeXbGdfQ5Pf7YiI9ImUDHSAb19URGNrG/Pf2eJ3KyIifSJlA71oxEC+PHkkT7y7lQOHm/1uR0Sk16VsoAPcdXERh5rb+MXirX63IiLS61I60CflDeLSM/N4fPEWDja2+N2OiEivSulAh+govb6xlV9plC4iKS7lA/2s/BwumTSCny/eQkNTq9/tiIj0mm4FupldamYbzKzSzL57gv2+YmbOzMq8azF+d18ygQOHW3hyyTa/WxER6TUnDXQzCwIPAZcBxcD1ZlbcxX7ZwHeA971uMl5TRg/mgom5/OztzRxu1ihdRFJTd0bo04FK59xm51wz8AxwVRf7/U/gh0BC/t7+nouL2Heomd+8v93vVkREekV3Aj0f2NHh+c7Ytk+ZWSkw2jn3hxMVMrPbzazCzCpqa2tPudl4lBUOZea4YTz29mYaW9r69NgiIn0h7i9FzSwA/Aj4h5Pt65x7zDlX5pwry83NjffQp+zuS4qoqW/i2YodJ99ZRCTJdCfQq4DRHZ4XxLa1ywbOAhaZ2VZgBrAw0b4YBZg5bhjTCofw00WbaGrVKF1EUkt3An0pMMHMxppZGJgLLGx/0TlX55wb7pwrdM4VAkuAK51zFb3ScRzMjLsvnsDuukZeWFZ18j8QEUkiJw1051wrcBfwR2Ad8Kxzbo2Zfd/MruztBr12/oThnDN6MA8vqqSlLeJ3OyIinunWOXTn3CvOuYnOufHOuR/Etn3PObewi30vTMTReTsz456Li9j5yRFeWqFRuoikjpT/pWhXLp40gjNHDeKhNytp1ShdRFJEvwz09nPpW/cd5uVVu/1uR0TEE/0y0AG+VHwaZ5yWzYNvVhKJOL/bERGJW78N9EDAuOviIiprGnj1o2q/2xERiVu/DXSAv508kvG5WTzwxscapYtI0ku+QN/wKjx9PUTi/zIzGBulr6+u58/r9njQnIiIf5Iv0JsPwYZXYPMbnpS74uxRnD4skwfe+BjnNEoXkeSVfIH+uSsgKxeW/tyTcqFggDsvLOKjqoMs2tC3E4aJiHgp+QI9lA6lX4eNr8EBb6bCnVOaT/7gAdz/ukbpIpK8ki/QAabeHL2veNyTcmnBAN++aDwrdxzgncq9ntQUEelryRnog0fDxMtg+RPQ2uRJyWumFjAyJ4MHXq/0pJ6ISF9LzkAHmHYrHN4Laz8znUyPpIeC3HHBOD7Yup8lm/d5UlNEpC8lb6CPuwiGjoOlP/Os5NzpY8jNTucnr3/sWU0Rkb6SvIEeCMC0ebDjfdi9ypOSGWnRUfq7m/axbNt+T2qKiPSV5A10gClfhdAAqPDmEkaAr547hqFZYX6ic+kikmSSO9AHDIHJX4FVz0JjnSclM8Mh5p0/lrc21vLhjgOe1BQR6QvJHegQPe3SchhWPu1Zya/PLGRwZhoPvKFz6SKSPJI/0EeVQH4ZLJ0PHv0oaGB6iFtmjeUv62pYs8ubkb+ISG9L/kCH6Ch938ew5W3PSn7jvEKy00M8+IbOpYtIckiNQD9zTvR8+tL5npXMGZDGTbMKefWjajZU13tWV0Skt6RGoKdlQMmNsP4PUOfdws+3zBpLVjjIg29qlC4iiS81Ah2g7BZwEVj+K89KDskKc+PMQl5etYtNtQ2e1RUR6Q2pE+hDx8KEL8KyX0Jbi2dl550/lvRQgIc0SheRBJc6gQ7RL0cb9sD6lz0rOXxgOjecezq/W7mLbfsOeVZXRMRrqRXoRV+AwWPgA+++HAW4/YJxBAPGw29u8rSuiIiXUivQA0EouxW2vQM16zwrO2JQBtdPG80Ly3ey85PDntUVEfFSagU6RK92CaZ7tkRduzs+Px4zeOQtjdJFJDGlXqBnDYtel/7hM9Dk3fXjowYP4Nqy0Ty7dCfVdY2e1RUR8UrqBTpEvxxtrodVv/W07Lc+P56Icxqli0hCSs1ALyiDkedET7t4uOjz6KGZzCnJ5+kPtlNTr1G6iCSW1Ax0s+govWYtbH/P09J3XlRES1uE+X/d4mldEZF4pWagA5x1DWTkeDq/C0Dh8CyumpLPr9/bxr4GbxaoFhHxQuoGejgTptwQXUS6fo+npe+8qIjG1jZ+/o5G6SKSOLoV6GZ2qZltMLNKM/tuF69/08xWm9lKM3vHzIq9b7UHym6BSAssf8LTskUjBnLZWXk8uWQbTa1tntYWEempkwa6mQWBh4DLgGLg+i4C+zfOucnOuSnAfwE/8rzTnhheBOMugmWPQ1urp6XnThvDwcZW3lhX42ldEZGe6s4IfTpQ6Zzb7JxrBp4Bruq4g3PuYIenWYB3l5bEa9o8OFgFG1/ztOysouGMyE7nheXeTdcrIhKP7gR6PrCjw/OdsW2dmNmdZraJ6Aj9nq4KmdntZlZhZhW1tbU96ffUTbwUBhXA0p95WjYYMOaU5LNoQ42+HBWRhODZl6LOuYecc+OBfwb+7Tj7POacK3POleXm5np16BMLhqDsJti8CPZ6u+jznNJ8WiOOl1ft9rSuiEhPdCfQq4DRHZ4XxLYdzzPA1fE05bmSr0MgDSp+4WnZSXmDKB45iAXLd3paV0SkJ7oT6EuBCWY21szCwFxgYccdzGxCh6dfBrwdCscr+zQovhJWPAXN3s5pXl6az4c766is0bqjIuKvkwa6c64VuAv4I7AOeNY5t8bMvm9mV8Z2u8vM1pjZSuDvgW/0Wsc9Ne02aKqDj17wtOyVU0YRDBgL9OWoiPjMnIdznZyKsrIyV1FR0XcHdA5+Ois6Z/odb0enB/DIzY9/wIbqet7554sJBLyrKyJyLDNb5pwr6+q11P2l6LHMYNqtUL0Kdnr7QTKntIBddY0s2bLP07oiIqei/wQ6wNl/B+Fsz+d3+VLxaWSnh3TaRUR81b8CPT0bplwPaxbAIe9G0xlpQf528kheXb2bw83e/iJVRKS7+legQ3TN0bZmWOHt/C7lpfkcam7jT2u8nQhMRKS7+l+gj5gEhedHr0mPeDex1rTCoeQPHsCCFTrtIiL+6H+BDtEvRw9sh8q/eFYyEDDKS/N55+Na9hzUakYi0vf6Z6BPuhwG5nn+5eicknwiDn63UqN0Eel7/TPQg2kw9Sb4+M+w37tFKsblDqRkzGBeWFaFX9f3i0j/1T8DHWDqN8ACns/vUl5awIY99azdffDkO4uIeKj/BvqgUTDpy7Di19ByxLOyl08eSVrQeFHXpItIH+u/gQ4w/TY48gmsecmzkkOywlw8aQQvrdxFa1vEs7oiIifTvwO98HwYPtHzxS/KSwvY29DEXyv3elpXRORE+negm0WXqKtaBlXLPSt70RkjGJyZpqkARKRP9e9ABzhnLqRlQcXPPSsZDgW44uxR/GlNNfWNLZ7VFRE5EQV6Rk500q7Vz0fPp3ukvDSfptYIr66u9qymiMiJKNAh+svR1sboikYemTJ6MOOGZ/GClqcTkT6iQAfImwyjZ0RPu0S8uTLFLDoVwPtb9rNj/2FPaoqInIgCvd20ebB/M2x+07OSV5fkA/CSJuwSkT6gQG9XfCVk5cJS774cLRiSybljh/LiCk0FICK9T4HeLpQOpV+Hja/CgR2elf1KaQGb9x5i5Y4DntUUEemKAr2jqTdF75c97lnJyybnkR4K6Jp0Eel1CvSOBo+BiZfC8iegtcmTktkZafzNmXn8ftUumlq9W1BDRORYCvRjTZsHh2ph3e89K1lems+Bwy28ub7Ws5oiIsdSoB9r3EUwdBx84N38LrOLhjN8YDovrtA16SLSexToxwoEogtJ71gC1as9KRkKBrh6yijeWF/DJ4eaPakpInIsBXpXpnwVQhmeXsJYXlpAS5vj5VW7PKspItKRAr0rmUNh8jWw6llorPOkZPGoQUzKy+YFXe0iIr1EgX480+ZByyH48BnPSpaX5rNyxwE21zZ4VlNEpJ0C/XhGlUD+VFg6Hzz6ledVU/IJGLyoqQBEpBco0E9k2jzYuxG2/tWTcqcNymD2hFwWLK8iEtFUACLiLQX6iZxZDgOGREfpHvlKaT5VB47wwdb9ntUUEQEF+omlZUDJjbDuZTjozdUpXyrOIyscZIHmSRcRjynQT6bsZnARWPYrT8oNCAe5bPJIXlldTWOLpgIQEe90K9DN7FIz22BmlWb23S5e/3szW2tmq8zsdTM73ftWfTJ0HBR9AZb9Etq8WR+0vDSfhqZW/rR2jyf1RESgG4FuZkHgIeAyoBi43syKj9ltBVDmnDsbeB74L68b9dX026ChGtb/wZNyM8YOY1ROhk67iIinujNCnw5UOuc2O+eagWeAqzru4Jx70znXvs7aEqDA2zZ9VvSF6EyMHn05GggYc0rzeXtjLTX1jZ7UFBHpTqDnAx1XfNgZ23Y8twKvdvWCmd1uZhVmVlFbm0QzDwaCUHZL9PLFmvWelJxTUkDEwcKVmgpARLzh6ZeiZnYDUAb8d1evO+cec86VOefKcnNzvTx07yu5EYLh6ELSHigaMZBzCnK08IWIeKY7gV4FjO7wvCC2rRMz+wLwP4ArnXPerA6RSLKGR69LX/k0NHnz0/3y0gLW7j7Iut0HPaknIv1bdwJ9KTDBzMaaWRiYCyzsuIOZlQCPEg3zGu/bTBDT5kFzPax+1pNyV5wzilDANBWAiHjipIHunGsF7gL+CKwDnnXOrTGz75vZlbHd/hsYCDxnZivNbOFxyiW3gjLIOxs+8GZ+l6FZYS6aNIKXVlTRpqkARCRO3TqH7px7xTk30Tk33jn3g9i27znnFsYef8E5d5pzbkrsduWJKyYps+govWYNbF/iScnyknxq6ptYXLnXk3oi0n/pl6KnavK1kJ7j2SWMF39uBIMyQromXUTipkA/VeFMKPkarP0dNMT/dUF6KMgV54zitTXVNDS1etCgiPRXCvSeKLsFIi2w3Jv5XcpLC2hsifDq6t2e1BOR/kmB3hPDJ8C4C2HpL6A1/kWfS8cMpnBYpq5JF5G4KNB7aubdUL8LVj8XdykzY05JAUu27KPqwBEPmhOR/kiB3lNFl8Bpk2HxjyESibvcnJJ8nIOXdE26iPSQAr2nzGD2vdEl6ja8Ene5McMymV44lAXLd+I8WsNURPoXBXo8iq+GIYXwzv/z5IdG5aX5bKo9xKqddfH3JiL9jgI9HsEQnHcPVFXAtsVxl7ts8kjCoYCmAhCRHlGgx2vKVyErNzpKj1POgDS+WHwaCz/cRXNr/OflRaR/UaDHK20AzPgWVP4Fdq+Ku9xXSvPZf6iZtzYm0XzxIpIQFOheKLsVwtnRK17idP6EXIYPDGsqABE5ZQp0LwwYDNNugTUvwv7NcZVKCwa44pxRvL6uhrrD3ixKLSL9gwLdKzO+DYEQvPtg3KW+UlpAc1uEl1dreToR6T4Fuley8+Cc62HFk3FP2nXmqEFMPG2gpgIQkVOiQPfSrO9AWzMs+WlcZcyM8tIClm37hK17D3nUnIikOgW6l4aNh+KronOlN8b346Crp+RjBgt0TbqIdJMC3Wuz74Wmg1DxeFxl8nIymDV+OC+u0FQAItI9CnSvjSqBcRfBkoehpTGuUuWl+ezYf4SKbZ941JyIpDIFem+YfR807IFVz8RV5m/OzCMzHNQ16SLSLQr03jD2guhIffH9EGnrcZms9BCXnpXHy6t209jS8zoi0j8o0HuDWXSUvn8zrFsYV6nykgLqG1v5y7o9HjUnIqlKgd5bJl0Ow4rinlp35vhh5A3K4EVdky4iJ6FA7y2BYPS69N0fwuY3e1wmGDCuLsln0cZa9jY0edigiKQaBXpvOvs6yB4J78Q3aVd5aT5tEcfClZoKQESOT4Hem0Lp0TletrwFVct