LabAssignmentRNNsToyTimeSeries.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "LabAssignmentRNNsToyTimeSeries.ipynb",
      "provenance": [],
      "toc_visible": true,
      "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/Maloepf/DataModels_Lab1/blob/main/LabAssignmentRNNsToyTimeSeries.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RdFq6io3oWcH"
      },
      "source": [
        "# Processing Sequences Using RNNs (and CNNs)\n",
        "\n",
        "*Credits:* Based on code written by [A. Géron](https://colab.research.google.com/github/ageron/handson-ml2/blob/master/15_processing_sequences_using_rnns_and_cnns.ipynb#scrollTo=AiINDLJHVNep) for his \"Hands-on ML\" book. Code realeased under [Apache 2.0 license](https://github.com/ageron/handson-ml2/blob/master/LICENSE)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vbgX6-aMHanl"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from tensorflow.keras import layers"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PyooZ0K4pCpC"
      },
      "source": [
        "In this activity we are going to explore different ways of modelling sequential data. We will try out simple baseline models as well as 1D CNNs and different RNNs to **forecast the next step in a time series**."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BR4QqhoDihkx"
      },
      "source": [
        "## Generating sample data\n",
        "\n",
        "To experiment with different models, we'll use some randomly generated univariate time series.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-_bQfrh_PGdz"
      },
      "source": [
        "def generate_time_series(batch_size, n_steps):\n",
        "    freq1, freq2, offsets1, offsets2 = np.random.rand(4, batch_size, 1)\n",
        "    time = np.linspace(0, 1, n_steps)\n",
        "    series = 0.5 * np.sin((time - offsets1) * (freq1 * 10 + 10))  #   wave 1\n",
        "    series += 0.2 * np.sin((time - offsets2) * (freq2 * 20 + 20)) # + wave 2\n",
        "    series += 0.1 * (np.random.rand(batch_size, n_steps) - 0.5)   # + noise\n",
        "    return series[..., np.newaxis].astype(np.float32)"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qSZCqVimq9Sz"
      },
      "source": [
        "np.random.seed(42)\n",
        "\n",
        "n_steps = 50\n",
        "series = generate_time_series(10000, n_steps + 1)"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jALDZroTkupK"
      },
      "source": [
        "Notice the shape of the series array:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Dj6Fw0yjrJ4_",
        "outputId": "59140aba-e632-4165-eb38-2a2af450ac54",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "series.shape"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(10000, 51, 1)"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Vd9b9zf7p9Xt"
      },
      "source": [
        "The meaning of each dimension is `[samples, time steps, sequence element size]`.\n",
        "Here each series element is a single scalar because we are making a univariate prediction.\n",
        "If we were modeling mutiple time series at once (eg. temperature and humidity) this would be a multi-variate prediction and our sequence element size would be 2."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KWUj5-Zaqi1D"
      },
      "source": [
        "### Creating predicion targets\n",
        "We want to predict one step further into the series. To do so , we can take the last time step of a series as the regression target, while using all the previous time steps as inputs.\n",
        "\n",
        "We also want to split our data into train, validation and test sets as usual."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XjSOnWaRVNer"
      },
      "source": [
        "X_train, y_train = series[:7000, :n_steps], series[:7000, -1]\n",
        "X_valid, y_valid = series[7000:9000, :n_steps], series[7000:9000, -1]\n",
        "X_test, y_test = series[9000:, :n_steps], series[9000:, -1]"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YItC9egeVNer",
        "outputId": "eb7315a2-efaf-4667-c7d4-3f7f97d666a4",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "X_train.shape, y_train.shape"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((7000, 50, 1), (7000, 1))"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rjIjQ4W23MZ-"
      },
      "source": [
        "### Visualizing the data\n",
        "This function plots the first series samples from a given batch, along with the next predicted value."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4y1Wg-eXVNer",
        "outputId": "b3964f98-13a9-4d8d-e4f2-9a1050e8fbc6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 399
        }
      },
      "source": [
        "def plot_series(series, y=None, y_pred=None, x_label=\"$t$\", y_label=\"$x(t)$\", legend=True):\n",
        "    plt.plot(series, \".-\")\n",
        "    if y is not None:\n",
        "        plt.plot(n_steps, y, \"bo\", label=\"Target\")\n",
        "    if y_pred is not None:\n",
        "        plt.plot(n_steps, y_pred, \"rx\", markersize=10, label=\"Prediction\")\n",
        "    plt.grid(True)\n",
        "    if x_label:\n",
        "        plt.xlabel(x_label, fontsize=16)\n",
        "    if y_label:\n",
        "        plt.ylabel(y_label, fontsize=16, rotation=0)\n",
        "    plt.hlines(0, 0, 100, linewidth=1)\n",
        "    plt.axis([0, n_steps + 1, -1, 1])\n",
        "    if legend and (y or y_pred):\n",
        "        plt.legend(fontsize=14, loc=\"upper left\")\n",
        "\n",
        "fig, axes = plt.subplots(nrows=1, ncols=3, sharey=True, figsize=(12, 4))\n",
        "for col in range(3):\n",
        "    plt.sca(axes[col])\n",
        "    plot_series(X_valid[col, :, 0], y_valid[col, 0],\n",
        "                y_label=(\"$x(t)$\" if col==0 else None),\n",
        "                legend=(col == 0))\n",
        "plt.show()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x400 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "78-2AbhOsvrX"
      },
      "source": [
        "## Prediction task 1 : forecast one time step\n",
        "\n",