6XGbiadlMzs9hwQpd7SIix6dA721Tb4KMnLhH6XNK8vmo6iAb99R705eIpBwFem/LGATTboN1v4e9H/e4zJVTRhEMmCbsEpHjUqD3hXO/GT39svj+HpcYPjCdCyfm8tKKKtoimgpARD5Lgd4XBuZCyY3w4TNwsOdfbJaXFlB9sJH3Nu3zsDkRSRUK9L5y3l3gItE5Xnroks+NIDsjpKkARKRLCvS+MqQQziqPzsJ4pGeTbWWkBbn87JG8tqaaQ02t3vYnIklPgd6XZt0LzQ3R+dJ7qLy0gMPNbfxxTbWHjYlIKuhWoJvZpWa2wcwqzey7Xbx+gZktN7NWM7vG+zZTRN5ZMOFLsOQRaD7coxJlpw9h9NAButpFRD7jpIFuZkHgIeAyoBi43syKj9ltO3AT8BuvG0w5s++Dw3th5VM9+nMzo7ykgMWb9rK77ojHzYlIMuvOCH06UOmc2+ycawaeAa7quINzbqtzbhUQ6YUeU8uYmTD6XHj3J9DWs/Pgc0rycQ5eWqGpAETkqO4Eej6wo8PznbFtp8zMbjezCjOrqK2t7UmJ5Nc+te6B7bDmxR6VKByexdTTh7BguZanE5Gj+vRLUefcY865MudcWW5ubl8eOrFM+BvInRTX1Lrlpfl8XNPAml0HPW5ORJJVdwK9Chjd4XlBbJv0VCAQveKlZg18/Ocelbh88ijCwQAv6Jp0EYnpTqAvBSaY2VgzCwNzgfiW4RGYfA0MKoiO0nsgJzONSz43goUrd9HSpq8uRKQbge6cawXuAv4IrAOedc6tMbPvm9mVAGY2zcx2AtcCj5rZmt5sOiUE0+C8u2H7u7D9/R6VKC8tYN+hZt7e2E+/jxCRTrp1Dt0594pzbqJzbrxz7gexbd9zzi2MPV7qnCtwzmU554Y5587szaZTRumNMGAoLO7Z1Lqfn5jL0Kww97/+MXVHWjxuTkSSjX4p6qdwFpx7B2x4BWrWnfqfhwL855zJrNt9kOsfW6Il6kT6OQW636bfDmmZPZ5a99Kz8vjZ18vYvLeBv3v0PXYd0I+NRPorBbrfModGVzVa/Vz02vQeuPCMETxxy7nUHmzi2kfeY8veQ972KCJJQYGeCGbeGb1/76Eel5g+dihP3z6DIy1tXPvIe6yv1vXpIv2NAj0R5BTA2dfB8ifgUM8XrzgrP4dn75hBMADXPbqEFdt7Nk2viCQnBXqimPUdaDkMHzwWV5miEdk8/83zyBmQxtfmv8+7m/Z61KCIJDoFeqLIPQPO+DJ88Cg0NcRVavTQTJ775kwKhgzgpseX8pe1ezxqUkQSmQI9kcy+L7qa0fIn4i512qAMfnv7TCblZXPHk8v43UrN1iCS6hToiWT0NDh9Nrz3ILQ2x11uSFaYp+adS9npQ7j3tyt5csk2D5oUkUSlQE80s++Dg1Xw0fOelMvOSONXt0znojNG8G8vfcRPF23ypK6IJB4FeqIpugROmwzv/Bgi3ky6lZEW5JEbpnL52SP54Wvr+a/X1msedZEUpEBPNGYw+17YuwE2vupZ2XAowP1zS5g7bTQPL9rEvy9cQySiUBdJJQr0RFR8NQw+Hf76ox4vgNGVYMD4X+WTue38sTzx3jb+8bkPadXUuyIpQ4GeiIIhmHUPVFXAtsWeljYz/vVvP8c/fHEiC1ZU8e2nltPU2ubpMUTEHwr0RDXla5CV2+MFME7EzLj7kgn8+xXF/GntHm79ZQWHm3u2YLWIJA4FeqJKGwAzvgWVf4Hq1b1yiJtnjeW/rzmbdzft5Yb572tOdZEkp0BPZGW3Qjg7esVLL7m2bDQPfbWU1VV1zNWc6iJJTYGeyAYMhrKbYc0C2L+l1w5z2eSRzP/GNLbsbeDvHtGc6iLJSoGe6GZ8GwIhePeBXj3M5yfm8utbz6W2XnOqiyQrBXqiGzQSzrkeVjwJDTW9eqhphZ3nVF+3W3OqiyQTBXoymPUdaGuG9x/p9UO1z6keChjXPfoeyzWnukjSUKAng2Hjofgq+GA+NPb+qLloRDbPfXMmQ7LC3DD/fd6t1JzqIslAgZ4sZt8LTXWw7PE+OdzooZk8d0dsTvVfLuXPmlNdJOEp0JPFqBIYd2F03dGWxj455IjYnOqfy8vmm5pTXSThKdCTyez7oGEPrHqmzw45JCvMU7fN0JzqIklAgZ5Mxn4+OlJffD9E+m7+lYHpIc2pLpIEFOjJxCw6St+/Gdb9vk8PnZEW5NEbp3LFOaM0p7pIggr53YCcokmXw7Ci6KRdxVdFQ76PpAUD/Pi6KQxMD/Lwok00NLXyH1ecSSDQdz2IyPFphJ5sAkE47x7YvRI2L+rzwwcDxn/OmcztF4zTnOoiCUaBnozOmQsD83plat3uMDP+5bJJ/OOXNKe6SCJRoCejUDrMvBO2vAVVy3xpwcy46+IJ/EeHOdU37qmnvlFT8Ir4xfz6YqusrMxVVFT4cuyU0HgQfnxW9MqX637tayvPL9vJPz3/Ie1LlA5MD5GXk8HI2C0vZ0DsPoNROQPIy8lgUEYI68Pz/yKpwsyWOefKunpNX4omq4xBMO02+Ov/hb2VMLzIt1aumVrAlNE5rN1dT3XdEXYdaKS6rpHdBxvZuKeWmvqmzyyNmhkOdgr49sCPfghEPwByBqQp9EVOQbdG6GZ2KXA/EATmO+f+9zGvpwNPAFOBfcB1zrmtJ6qpEboHGmqjo/RACMIDIRSGYPstDYLpRx+H0mPbwp+9hY7zN8HwMX/X4XFXxwoEAQMLxG4GZrREoLahhT31TVQfbKL6YDO76pqprm9kV10zu+uaqK5vIuI6h3dGWuDTcM/rMNof9enzAQzJVOhL/xLXCN3MgsBDwBeBncBSM1vonFvbYbdbgU+cc0VmNhf4IXBd/K3LCQ3MhTmPwNbF0NYEbS3RWRlbOzxua4HmBjiyP/q402sdHzf3WptpwKjY7bjSo3fOAjgsdgsQOQSRBiNSZbS56PYIR+/3EYheNmkBAoEgFgjgLERbII1IIEyk4+NAWux29LGLPXeBNCLB6L0LhHHBNFwwfPRxIPrh9enjUBgCYQil4Tp8CFowjAuFsUAaFggSMINAkIDxaX9mAQKBAMFAtHczi75uRsAMMwgEOm/r9HqATtssdh+MPdYHXP/VnVMu04FK59xmADN7BrgK6BjoVwH/EXv8PPCgmZnTL09635lzord4Odc53D+9tX8INHf9QfDpB0QTuEjs5mK3COCO2X7sNo4+xmEugnXc99P9HZFIG43NrRxqauFwUzOHm1o50tzCkfb75laaG1sIulZCtBKO3dJoJc0aOz+nlbC1RO9pJY02wrQQsr6/BPPYDypH188dECHQ6d4d8zwSu86h/b79b9udMOqty4ef1jnB7sct3nFTVzVO5Rgn+/voPifh0WddvGVqp95H2ZfnedJLR90J9HxgR4fnO4Fzj7ePc67VzOqAYUCneVfN7HbgdoAxY8b0sGXpFWbRUWco7HcnxxUAMmO3k3HO4RxEnCMSuz/6PLYtEn3c7KCpfVtbK5G2ZmhtxrU141qbcbEPNNe+vfWzH3aurQXr8EFnbc04HEQiOBwucvRDzbkIrtPj6AeZi32IuU8/zNyn+x3d3vlDseO+nT8sHe0fhO0x92nYOTo/b398zA5Hn3aOyRPV6VzrM0fo4n+ok7x+CuzYvj9zrMQZX6ZnD+2Vun36pahz7jHgMYieQ+/LY0v/Yu2nIrwakokkge5ch14FjO7wvCC2rct9zCwE5BD9clRERPpIdwJ9KTDBzMaaWRiYCyw8Zp+FwDdij68B3tD5cxGRvnXSUy6xc+J3AX8ketniL5xza8zs+0CFc24h8HPg12ZWCewnGvoiItKHunUO3Tn3CvDKMdu+1+FxI3Ctt62JiMip0FwuIiIpQoEuIpIiFOgiIilCgS4ikiJ8mz7XzGqBni4hP5xjfoXaz+n96Ezvx1F6LzpLhffjdOdcblcv+Bbo8TCziuPNNtYf6f3oTO/HUXovOkv190OnXEREUoQCXUQkRSRroD/mdwMJRu9HZ3o/jtJ70VlKvx9JeQ5dREQ+K1lH6CIicgwFuohIiki6QDezS81sg5lVmtl3/e7HL2Y22szeNLO1ZrbGzL7jd0+JwMyCZrbCzF72uxe/mdlgM3vezNab2Tozm+l3T34xs/ti/04+MrOnzSzD7556Q1IFeocFqy8DioHrzRWIxdUAAAH9SURBVKzY36580wr8g3OuGJgB3NmP34uOvgOs87uJBHE/8JpzbhJwDv30fTGzfOAeoMw5dxbRacBTcorvpAp0OixY7ZxrBtoXrO53nHO7nXPLY4/rif5jzfe3K3+ZWQHwZWC+3734zcxygAuIrlWAc67ZOXfA3658FQIGxFZUywR2+dxPr0i2QO9qwep+HWIAZlYIlADv+9uJ734M/BMQ8buRBDAWqAUej52Cmm9mWX435QfnXBXwf4DtwG6gzjn3J3+76h3JFuhyDDMbCLwA3OucO+h3P34xs8uBGufcMr97SRAhoBT4qXOuBDgE9MvvnMxsCNH/kh8LjAKyzOwGf7vqHckW6N1ZsLrfMLM0omH+lHNugd/9+GwWcKWZbSV6Ku5iM3vS35Z8tRPY6Zxr/6+254kGfH/0BWCLc67WOdcCLADO87mnXpFsgd6dBav7BTMzoudH1znnfuR3P35zzv2Lc67AOVdI9P8XbzjnUnIU1h3OuWpgh5mdEdt0CbDWx5b8tB2YYWaZsX83l5CiXxB3a03RRHG8Bat9bssvs4AbgdVmtjK27V9j67+KANwNPBUb/GwGbva5H1845943s+eB5USvDltBik4BoJ/+i4ikiGQ75SIiIsehQBcRSREKdBGRFKFAFxFJEQp0EZEUoUAXEUkRCnQRkRTx/wEj7TcWHsiOlQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 0s 4ms/step - loss: 4.3619e-05 - mae: 0.0057\n",
            "loss= 4.3618994823191315e-05\n",
            "mae= 0.005686040036380291\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAD4CAYAAADhNOGaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dfZBUd53v8feXyUAm2esOJBgDYQKJbKIpE+L2DXFTtWrMk1YtoMZADLtkN5Gr1+yWRrnChTIPa1aUUuLWplYxRqPmAhGTdlJG5+aJtSo3ZBlqgAnJIgQ10MHABtB7A/IwfO8f5zSeNN0950yffv68qqam+3d+p+fLmaG/fX6P5u6IiEj7GlXvAEREpL6UCERE2pwSgYhIm1MiEBFpc0oEIiJt7pR6BzASZ555pk+ePLneYYiINJUNGzb8p7uPLyxvykQwefJk+vv76x2GiEhTMbPfFCtX05CISJtTIhARaXNKBCIibU6JQESkzSkRiIi0uaYcNSQi0k4uvuPn/P7w0InnbxnTwea7rkvt9VO5IzCzB8xsj5m9UOK4mdk/m9l2M9tsZu+OHJtnZtvCr3lpxCMi0iouXPz4m5IAwO8PD3HxHT9P7Wek1TT0PaBcevogMDX8mg/8K4CZjQPuAKYDlwF3mNnYlGISEWla2YEcF33x5/xhqPhWAYXJoRKpNA25+y/MbHKZKjOB73uw+cE6M+s2s7OB9wFPuPs+ADN7giChrEwjLhGRZnTTt5/j2Zf31ezn1aqzeCKwM/J8V1hWqvwkZjbfzPrNrH/v3r1VC1REpJ5qnQSgiUYNufsKd8+4e2b8+JOWyhARaXr1SAJQu1FDOWBS5Pk5YVmOoHkoWr62RjGJiDSE7ECO/7FmE0dK9AcUc+/saan9/Folgl7gNjNbRdAx/Dt3321mfcA/RTqIrwEW1SgmEZG6u/rra9m2543Y9bs6R/Hlj1zMrEuLtqKPSCqJwMxWEnyyP9PMdhGMBOoEcPdvAo8DHwK2AweBvw2P7TOzfwTWhy91d77jWESk1U2/5wle+79HYtefe3kPX5r1rtTjSGvU0I3DHHfg0yWOPQA8kEYcIiLNIDuQY/Gjg7xxJN4Q0NEdxlevvyTVu4AozSwWEamh7ECORY8McuhovCRwxfnjeOgT76lqTEoEIiI1kB3IsaxvK7kDh2KfU4skAEoEIiJVtSQ7yA/XvZLonA6Dr90wrWpNQYWUCEREqmQkSeCs/zKa5xdfXaWIimuaCWUiIs0kO5BLlASMYFRQrZMA6I5ARCRV2YEcdz22hf0Hj8aqP7G7iwXXXlCzZqBilAhERFKSZIkIA5bPrl0/QDlqGhIRScGS7GCidYJuurynIZIA6I5ARKRiSfoDqrFERKWUCERERiBpX0BXZwdf/si7GioB5CkRiIgklB3IsWDNJo7GXC20u6uTO2dc1JBJAJQIREQSW9a3NXYSqNXs4EooEYiIxJBfIuLVA4eIu2tAtVYLTZsSgYhIGdmBHHf2buHAoXh9AdDY/QHFKBGIiJSQdKVQgLGndXLHXzVuf0AxaW1Mcx3wDaADuN/dlxYcXw68P3x6GvBWd+8Ojw0Bg+GxV9x9RhoxiYhUIjuQ43MPb2LIW6NDuJyKE4GZdQD3AVcDu4D1Ztbr7i/m67j7ZyP1/x64NPISh9w9vc03RUQqkGRY6MTuLp5deGUNoqquNGYWXwZsd/cd7n4EWAXMLFP/RmBlCj9XRCRV+aagOEmgq7ODBddeUIOoqi+NpqGJwM7I810EG9SfxMzOBaYAT0eKTzWzfuAYsNTdsyXOnQ/MB+jp6UkhbBGRwJLsICuf3xm7GagZ+wHKqXVn8RxgjbtHe17OdfecmZ0HPG1mg+7+cuGJ7r4CWAGQyWTijt4SESkryZ4BHWZ87Ybq7R1cL2kkghwwKfL8nLCsmDkUbGLv7rnw+w4zW0vQf3BSIhARSctI5gQ025DQJNJIBOuBqWY2hSABzAE+XljJzC4ExgLPRcrGAgfd/bCZnQlcAXw1hZhERIpakh3koXWvxE4A0NwjguKoOBG4+zEzuw3oIxg++oC7bzGzu4F+d+8Nq84BVrm/qRHuHcC3zOw4Qcf10uhoIxGRNGUHcomSQCNsGlML5jE7RxpJJpPx/v7+eochIk0k6byAZlkeIgkz2+DumcJyzSwWkZaXHxYaJwl0mHHj9EktlwTKUSIQkZYT7Qye0N3FwSPHyi4T0UjbRtaDEoGItJTCzuDcgUNl6xuNtW1kPSgRiEjLSDInAFp3XkBS2rxeRFpCkn2DIZgXoCQQ0B2BiDS9/Iigcrq7Ojl9zCkn+g3aYVhoXEoEItLU4o4IauUJYZVSIhCRplI4IuiNw+VHBAGcPrpDSaAMJQIRaRqFO4YNNyIIYJTBPR9unzkBI6FEICJNIenMYGj9NYLSokQgIg0vycxgaO2VQqtBiUBEGla+P2C4JqCxp3Vy2miNCBopJQIRaUiF/QGldHV2tNRuYfWgCWUi0pCW9W0dNgl0mKkJKAVKBCLSkF4dpjlIM4PTo6YhEambwjkB0bb9Cd1dJfsG2mXDmFpJ5Y7AzK4zs61mtt3MFhY5frOZ7TWzjeHXrZFj88xsW/g1L414RKTx5fsAcuG+wbkDh1j0yCDZgWDL8wXXXkBXZ8ebzunq7ODe2dN4duGVSgIpqviOwMw6gPuAq4FdwHoz6y2y5eRqd7+t4NxxwB1ABnBgQ3ju/krjEpHGVWpOwKGjQyzr28qsSyeeeKMvdccg6UmjaegyYLu77wAws1XATCDO3sPXAk+4+77w3CeA64CVKcQlIg0mO5Djzt4tHDh0tGSdaN9ANCFI9aTRNDQR2Bl5vissK/RRM9tsZmvMbFLCczGz+WbWb2b9e/fuTSFsEamlfFNQuSQAQd+A1FatRg09Bkx294uBJ4AHk76Au69w94y7Z8aPH596gCJSPUuyg3xm9cZYcwIWXHtBjaKSvDSahnLApMjzc8KyE9z99cjT+4GvRs59X8G5a1OISUQaQJIdwzQnoH7SuCNYD0w1sylmNhqYA/RGK5jZ2ZGnM4CXwsd9wDVmNtbMxgLXhGUi0uSSJAHNCaiviu8I3P2Ymd1G8AbeATzg7lvM7G6g3917gX8wsxnAMWAfcHN47j4z+0eCZAJwd77jWESaU9z1gfLGntapJSLqzDzBkq6NIpPJeH9/f73DEJGIOCOCCt07e5oSQA2Z2QZ3zxSWa2axiFQs7gJxUXMv71ESaBBKBCIyYkmbgfKuOH8cX5qlXcMahRKBiIzISO4CTh/dwT0f1sigRqNEICKJJL0L0G5hjU+JQERiS3oXoBFBzUGJQESGlfQuQMtENxclAhEpK8ldgJqBmpMSgYicZEl2kJXP7zxpmehydBfQvJQIRORNkiwNAboLaAVKBCJyQtIkoLuA1qBEICJA8kXidBfQOpQIRNpYdPP4OL0BBtoysgUpEYi0qaRzAuZe3qNlIVqUEoFImxnJ+kBKAq1NiUCkjYx0lVAlgdaWSiIws+uAbxBsTHO/uy8tOH47cCvBxjR7gb9z99+Ex4aAwbDqK+4+I42YROSPsgM57npsC/sPxt8roMOMG6dPUhJoAxUnAjPrAO4DrgZ2AevNrNfdX4xUGwAy7n7QzD5FsGfx7PDYIXefVmkcInKy7ECOxY8O8saReHcAGg3UntLYs/gyYLu773D3I8AqYGa0grs/4+4Hw6frCDapF5Eqyg7kWLBmU+wkMLG7S0mgTaXRNDQR2Bl5vguYXqb+LcDPIs9PNbN+gmajpe6eTSEmkbamOQGSRE07i81sLpAB3hspPtfdc2Z2HvC0mQ26+8tFzp0PzAfo6empSbwizShJEugwUxKQVBJBDpgUeX5OWPYmZnYVsBh4r7sfzpe7ey78vsPM1gKXAiclAndfAayAYPP6FOIWaRkjWSQO4Gs3XKIkIKn0EawHpprZFDMbDcwBeqMVzOxS4FvADHffEykfa2ZjwsdnAlcA0U5mERlG/g4gaRLQ5vGSV/EdgbsfM7PbgD6C4aMPuPsWM7sb6Hf3XmAZ8CfAj8wM/jhM9B3At8zsOEFSWlow2khEiki6NESUdg2TQuYJP0U0gkwm4/39/fUOQ6QuRjIpDDQxTMDMNrh7prBcM4tFmkh2IMfnHt6UqBlIE8NkOEoEIk0ifycQNwnoDkDiUiIQaRLL+rbGag7SHYAkpUQg0sDidgprUphUQolApIFEl4g2iDUiSJPCpFJKBCINIjuQY8GPNnH0ePD2HycJ6E5A0qBEINIg7uzdciIJDEdbRkqalAhEGsSBQ/H2CpjY3cWzC6+scjTSTpQIROpkJJvFdHV2sODaC6oYlbQjJQKROsjvFXB0KP7EMC0NIdWiRCBSQyNZJbS7q5M7ZygBSPUoEYjUQDAreDOHjh6PVV+dwVJLSgQiVfbHReLiJQF1BkutKRGIVFHSReI6R5k6g6XmlAhEqiA7kOPO3i2xh4SC+gKkfpQIRFK2JDvIQ+teib1hjFYJlXpLJRGY2XXANwh2KLvf3ZcWHB8DfB/4c+B1YLa7/zo8tgi4BRgC/sHd+9KISaSWomsExTXK4OPTlQSk/ipOBGbWAdwHXA3sAtabWW/BlpO3APvd/e1mNgf4CjDbzN5JsMfxRcAE4Ekz+zN3T7b1kkgdJd0xrMNMm8ZLQ0lj8/rLgO3uvsPdjwCrgJkFdWYCD4aP1wAfsGDz4pnAKnc/7O6/AraHryfSNOLuEwDBzGAlAWk0aTQNTQR2Rp7vAqaXqhNudv874IywfF3BuUX/h5jZfGA+QE9PTwphi4xMdI+ACd1dsZuDNDNYGlXTdBa7+wpgBQSb19c5HGlThUtDxEkCBtykDmFpYGkkghwwKfL8nLCsWJ1dZnYK8KcEncZxzhWpu6SdwflNZSZqdrA0gTQSwXpgqplNIXgTnwN8vKBOLzAPeA64Hnja3d3MeoH/ZWZfJ+gsngr8ewoxiaQmbmfwxO6uE81FevOXZlJxIgjb/G8D+giGjz7g7lvM7G6g3917ge8APzCz7cA+gmRBWO9h4EXgGPBpjRiSRhDtBxhlFmtmsJaFkGaVSh+Buz8OPF5Q9sXI4z8AHytx7j3APWnEIZKGwjuAOEmgu6uz2mGJVE3TdBaLVNtIJoVBsD7QnTMuqlJUItWnRCBC8mUh1BksrUSJQNpediAXKwl0mHHcXZ3B0nKUCKRtJWkK6urs4MsfeZfe/KUlKRFIW0rSFKTmH2l1SgTSduI2BRmwfPY0JQBpeUoE0jaSNAXll4VQEpB2oEQgLS/piCA1BUm7USKQlpQdyHHXY1vYfzD+VpFqCpJ2pUQgLSc7kONzP9rE0PH4i9SqKUjamRKBtJzFjw4mSgJqCpJ2p0QgLWEky0NoboBIQIlAml7SzmCA00d3cM+HlQREQIlAmtzVX1/Ltj1vxK5vBjdN125hIlFKBNJ0RrpK6FxtFylSlBKBNJW4u4VFac9gkfJGVXKymY0zsyfMbFv4fWyROtPM7Dkz22Jmm81sduTY98zsV2a2MfyaVkk80vqW9W1NlAQmdnexfPY0JQGRMiq9I1gIPOXuS81sYfj8CwV1DgJ/4+7bzGwCsMHM+tz9QHh8gbuvqTAOaXFJm4PUDCQSX0V3BMBM4MHw8YPArMIK7v5Ld98WPn4V2AOMr/DnShvJNwfFTQJT33q6koBIApUmgrPcfXf4+LfAWeUqm9llwGjg5UjxPWGT0XIzG1Pm3Plm1m9m/Xv37q0wbGkmSZqDrjh/HE/c/r7qBiTSYoZtGjKzJ4G3FTm0OPrE3d3MSg7lNrOzgR8A89z9eFi8iCCBjAZWEDQr3V3sfHdfEdYhk8kkGTIuTSLa/NNhxpA7E7u7yt4JaFawSOWGTQTuflWpY2b2mpmd7e67wzf6PSXqvQX4KbDY3ddFXjt/N3HYzL4LfD5R9NIyCtcHGvLge+7AoRP7Axea2N3FswuvrF2QIi2q0qahXmBe+Hge8JPCCmY2GngU+H5hp3CYPDAzI+hfeKHCeKQJZQdyfHb1xpLrAznBENCors4OFlx7QdVjE2kHlSaCpcDVZrYNuCp8jpllzOz+sM4NwF8CNxcZJvqQmQ0Cg8CZwJcqjEeaTL4jeLi2Pie4A7Dwu9YIEkmPuTdfc3smk/H+/v56hyEVSLpfgJqBRCpnZhvcPVNYrpnFUnPZgRwL1mzi6FC8DyFqBhKprkqbhkQSW9a3NXYSOK1zlJqBRKpMdwRSc68m2DxeE8NEqk+JQKoqPzfg1QOHmBCO+Z9QZm5Ahxlfu+ES3QGI1JASgaQuOjEsOgcgd+AQix4Z5KN/PpHV63ee1DzUOcpY9jElAZFaUx+BpKpwXaDCnoBDR4d45j/2suz6Sxh7WueJ8u6uTiUBkTrRHYGkKs66QK8eOMSsSyfqTV+kQSgRSCqSLBM9oburBhGJSFxKBFKRpBPDNCdApPEoEciILckO8tC6V4ZdHiLfYayVQkUakxKBjEh2IBcrCejNX6TxKRFIIkn6ArQ+kEhzUCKQWNQXINK6lAhkWPm5AXG3i+zu6uTOGRepOUikSSgRSElJmoFA6wOJNKuKEoGZjQNWA5OBXwM3uPv+IvWGCDafAXjF3