        "We want to predict a single continuous value for each input sequence. This can be seen as a sequence-to-vector regression task. As seen in class, a suitable loss function is the `keras.losses.mean_squared_error` (as derived by MLE under a Gaussian assumption on the prediciton targets).\n",
        "\n",
        "\n",
        "As we try out different models, keep the models and their validation performances. In the end you will compare their performances in validation to pick the best model (which will be applied to the test set)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hq8h_fRJu2cb"
      },
      "source": [
        "\n",
        "### Simple baseline models\n",
        "\n",
        "Before trying complex models, it is important to set up a baseline performance obtained with a simple model. This way we will only be interested in the more complex models if they do better than our baseline."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BH8ELltMrrmg"
      },
      "source": [
        "#### Baseline 1: Naive baseline\n",
        "\n",
        "Our first baseline will follow a very simple rule: predict the series will not change. That means the output will be a repetition of the series value at the last observed time step.\n",
        "\n",
        "This is very simple to implement: our predictions `y_pred` will simply be equal to the last time step in the `X` array.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bRxCh_Cfwf9K"
      },
      "source": [
        "\n",
        "##### TODO: write the naive baseline code bellow\n",
        "Compute the naive predictions on the validation set and save them to a `y_pred` array."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "a0F_BH9uqUTL"
      },
      "source": [
        "y_pred = X_valid[:,-1]"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I7EtFfee3k3H"
      },
      "source": [
        "Here is the first series in the validation set with its predicted value:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iqrIG3bXVNet",
        "outputId": "97e0ff3b-2dbd-4aaf-c4bf-6473d6a78de4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 461
        }
      },
      "source": [
        "plot_series(X_valid[0, :, 0], y_valid[0, 0], y_pred[0, 0])\n",
        "plt.show()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YJJoJTwqsiCr"
      },
      "source": [
        "#### Baseline 2: linear baseline\n",
        "\n",
        "Our second basline model will be a linear regression. We can implement it using keras `Sequential` API by simply using a `Dense` layer with no activation functions.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xO4mpDFUwhiw"
      },
      "source": [
        "\n",
        "##### TODO: write a linear regression model using a Dense layer\n",
        "- Don't forget to use `Flatten` and define the expected input shape using the keyword argument `input_shape`\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5UoHY76NuIps",
        "outputId": "644c1820-0ac3-478a-bd22-ce955530cec2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 249
        }
      },
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Flatten\n",
        "\n",
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "def create_linear_regression_model(input_shape):\n",
        "  model = Sequential([\n",
        "    Flatten(input_shape=input_shape),\n",
        "    Dense(1)\n",
        "  ])\n",
        "\n",
        "  return model\n",
        "\n",
        "model_baseline_2=create_linear_regression_model(X_train.shape[1:])\n",
        "model_baseline_2.summary()\n",
        ""
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/reshaping/flatten.py:37: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential_1\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_1\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ flatten_1 (\u001b[38;5;33mFlatten\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m)                  │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)                   │              \u001b[38;5;34m51\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ flatten_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>)                  │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)                   │              <span style=\"color: #00af00; text-decoration-color: #00af00\">51</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m51\u001b[0m (204.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">51</span> (204.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m51\u001b[0m (204.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">51</span> (204.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4c5GO6XVuJR_"
      },
      "source": [
        "This model has parameters to train: we will call `fit` to learn them.\n",
        "\n",
        "##### TODO: train the linear regression model\n",
        " - Use the Adam optimizer with the default learning rate\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JIps0Re0uxeL",
        "outputId": "dde96ff7-97bd-4629-fbde-df05ae065e99",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "from tensorflow.keras.optimizers import Adam\n",
        "\n",
        "model_baseline_2.compile(optimizer=Adam(), loss='mse')\n",
        "\n",
        "history = model_baseline_2.fit(\n",
        "    X_train, y_train,\n",
        "    validation_data=(X_valid, y_valid),\n",
        "    epochs=20,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss (MSE)')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "# Evaluate performance on the validation set\n",
        "val_loss = model_baseline_2.evaluate(X_valid, y_valid, verbose=1)\n",
        "print(f'Validation MSE: {val_loss}')"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.2790 - val_loss: 0.0485\n",
            "Epoch 2/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - loss: 0.0377 - val_loss: 0.0205\n",
            "Epoch 3/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - loss: 0.0180 - val_loss: 0.0119\n",
            "Epoch 4/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - loss: 0.0112 - val_loss: 0.0087\n",