WeE5VOAVcAZwAbgr939SCUxSTpu+vZzPPvyvtj11Sks0rwqXWJiIfCUu08FngqfF3PI3aeFXzMi5V8Blrv724H9wC0VxiMpWJIdjJ0Eujo7uHf2NJ5deKWSgEiTqjQRzAQeDB8/SLDvcCzhPsVXAvl9jBOdL9Wz8vmdsep1d3VqrwCRFlBpH8FZ7r47fPxb4KwS9U41s37gGLDU3bMEzUEH3P1YWGcXUPIdxczmA/MBenp6KgxbyhkaZvtSNQOJtJZhE4GZPQm8rcihxdEn7u5mVuod5Fx3z5nZecDT4Yb1v0sSqLuvAFZAsGdxknMlmQ6zksng3tnTlABEWsywicDdryp1zMxeM7Oz3X23mZ0N7CnxGrnw+w4zWwtcCvwY6DazU8K7gnOA3Aj+DZJQdDRQ/k0/+in/xumT+OG6V04674rzxykJiLSgSvsIeoF54eN5wE8KK5jZWDMbEz4+E7gCeNHdHXgGuL7c+ZKuJdlBPrt644khoflP/vlNY7IDOb40613MvbyHDjMguEOYe3kPD33iPXWLW0Sqx3yY9uCyJ5udATwM9AC/IRg+us/MMsAn3f1WM/sL4FvAcYLEc6+7fyc8/zyC4aPjgAFgrrsfHu7nZjIZ7+/vH3Hc7So7kOOzqzeWXR9Iy0KItC4z2+DumcLyijqL3f114ANFyvuBW8PH/wcoOsPI3XcAl1USg8S3rG/rsIvExdlYXkRai7aqbCNx3uS1aYxI+9ESEy2m2OJwoww+Pr2HCd1dZZeL0EJxIu1JdwQtJDuQ4/aHN560Quhxhx+ue4XJZ3TR1dlR9NyJ3V2aHCbSpnRH0EIWPzrI8TKdAOt27OdrN1zCsr6tvHrgEBM0MUxEUCJoKW8cKb9M9JA7sy6dqDd+EXkTNQ21kfy8ABGRKCWCFjLc2/yN0yfVJA4RaS5KBC3kpstLL8Y3VxvGiEgJ6iNoEvn1gcp18ubf6Fc+v5MhdzrMuHH6JCUAESmroiUm6qXdlphYkh3koXWvvGlWcFdnh4Z7ikgipZaYUNNQg8sO5E5KAgCHjg6xrG9rXWISkdaipqEGFWfjeK0LJCJpUCJoQNmBHIseGeTQ0fLzArQukIikQU1DDWhZ39Zhk4CB1gUSkVTojqBBREcFDdd9bwRDRdVRLCJpUCKosyXZwaLbQpaijeNFJG0VJQIzGwesBiYDvybYoWx/QZ33A8sjRRcCc9w9a2bfA97LHzeyv9ndN1YSU7PIDuT4wo83c/jY8Vj1NVxURKql0j6ChcBT7j4VeCp8/ibu/oy7T3P3acCVwEHgf0eqLMgfb5cksCQ7yGdWb4yVBAwtES0i1VVp09BM4H3h4weBtcAXytS/HviZux+s8Oc2pexAjv/5yGYOHo13F6D9g0WkFiq9IzjL3XeHj38LnDVM/TnAyoKye8xss5ktN7MxpU40s/lm1m9m/Xv37q0g5PrI3wXETQIaFSQitTJsIjCzJ83shSJfM6P1PFirouSAFzM7m2AT+75I8SKCPoP/CoyjzN2Eu69w94y7Z8aPHz9c2A0lPzs4CY0KEpFaGbZpyN2vKnXMzF4zs7PdfXf4Rr+nzEvdADzq7if2UYzcTRw2s+8Cn48Zd1NYkh08sQBcXKM7jK9ef4mSgIjUTKV9BL3APGBp+P0nZereSHAHcEIkiRgwC3ihwngaxtVfX8u2PW8kOkdLRYtIPVSaCJYCD5vZLcBvCD71Y2YZ4JPufmv4fDIwCfi3gvMfMrPxBE3iG4FPVhhPQ7jp288lSgKndY7inz5yse4CRKQuKkoE7v468IEi5f3ArZHnvwZOepdz95YaElNsuehyujpH8WUlABGpM80sTkmSGcKaHSwijUSJoALZgRx3PbaF/QePDl85pH4AEWk0SgQjlB3IsWDNJo4OxR8RNPWtpysJiEjDUSIYgaQLxQFccf44HvrEe6oUkYjIyCkRJJB0obg8NQeJSCNTIohpJHcBGhUkIs1AiWAYSTuEO0cZyz6mmcEi0jyUCErIDuRY/Oggbxwpv2VkVHdXJ3fOuEhJQESaihJBEdmBHLc/vJHj8QcEqR9ARJqWEkGB7ECOz67eGHt2MAQjgpQERKRZKRFE3PTt53j25X2x62uNIBFpBUoEJF8jSH0BItJK2j4RXHzHz/n94fgdwuoLEJFW07aJIDuQ43MPbyTBChFKAiLSktouEWQHcix6ZDOHYu4dDDD2tE7u+Cs1BYlIa6po83oz+5iZbTGz4+FmNKXqXWdmW81su5ktjJRPMbPnw/LVZja6kniGc9O3n+MzqzcmSgJXnD+OgS9eoyQgIi2rokRAsLXkR4BflKpgZh3AfcAHgXcCN5rZO8PDXwGWu/vbgf3ALRXGU9KUhT9NNCIIgqYgLRQnIq2u0h3KXgIIthwu6TJgu7vvCOuuAmaa2UvAlcDHw3oPAncC/1pJTMVcuPjxRPMC3jKmg813XZd2GCIiDanSO4I4JgI7I893hWVnAAfc/VhBeer+kHDPACUBEWknw94RmNmTwNuKHFrs7j9JP6SSccwH5gP09PSk/voaESQi7WrYRODuV1X4M3LApMjzc8Ky14FuMzslvCvIl5eKYwWwAiCTySRp6RnWvbOnqTNYRNpWLZqG1gNTwxFCo4E5QK+7O/AMcH1Ybx5QlTuMUztK92Fccf44JQERaWuVDh/9sJntAt4D/NTM+sLyCWb2OED4af82oA94CXjY3beEL/EF4HYz207QZ/CdSuIp5T/u+VDRZKDtI0VEwIIP5s0lk8l4f39/vcMQEWkqZrbB3U+a81WLpiEREWlgSgQiIm1OiUBEpM0pEYiItDklAhGRNteUo4bMbC/wmxGefibwnymGkxbFlYziSkZxJdOqcZ3r7uMLC5syEVTCzPqLDZ+qN8WVjOJKRnEl025xqWlIRKTNKRGIiLS5dkwEK+odQAmKKxnFlYziSqat4mq7PgIREXmzdrwjEBGRCCUCEZE215KJwMw+ZmZbzOy4mZUcamVm15nZVjPbbmYLI+VTzOz5sHx1uI9CGnGNM7MnzGxb+H1skTrvN7ONka8/mNms8Nj3zOxXkWPTahVXWG8o8rN7I+X1vF7TzOy58Pe92cxmR46ler1K/b1Ejo8J//3bw+sxOXJsUVi+1cyurSSOEcR1u5m9GF6fp8zs3Mixor/TGsV1s5ntjfz8WyPH5oW/921mNq/GcS2PxPRLMzsQOVaV62VmD5jZHjN7ocRxM7N/DmPebGbvjhyr/Fq5e8t9Ae8ALgDWApkSdTqAl4HzgNHAJuCd4bGHgTnh428Cn0oprq8CC8PHC4GvDFN/HLAPOC18/j3g+ipcr1hxAf+vRHndrhfwZ8DU8PEEYDfQnfb1Kvf3Eqnz34Fvho/nAKvDx+8M648BpoSv01HDuN4f+Rv6VD6ucr/TGsV1M/AvRc4dB+wIv48NH4+tVVwF9f8eeKAG1+svgXcDL5Q4/iHgZ4ABlwPPp3mtWvKOwN1fcvetw1S7DNju7jvc/QiwCphpZgZcCawJ6z0IzEoptJnh68V93euBn7n7wZR+filJ4zqh3tfL3X/p7tvCx68Ce4CTZk6moOjfS5l41wAfCK/PTGCVux92918B28PXq0lc7v5M5G9oHcG2sNUW53qVci3whLvvc/f9wBPAdXWK60ZgZUo/uyR3/wXBh75SZgLf98A6gm1+zyala9WSiSCmicDOyPNdYdkZwAEPdlaLlqfhLHffHT7+LXDWMPXncPIf4T3hreFyMxtT47hONbN+M1uXb66iga6XmV1G8Cnv5UhxWter1N9L0Trh9fgdwfWJc24144q6heCTZV6x32kt4/po+PtZY2b5vc0b4nqFTWhTgKcjxdW6XsMpFXcq12rYzesblZk9CbytyKHF7l6VvY/jKBdX9Im7u5mVHLsbZvt3EWzxmbeI4A1xNMF44i8Ad9cwrnPdPWdm5wFPm9kgwZvdiKV8vX4AzHP342HxiK9XKzKzuUAGeG+k+KTfqbu/XPwVUvcYsNLdD5vZfyO4m7qyRj87jjnAGncfipTV83pVTdMmAne/qsKXyAGTIs/PCcteJ7jtOiX8VJcvrzguM3vNzM52993hG9eeMi91A/Coux+NvHb+0/FhM/su8PlaxuXuufD7DjNbC1wK/Jg6Xy8zewvwU4IPAesirz3i61VEqb+XYnV2mdkpwJ8S/D3FObeacWFmVxEk1/e6++F8eYnfaRpvbMPG5e6vR57eT9AnlD/3fQXnrk0hplhxRcwBPh0tqOL1Gk6puFO5Vu3cNLQemGrBiJfRBL/0Xg96YJ4haJ8HmAekdYfRG75enNc9qW0yfDPMt8vPAoqOMKhGXGY2Nt+0YmZnAlcAL9b7eoW/u0cJ2k/XFBxL83oV/XspE+/1wNPh9ekF5lgwqmgKMBX49wpiSRSXmV0KfAuY4e57IuVFf6c1jOvsyNMZwEvh4z7gmjC+scA1vPnOuKpxhbFdSND5+lykrJrXazi9wN+Eo46C91IAAAEOSURBVIcuB34XftBJ51pVowe83l/Ahwnayg4DrwF9YfkE4PFIvQ8BvyTI6Isj5ecR/EfdDvwIGJNSXGcATwHbgCeBcWF5Brg/Um8yQaYfVXD+08AgwRvaD4E/qVVcwF+EP3tT+P2WRrhewFzgKLAx8jWtGter2N8LQVPTjPDxqeG/f3t4Pc6LnLs4PG8r8MGU/96Hi+vJ8P9B/vr0Dvc7rVFcXwa2hD//GeDCyLl/F17H7cDf1jKu8PmdwNKC86p2vQg+9O0O/5Z3EfTlfBL4ZHjcgPvCmAeJjIZM41ppiQkRkTbXzk1DIiKCEoGISNtTIhARaXNKBCIibU6JQESkzSkRiIi0OSUCEZE29/8BPmVR4HFqkbkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "M1sV7tBtAWEz"
      },
      "source": [
        "## GRU"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 860
        },
        "id": "usltj8hI3bdL",
        "outputId": "ffaded0d-f60b-4ab3-846c-fdde01befdba"
      },
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten, Input\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "# input_shape = (5,1)\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(GRU(128))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mae\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=1, validation_split=0.1, batch_size=128)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mae = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mae=\", mae)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x)\n",
        "plt.scatter(test_y, y_)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " gru (GRU)                   (None, 128)               50304     \n",
            "                                                                 \n",
            " flatten_2 (Flatten)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 10)                1290      \n",
            "                                                                 \n",
            " dense_5 (Dense)             (None, 1)                 11        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 51,605\n",
            "Trainable params: 51,605\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 0s 3ms/step - loss: 3.9383e-06 - mae: 0.0016\n",
            "loss= 3.938296686101239e-06\n",
            "mae= 0.0016246178420260549\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jCBiKCg9AeHT"
      },
      "source": [
        "## Bidirectional + LSTM"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "azGH_FDbwqHT",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 878
        },
        "outputId": "5fcff34a-ed78-4ceb-cf01-eb70543a203d"
      },
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten, Input\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "# input_shape = (5,1)\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(Bidirectional(LSTM(128)))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mae\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=1, validation_split=0.1, batch_size=128)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mae = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mae=\", mae)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x)\n",
        "plt.scatter(test_y, y_)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_3\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " bidirectional (Bidirectiona  (None, 256)              133120    \n",
            " l)                                                              \n",
            "                                                                 \n",
            " flatten_3 (Flatten)         (None, 256)               0         \n",
            "                                                                 \n",
            " dense_6 (Dense)             (None, 10)                2570      \n",
            "                                                                 \n",
            " dense_7 (Dense)             (None, 1)                 11        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 135,701\n",
            "Trainable params: 135,701\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 0s 4ms/step - loss: 6.1372e-05 - mae: 0.0058\n",
            "loss= 6.137207674328238e-05\n",
            "mae= 0.005842313636094332\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TfY_p_SJA5Rf"
      },
      "source": [
        "## Bidirectional + GRU"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OcjycCe8wqEH",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 878
        },
        "outputId": "7214acef-2b78-44cd-9d43-cfc388f32dae"
      },
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten, Input\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "# input_shape = (5,1)\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(Bidirectional(GRU(128)))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mae\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=1, validation_split=0.1, batch_size=128)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mae = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mae=\", mae)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x)\n",
        "plt.scatter(test_y, y_)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_4\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " bidirectional_1 (Bidirectio  (None, 256)              100608    \n",
            " nal)                                                            \n",
            "                                                                 \n",
            " flatten_4 (Flatten)         (None, 256)               0         \n",
            "                                                                 \n",
            " dense_8 (Dense)             (None, 10)                2570      \n",
            "                                                                 \n",
            " dense_9 (Dense)             (None, 1)                 11        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 103,189\n",
            "Trainable params: 103,189\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 0s 3ms/step - loss: 1.1252e-06 - mae: 8.6885e-04\n",
            "loss= 1.1252236618020106e-06\n",
            "mae= 0.000868848292157054\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P3aO3WEwA-lR"
      },
      "source": [
        "## stacked"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xIYi9GUDwqA6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 895
        },
        "outputId": "1e4829a3-9c62-40ae-815e-56821e9765d6"
      },
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten, Input\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "# input_shape = (5,1)\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(LSTM(128, return_sequences=True))\n",
        "model.add(LSTM(128))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mae\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=1, validation_split=0.1, batch_size=128)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mae = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mae=\", mae)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x)\n",
        "plt.scatter(test_y, y_)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_5\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " lstm_2 (LSTM)               (None, 5, 128)            66560     \n",
            "                                                                 \n",
            " lstm_3 (LSTM)               (None, 128)               131584    \n",
            "                                                                 \n",
            " flatten_5 (Flatten)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_10 (Dense)            (None, 10)                1290      \n",
            "                                                                 \n",
            " dense_11 (Dense)            (None, 1)                 11        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 199,445\n",
            "Trainable params: 199,445\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 0s 5ms/step - loss: 5.2776e-05 - mae: 0.0056\n",
            "loss= 5.2775787480641156e-05\n",
            "mae= 0.0056479270569980145\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OVGRQYg5beg_"
      },
      "source": [
        "## CNN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "pXpoPikkbhpC",
        "outputId": "7f675bd0-79a3-4e5f-ed00-10cd7989c9ce"
      },
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten, Input\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Conv1D, MaxPooling1D, Input, Reshape\n",
        "1\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "# input_shape = (5,1)\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(Conv1D(32, 3, padding=\"same\"))\n",
        "model.add(Conv1D(32, 3, padding=\"same\"))\n",
        "model.add(MaxPooling1D())\n",
        "model.add(Conv1D(32, 3, padding=\"same\"))\n",
        "model.add(Conv1D(32, 3, padding=\"same\"))\n",
        "model.add(MaxPooling1D())\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mae\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=1, validation_split=0.1, batch_size=128)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mae = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mae=\", mae)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x)\n",
        "plt.scatter(test_y, y_)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_6\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv1d (Conv1D)             (None, 5, 32)             128       \n",
            "                                                                 \n",
            " conv1d_1 (Conv1D)           (None, 5, 32)             3104      \n",
            "                                                                 \n",
            " max_pooling1d (MaxPooling1D  (None, 2, 32)            0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv1d_2 (Conv1D)           (None, 2, 32)             3104      \n",
            "                                                                 \n",
            " conv1d_3 (Conv1D)           (None, 2, 32)             3104      \n",
            "                                                                 \n",
            " max_pooling1d_1 (MaxPooling  (None, 1, 32)            0         \n",
            " 1D)                                                             \n",
            "                                                                 \n",
            " flatten_6 (Flatten)         (None, 32)                0         \n",
            "                                                                 \n",
            " dense_12 (Dense)            (None, 10)                330       \n",
            "                                                                 \n",
            " dense_13 (Dense)            (None, 1)                 11        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 9,781\n",
            "Trainable params: 9,781\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 0s 4ms/step - loss: 3.3883e-05 - mae: 0.0048\n",
            "loss= 3.388255208847113e-05\n",
            "mae= 0.004762789234519005\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0MW7OTLyfph4"
      },
      "source": [
        "## CNN + LSTM"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 930
        },
        "id": "rjCgy-f1fph4",
        "outputId": "1be40ae0-8e00-4e9b-b54d-a28090d4397d"
      },
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten, Input\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Conv1D, MaxPooling1D, Input, Reshape\n",
        "1\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "# input_shape = (5,1)\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(Conv1D(32, 3, padding=\"same\"))\n",
        "model.add(Conv1D(32, 3, padding=\"same\"))\n",
        "model.add(LSTM(128))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mae\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=1, validation_split=0.1, batch_size=128)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mae = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mae=\", mae)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x)\n",
        "plt.scatter(test_y, y_)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_7\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv1d_4 (Conv1D)           (None, 5, 32)             128       \n",
            "                                                                 \n",
            " conv1d_5 (Conv1D)           (None, 5, 32)             3104      \n",
            "                                                                 \n",
            " lstm_4 (LSTM)               (None, 128)               82432     \n",
            "                                                                 \n",
            " flatten_7 (Flatten)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_14 (Dense)            (None, 10)                1290      \n",
            "                                                                 \n",
            " dense_15 (Dense)            (None, 1)                 11        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 86,965\n",
            "Trainable params: 86,965\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de3QU93338fd3V1eEpAUjbECLARtisLHWjoxJ3Dh26jjkJDVu69bYcW03aWiauEmaPE/qpDlJjxs3bpMntxOnNb40T5oLcf20Ce0hcS4mV5uLbIMJYAjGNkhgEFdJCF33+/yxI1gUAStppJVmP69z9mhn5je//e4e+OzszPxmzN0REZHoiuW7ABERGVkKehGRiFPQi4hEnIJeRCTiFPQiIhFXlO8C+psyZYrPmjUr32WIiIwrzz777EF3rxlo2ZgL+lmzZtHQ0JDvMkRExhUze/VMy7TrRkQk4hT0IiIRl1PQm9kSM9tuZjvN7N4Blr/PzDab2UYz+5WZLcha9vFgve1m9rYwixcRkXM7Z9CbWRx4EHg7sAC4LTvIA99294XungL+GfhCsO4CYBlwKbAE+FrQn4iIjJJctugXATvdfZe7dwErgaXZDdy9JWuyAui7gM5SYKW7d7r7y8DOoD8RERkluZx1MwPYkzXdCFzdv5GZfQD4CFACvCVr3bX91p0xwLrLgeUAM2fOzKVuERHJUWgHY939QXe/CPhb4JODXHeFu9e7e31NzYCngYqIyBDlEvRNQDJrujaYdyYrgZuHuO6QHW3v4is//S2bG4+NRPciIuNWLkG/AZhrZrPNrITMwdVV2Q3MbG7W5DuA3wbPVwHLzKzUzGYDc4H1wy/7d8Vixhd/soM12w+MRPciIuPWOffRu3uPmd0DPAnEgcfcfYuZ3Qc0uPsq4B4zuwHoBo4AdwXrbjGzx4GtQA/wAXfvHYk3UlVWzEU1E9m05+hIdC8iMm7ldAkEd18NrO4371NZzz90lnXvB+4faoGDkUomWPPiAdwdMxuNlxQRGfMiNTK2Lpng0PEuGo+cyHcpIiJjRqSCPlWbAGCjdt+IiJwUqaC/ZFolJUUx7acXEckSqaAvjse4bHoVmxoV9CIifSIV9ACp5CQ2Nx2juzed71JERMaEyAV9XbKaju40O/a35rsUEZExIXJBn0rqgKyISLbIBf3MyROYNKFYB2RFRAKRC3ozoy6ZYNMeXfNGRAQiGPSQ2X2z40ArbZ09+S5FRCTvIhn0dckE7uhKliIiRDXogxGyOp9eRCSiQT+5ooQLz5vAxt0KehGRSAY9ZLbqtUUvIhLhoE8lE+w71sH+lo58lyIikleRDfo6DZwSEQEiHPSXTq+iKGYaOCUiBS+yQV9WHGf+tCpt0YtIwYts0EPmAmcvNB4jnfZ8lyIikjeRDvpUchJtnT281NyW71JERPIm4kFfDeiArIgUtkgH/ZwpE6ksLdL59CJS0CId9LGYcXmyWlv0IlLQIh30kBkh++K+Vjq6e/NdiohIXkQ+6FPJBD1pZ8teXclSRApTTkFvZkvMbLuZ7TSzewdY/hEz22pmL5jZT83swqxlvWa2MXisCrP4XJy6taCCXkQKU9G5GphZHHgQeCvQCGwws1XuvjWr2fNAvbu3m9lfAf8M3BosO+HuqZDrztnUqjKmVZdphKyIFKxctugXATvdfZe7dwErgaXZDdx9jbu3B5NrgdpwyxyeVDKhA7IiUrByCfoZwJ6s6cZg3pm8B/hB1nSZmTWY2Vozu3mgFcxsedCmobm5OYeSBqcumWD34XYOH+8KvW8RkbEu1IOxZnYHUA98Lmv2he5eD9wOfMnMLuq/nruvcPd6d6+vqakJsyTg1H567b4RkUKUS9A3Acms6dpg3mnM7Abg74Cb3L2zb767NwV/dwE/A64YRr1DsnBGNTHTCFkRKUy5BP0GYK6ZzTazEmAZcNrZM2Z2BfAQmZA/kDV/kpmVBs+nANcA2QdxR0VFaRFzp1ZqhKyIFKRzBr279wD3AE8C24DH3X2Lmd1nZjcFzT4HTAT+o99plPOBBjPbBKwBHuh3ts6oSSUTbNpzFHddyVJECss5T68EcPfVwOp+8z6V9fyGM6z3NLBwOAWGpS6Z4LsNe9h9uJ0Lz6vIdzkiIqMm8iNj+6R0a0ERKVAFE/Tzzp9IWXFMQS8iBadggr4oHmPhjGqdYikiBadggh4yu29+s7eFrp50vksRERk1BRX0dckEXT1ptr/Wmu9SRERGTUEF/akDskfyXImIyOgpqKCfkShnysQSXbJYRApKQQW9mVFXm9AIWREpKAUV9JDZffNScxstHd35LkVEZFQUXNDXJRO4w+ZG7b4RkcJQeEFfqxGyIlJYCi7oqycUM2dKhYJeRApGwQU9ZHbfbNSVLEWkQBRk0KeSCZpbO9l3rCPfpYiIjLiCDPo63VpQRApIQQb9/GmVlMR1JUsRKQwFGfSlRXHmT69S0ItIQSjIoAdI1VazuekYvWkdkBWRaCvcoJ+ZoL2rl98e0JUsRSTaCjbo+wZO6YCsiERdwQb97CkVVJUV6UqWIhJ5BRv0ZnZy4JSISJQVbNBDZuDUjv2ttHf15LsUEZERU/BB35t2ftPUku9SRERGTE5Bb2ZLzGy7me00s3sHWP4RM9tqZi+Y2U/N7MKsZXeZ2W+Dx11hFj9cGiErIoXgnEFvZnHgQeDtwALgNjNb0K/Z80C9u18OPAH8c7DuZODTwNXAIuDTZjYpvPKHZ8rEUmonlbNRd5wSkQjLZYt+EbDT3Xe5exewElia3cDd17h7ezC5FqgNnr8N+LG7H3b3I8CPgSXhlB6OumSCjbsV9CISXbkE/QxgT9Z0YzDvTN4D/GAw65rZcjNrMLOG5ubmHEoKT6o2QdPREzS3do7q64qIjJZQD8aa2R1APfC5wazn7ivcvd7d62tqasIs6ZxSM7WfXkSiLZegbwKSWdO1wbzTmNkNwN8BN7l752DWzafLplcTjxmbtJ9eRCIql6DfAMw1s9lmVgIsA1ZlNzCzK4CHyIT8gaxFTwI3mtmk4CDsjcG8MaO8JM7rzq/UwCkRiaxzBr279wD3kAnobcDj7r7FzO4zs5uCZp8DJgL/YWYbzWxVsO5h4B/IfFlsAO4L5o0pdckEm/YcJa0rWYpIBBXl0sjdVwOr+837VNbzG86y7mPAY0MtcDSkktV8Z/1uXjl0nDk1E/NdjohIqAp6ZGyfVDJzar9234hIFCnogYunTqSiJK4zb0QkkhT0QDxmLKytZmOjLlksItGjoA/UJRNs29tCZ09vvksREQmVgj6Qqk3Q1Ztm2z7dWlBEokVBH+gbIbtx95E8VyIiEi4FfeCCqjKmVpaySfvpRSRiFPQBMyMVDJwSEYkSBX2WumSCXQePc6y9O9+liIiERkGfJdV3xyld4ExEIkRBn2VhbTVmGiErItGioM9SVVbMRTUTtZ9eRCJFQd9PKplgU+NR3HUlSxGJBgV9P3XJBAfbumg8ciLfpYiIhEJB30+qVgdkRSRaFPT9XDKtkpKiGBt3K+hFJBoU9P0Ux2NcNr1KW/QiEhkK+gGkkpPY3HSMnt50vksRERk2Bf0A6pLVdHSn2b5fV7IUkfFPQT+AkyNk9+gCZyIy/inoBzBz8gQmTShm4x5dslhExj8F/QDMjLpkQlv0IhIJCvozSCUT7DjQSltnT75LEREZFgX9GdQlE7jDZt2IRETGuZyC3syWmNl2M9tpZvcOsPxaM3vOzHrM7JZ+y3rNbGPwWBVW4SOtTiNkRSQiis7VwMziwIPAW4FGYIOZrXL3rVnNdgN3A/9rgC5OuHsqhFpH1eSKEi48b4JGyIrIuJfLFv0iYKe773L3LmAlsDS7gbu/4u4vAJEaYVRXm9AWvYiMe7kE/QxgT9Z0YzAvV2Vm1mBma83s5kFVl2epZIJ9xzrY39KR71JERIZsNA7GXuju9cDtwJfM7KL+DcxsefBl0NDc3DwKJeWmLhg4pTtOich4lkvQNwHJrOnaYF5O3L0p+LsL+BlwxQBtVrh7vbvX19TU5Nr1iLt0ehVFMdMdp0RkXMsl6DcAc81stpmVAMuAnM6eMbNJZlYaPJ8CXANsPftaY0dZcZz506q0RS8i49o5g97de4B7gCeBbcDj7r7FzO4zs5sAzOwqM2sE/gR4yMy2BKvPBxrMbBOwBnig39k6Y15dspoXGo+RTuvWgiIyPp3z9EoAd18NrO4371NZzzeQ2aXTf72ngYXDrDGvUslJfHPtbnYdbOPiqZX5LkdEZNA0MvYcUslqAJ7X+fQiMk4p6M9hzpSJVJYW6Xx6ERm3FPTnEIsZlyerdSVLERm3FPQ5qKtNsG1fCx3dvfkuRURk0BT0OUglE/SknS17W/JdiojIoCnoc5DSCFkRGccU9DmYWlXGtOoyjZAVkXFJQZ+jVFJXshSR8UlBn6O6ZIJXD7Vz+HhXvksRERkUBX2O+vbTa6teRMYbBX2OFs6oJmbojlMiMu4o6HNUUVrE3KmV2qIXkXFHQT8IqWSCTXuO4q4rWYrI+KGgH4S6ZIIj7d3sPtye71JERHIWraB/5VdwYuR2rWjglIiMR9EJ+kMvwdffAWu/NmIvMe/8iZQVxxT0IjKuRCfoz7sI5t8Ez3wN2g+PyEsUxWMsnFGtEbIiMq5EJ+gBrv8EdLXB018ZsZdIJRP8Zm8L3b3pEXsNEZEwRSvop86Hy/4Y1j0Ebc0j8hJ1yQRdPWle3Nc6Iv2LiIQtWkEPcN290NMBv/7SiHR/8oCszqcXkXEiekE/ZS5cvgw2PAIt+0LvfkainCkTSzRCVkTGjegFPcCbPwbpHvjVF0Lv2syoq9WVLEVk/Ihm0E+eDal3wbNfh6N7Qu8+lUzwUnMbLR3dofctIhK2aAY9wLX/O/P3l58Pveu6ZAJ32NyoG4aLyNgX3aBPJOH1d8Pz34TDL4fadV2tRsiKyPiRU9Cb2RIz225mO83s3gGWX2tmz5lZj5nd0m/ZXWb22+BxV1iF5+RNH4VYEfzic6F2Wz2hmDlTKhT0IjIunDPozSwOPAi8HVgA3GZmC/o12w3cDXy737qTgU8DVwOLgE+b2aThl52jygvgqr+ATd+BgztD7boumWCjrmQpIuNALlv0i4Cd7r7L3buAlcDS7Abu/oq7vwD0Hy76NuDH7n7Y3Y8APwaWhFB37q75MBSVw88fCLXbVDJBc2snr7V0hNqviEjYcgn6GUD2qSuNwbxc5LSumS03swYza2huDnlE68QauHo5bH4CDmwLrdu6voFTOp9eRMa4MXEw1t1XuHu9u9fX1NSE/wJv/CCUTISffTa0LudPq6QkHtMIWREZ83IJ+iYgmTVdG8zLxXDWDc+EyfCG98PW78O+F0LpsrQozvzpVdqiF5ExL5eg3wDMNbPZZlYCLANW5dj/k8CNZjYpOAh7YzBv9C1+P5RVw5p/DK3LVG01m5uO0ZvWAVkRGbvOGfTu3gPcQyagtwGPu/sWM7vPzG4CMLOrzKwR+BPgITPbEqx7GPgHMl8WG4D7gnmjrzwBb/xr2PEDaHw2lC5TMxO0d/Wy80BbKP2JiIyEnPbRu/tqd5/n7he5+/3BvE+5+6rg+QZ3r3X3Cnc/z90vzVr3MXe/OHj828i8jRxd/T4onwxr7g+lu1MDp46E0p+IyEgYEwdjR01pJfzeh+Gln8LutcPubvaUCqrKiti4R5dCEJGxq7CCHuCq90LFVHjqM8PuysxODpwSERmrCi/oSybAmz4Cr/wSXv7FsLtLJRPs2N9Ke1dPCMWJiISv8IIe4PV/DpXT4an7YZiXMEglE/SmnS17W0IqTkQkXIUZ9MVlcO1HYc/azP76YdAIWREZ6woz6AGuuBOqZw57q37KxFJqJ5VrhKyIjFmFG/RFJZlbDu59Dnb8cFhd1SUT2qIXkTGrcIMeoO42mDwnc159uv+FN3OXqk3QdPQEza2dIRYnIhKOwg76eBG8+V54bTO8+N9D7iY1M7Of/gXtvhGRMaiwgx5g4S0w5XWw5rOQ7h1SF5dNryYeM51PLyJjkoI+Fofr7oXmbbDlv4bURXlJnNedX6mgF5ExSUEPsOBmOP+yzPXqe4c28KkumWCTbi0oImOQgh4gFoPrPg6HdsLmx4fURSpZTUtHDy8fPB5ycSIiw6Og73PJO2BaCn72APR2D3r1VDJzz/NNOiArImOMgr6PGVz/d3D0VXj+m4Ne/eKpE6koiet8ehEZcxT02ea+FWqvgl98HnoGd058PGYsrK1mY6MuWSwiY4uCPlvfVn1LIzz7fwe9el0ywba9LXT2DO00TRGRkaCg72/OdXDhNfDL/wPdJwa1aqo2QVdvmm37WkekNBGRoVDQ99e3Vd/2Gmx4dFCr9o2Q3aTz6UVkDFHQD2TWNZkt+199ETpzv/H3BVVlTK0s1cApERlTFPRncv0nof0grF+R8ypmRioYOCUiMlYo6M8keRXMfRs8/RXoyP3uUXXJBLsOHudY++DPxRcRGQkK+rO5/hNw4gis/ZecV1k8ZzIAn/z+b+jpHfqlj0VEwqKgP5vpKbjknfDMg5nAz8HrL5zM3y65hP/etJcPrnyeboW9iORZTkFvZkvMbLuZ7TSzewdYXmpm3w2WrzOzWcH8WWZ2wsw2Bo9/Dbf8UXD9J6CzBZ7+as6r/NV1F/HJd8xn9ebXuOfbz9HVo7AXkfw5Z9CbWRx4EHg7sAC4zcwW9Gv2HuCIu18MfBH4p6xlL7l7Kni8L6S6R8/5l8Klfwjr/hWOH8p5tb940xz+/g8W8OSW/bz/W89qEJWI5E0uW/SLgJ3uvsvdu4CVwNJ+bZYCfUNJnwB+38wsvDLz7LqPQ3c7/PpLg1rt7mtm85mbL+Mn2w7wl//+LB3dCnsRGX25BP0MYE/WdGMwb8A27t4DHAPOC5bNNrPnzeznZvamYdabHzXzYOGfwvqHoXX/oFa9Y/GFPPBHC/n5jmbe+40GTnQp7EVkdI30wdh9wEx3vwL4CPBtM6vq38jMlptZg5k1NDc3j3BJQ/Tmj0FvV2YQ1SAtWzSTz91Sx692HuTdX99Ae9fQbm4iIjIUuQR9E5DMmq4N5g3YxsyKgGrgkLt3uvshAHd/FngJmNf/Bdx9hbvXu3t9TU3N4N/FaDjvIkjdDg2PwrH+b//cbnl9LV/80xTrXj7E3Y9toK1TYS8ioyOXoN8AzDWz2WZWAiwDVvVrswq4K3h+C/CUu7uZ1QQHczGzOcBcYFc4pefBmz8G7vDLzw9p9ZuvmMFXbruCZ3cf4a7H1tPaoUFVIjLyzhn0wT73e4AngW3A4+6+xczuM7ObgmaPAueZ2U4yu2j6TsG8FnjBzDaSOUj7Pnc/HPabGDWJmXDlnfDcv8ORV4fUxTsvn85Xb7uCTXuOcsej6zl2QmEvIiPLxtrNrOvr672hoSHfZZxZy174cgou/xNY+uCQu/nRltf4wLef43UXVPLN91xNYkJJiEWKSKExs2fdvX6gZRoZO1hV06H+3bDxO3DopSF3c+OlF7Diz+rZsb+N2x5ex+HjXSEWKSJyioJ+KH7vbyBeAj//p3O3PYvrL5nKI3fWs6u5jdsfXsvBtsHdvlBEJBcK+qGoPB8WvRdeeByatw+rq2vn1fDY3VfxyqHj3LZiLQdaO0IqUkQkQ0E/VNd8GEoq4GefHX5XF0/h63++iKajJ1j20FpeO6awF5HwKOiHquI8WPxXsOW/4LXfDLu7xXPO4xvvXsSB1k5uXfEMe48O7n61IiJnoqAfjjd8AEqrQ9mqB6ifNZlvvGcRh9u6uHXFM+w53B5KvyJS2BT0w1E+Cd54D7z4P7D3+VC6vHLmJL713qs51t7NshVr2X1IYS8iw6OgH66r35cJ/DX/GFqXl9cm+PZ7F3O8q4dbVzzDywePh9a3iBQeBf1wlVXBNR+C3/4I9qwPrdvLZlTznfcuprMnza0PPcPOA22h9S0ihUVBH4ZFy6GiBtbcH2q386dVsXL5YtIOy1asZcf+1lD7F5HCoKAPQ0lFZhDVrp/BK78Otet551eycvliYga3rVjLtn0tofYvItGnoA9L/buhcho89ZnMFS5DdPHUiXz3L99AcTzG7Q+vZcveY6H2LyLRpqAPS3E5vOmjsPtp2LUm9O5nT6ngu3+5mAklRdz+8Do2NyrsRSQ3CvowXXknVNXCU/dDOh169xeeV8HK5YupKi/i9kfW8vzuI6G/hohEj4I+TEWlmZuTNDXAlxbCDz8Ou9eGGvrJyRP47vI3MLmihD97dD0Nr4zfy/uLyOhQ0Iftyjvhjx+FaZfDhkfgsbfBFxfA6o9lDtSmh39z8OmJcr67/A1MrSzlzsfWs27XoRAKF5Go0o1HRlJHC+x4ErZ+D3b+BHo6oGIqzP8DuPRmmPlGiBcNufsDLR3c/sg6mo6c4NG76nnjxVNCLF5ExpOz3XhEQT9aOtsyg6q2fg92/Ah6TsCEKTD/nbDgZpj1piGF/sG2Tt718DpeOXSch++s59p5Y/Tm6iIyohT0Y03X8cwW/tbvw/YfQvdxKJ8Ml7wjE/pz3gzx4py7O3y8i3c9so6Xmtt46I7Xc/0lU0eweBEZixT0Y1n3Cdj50yD0fwBdrVCWCEJ/Kcy5LnOQ9xyOtndxx6Pr2PFaG19715XcsOD8ES9dRMYOBf140d2ROQd/6/fhxdXQeQxKq+B1b89s6V/0FiguO+Pqx050c+dj69nSdIyv3n4lSy67YBSLF5F8UtCPRz1dmUsqbP1+5jLIHUehZCLMW5I5kHvxDZlBWv20dHRz92Pr2dR4jAf+aCE3zD+fxIRizGz034OIjBoF/XjX2w0v/zwT+tv+B04chuIKmHdjZvfO3Bsz19sJtHX28O5/28D64Bz7ipI40xPlzJhUzoysv7WTypmRmMDUylJiMX0RiIxnCvoo6e2BV38FW74H2/4b2g9CUTnMfWsm9Oe9DUor6eju5ec7mtlzuJ2moydoOnIi8/foCY62d5/WZXHcmFZ9+pdA9vNpiTJKi+J5esMikgsFfVSle+HVp4Mt/VXQth+KyjK7dRYszfwtnwT9dtsc7+xh79ETNGZ/AWT93d/acdp12cygZmLpab8Iak9+EUxgxqRyJpYOfTyAiAzfsIPezJYAXwbiwCPu/kC/5aXAN4DXA4eAW939lWDZx4H3AL3AB939ybO9loJ+iNK9sGddJvS3fh9a92XmWwxKKzMHdUsrM/v5SyuzHlVQevq87qKJHOou5bXOIprai9h9PM4rrTEaW7ppOnKCvUc76Oo9/bIO1eXFzEiUM/3kLqFTvwimVJZSEo9REo9RXGQUx2MUxUzHDURCNKygN7M4sAN4K9AIbABuc/etWW3eD1zu7u8zs2XAH7r7rWa2APgOsAiYDvwEmOfuZ7wOgII+BOk0NG6APWszo3O72qCzFTpbgr/Zj7bMKZ25KCqH0kq8dCI9RRPpiE3gOBNo8TKO9JZxqKuY/Z0l7O0o5lB3CW2U08oEOn3gMQHxeJyiGBTFYxTHYhTFjaLgS6AoHiMes5Pzi+NGPBYPnsdOrhePZb5ATq3X10ffPDLt40YsFscsRize9zeGWRyLxYjFijJ/g3mxWByLZdaJx/vaxLFYnHgsnmkXixHvmxcvIhYzYjEjbkbMjFgMYpaZNkNfbDKizhb0ufzeXgTsdPddQWcrgaXA1qw2S4G/D54/AXzVMv+qlwIr3b0TeNnMdgb9PTOUNyI5isVg5tWZRy7S6awvg+DR1f8L4dTDOlspDh6Vnc1c0Ne2owXS3WBAySDqdTK/94Z/GaC863UjTYw0hmN0E6OTU/PSGJkPKMOD587vfgmctsz65mWe+gB9cLL97/Yx0GsO8JJnNFB9Z5d7+8H3PXb5ML/MD1bM48qPfi+kak7JJehnAHuyphuB/glyso2795jZMeC8YP7afuvO6P8CZrYcWA4wc+bMXGuXsMRimXvfllUNv6+ezqxfD8GXR8+J3213xh+SZ1gw4C/P0+f1pp3etNOT7qU3DT29aXrSTk86TU9PmrSn8XQa9zTpdBpP92am071wclnfvEw7+uZlP0+nwR33zHqkezPTQT94GnfPWpbOevSefDt2sn4/bZ7/zvv1oMnp7U899aw2p2Yb6X7zgkj1fn2c9ZMfIIb9rJOn3kMuRuwYYR6OPYbwXrqqZg2/jgGMiSNo7r4CWAGZXTd5LkeGo6g086gY/QusxYPHYH5MiBSCXC5T3AQks6Zrg3kDtjGzIqCazEHZXNYVEZERlEvQbwDmmtlsMysBlgGr+rVZBdwVPL8FeMozR3lXAcvMrNTMZgNzgfXhlC4iIrk4566bYJ/7PcCTZH4ZP+buW8zsPqDB3VcBjwL/HhxsPUzmy4Cg3eNkDtz2AB842xk3IiISPg2YEhGJgLOdXqlbCYqIRJyCXkQk4hT0IiIRp6AXEYm4MXcw1syagVeH0cUU4GBI5Yx3+ixOp8/jdPo8TonCZ3Ghu9cMtGDMBf1wmVnDmY48Fxp9FqfT53E6fR6nRP2z0K4bEZGIU9CLiERcFIN+Rb4LGEP0WZxOn8fp9HmcEunPInL76EVE5HRR3KIXEZEsCnoRkYiLTNCb2RIz225mO83s3nzXk09mljSzNWa21cy2mNmH8l1TvplZ3MyeN7P/yXct+WZmCTN7wsxeNLNtZvaGfNeUT2b2N8H/k9+Y2XfMrCzfNYUtEkEf3MD8QeDtwALgtuDG5IWqB/iouy8AFgMfKPDPA+BDwLZ8FzFGfBn4obtfAtRRwJ+Lmc0APgjUu/tlZC7Fviy/VYUvEkFP1g3M3b0L6LuBeUFy933u/lzwvJXMf+TfuVdvoTCzWuAdwCP5riXfzKwauJbMPSRw9y53P5rfqvKuCCgP7o43Adib53pCF5WgH+gG5gUbbNnMbBZwBbAuv5Xk1ZeAjwHpfBcyBswGmoF/C3ZlPWJmFfkuKl/cvQn4PLAb2Accc/cf5beq8EUl6GUAZjYR+H/Ah929Jd/15IOZvRM44O7P5ruWMaIIuBL4F3e/AjgOFOwxLTObRObX/2xgOlBhZnfkt6rwRSXodRPyfsysmEzIf8vd/zPf9eTRNcBNZvYKmV16bzGzb+a3pOigWxoAAADXSURBVLxqBBrdve8X3hNkgr9Q3QC87O7N7t4N/CfwxjzXFLqoBH0uNzAvGGZmZPbBbnP3L+S7nnxy94+7e627zyLz7+Ipd4/cFluu3P01YI+ZvS6Y9ftk7ulcqHYDi81sQvD/5veJ4MHpc94cfDw40w3M81xWPl0D/Bmw2cw2BvM+4e6r81iTjB1/DXwr2CjaBfx5nuvJG3dfZ2ZPAM+ROVvteSJ4OQRdAkFEJOKisutGRETOQEEvIhJxCnoRkYhT0IuIRJyCXkQk4hT0IiIRp6AXEYm4/w8rCFvDXa3OJwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 0s 3ms/step - loss: 6.2293e-05 - mae: 0.0066\n",
            "loss= 6.229300197446719e-05\n",
            "mae= 0.006574156228452921\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iLwCRvt6fyHN"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}