            "Epoch 5/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0084 - val_loss: 0.0075\n",
            "Epoch 6/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - loss: 0.0073 - val_loss: 0.0070\n",
            "Epoch 7/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0068 - val_loss: 0.0066\n",
            "Epoch 8/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1ms/step - loss: 0.0064 - val_loss: 0.0063\n",
            "Epoch 9/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0061 - val_loss: 0.0060\n",
            "Epoch 10/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0059 - val_loss: 0.0058\n",
            "Epoch 11/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0056 - val_loss: 0.0056\n",
            "Epoch 12/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0054 - val_loss: 0.0054\n",
            "Epoch 13/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0053 - val_loss: 0.0052\n",
            "Epoch 14/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0051 - val_loss: 0.0050\n",
            "Epoch 15/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - loss: 0.0049 - val_loss: 0.0048\n",
            "Epoch 16/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0048 - val_loss: 0.0047\n",
            "Epoch 17/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - loss: 0.0047 - val_loss: 0.0045\n",
            "Epoch 18/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - loss: 0.0045 - val_loss: 0.0044\n",
            "Epoch 19/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0044 - val_loss: 0.0043\n",
            "Epoch 20/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.0043 - val_loss: 0.0042\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - loss: 0.0040  \n",
            "Validation MSE: 0.004179316572844982\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZUtD75qcxWNj"
      },
      "source": [
        "### Recurrent models"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_7lXYg0luxzb"
      },
      "source": [
        "#### Model 1: Simple RNN\n",
        "Our first recurrent model will use simple recurrent units. You can implement it using the built-in `keras.layers.SimpleRNN`.\n",
        "\n",
        "Here again we will specify the expected input shape (ignoring the batch dimension). We want the time-steps dimension to be variable: thus we set it to `None`. The sequence element dimension continues to be 1."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AD0qhLwLvfAL",
        "outputId": "672e7d73-51d5-4419-a79b-cc0c62a6316c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 217
        }
      },
      "source": [
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "model_rnn = keras.models.Sequential([\n",
        "    keras.layers.SimpleRNN(1, input_shape=[None, 1])\n",
        "])\n",
        "\n",
        "model_rnn.summary()"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential_4\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_4\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ simple_rnn_2 (\u001b[38;5;33mSimpleRNN\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)                   │               \u001b[38;5;34m3\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ simple_rnn_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SimpleRNN</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)                   │               <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m3\u001b[0m (12.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3</span> (12.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m3\u001b[0m (12.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3</span> (12.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OQrlBTNQvnnj"
      },
      "source": [
        "##### TODO: train the simple RNN model\n",
        " - Use the Adam optimizer with the a learning rate of 1e-5\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KNEebRjExJbL",
        "outputId": "03648718-fa47-4adf-8a9c-808d8e1f6358",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "model_rnn.compile(optimizer=Adam(learning_rate=1e-5),loss='mse')\n",
        "\n",
        "history = model_rnn.fit(\n",
        "    X_train, y_train,\n",
        "    validation_data=(X_valid, y_valid),\n",
        "    epochs=20,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss (MSE)')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "# Evaluate performance on the validation set\n",
        "val_loss = model_rnn.evaluate(X_valid, y_valid, verbose=1)\n",
        "print(f'Validation MSE: {val_loss}')"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 9ms/step - loss: 0.0214 - val_loss: 0.0209\n",
            "Epoch 2/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0211 - val_loss: 0.0207\n",
            "Epoch 3/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 10ms/step - loss: 0.0209 - val_loss: 0.0204\n",
            "Epoch 4/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 11ms/step - loss: 0.0207 - val_loss: 0.0202\n",
            "Epoch 5/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0204 - val_loss: 0.0200\n",
            "Epoch 6/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 8ms/step - loss: 0.0202 - val_loss: 0.0198\n",
            "Epoch 7/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0200 - val_loss: 0.0196\n",
            "Epoch 8/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0198 - val_loss: 0.0194\n",
            "Epoch 9/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 11ms/step - loss: 0.0196 - val_loss: 0.0192\n",
            "Epoch 10/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 10ms/step - loss: 0.0194 - val_loss: 0.0190\n",
            "Epoch 11/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0192 - val_loss: 0.0188\n",
            "Epoch 12/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 9ms/step - loss: 0.0190 - val_loss: 0.0186\n",
            "Epoch 13/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0188 - val_loss: 0.0184\n",
            "Epoch 14/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 8ms/step - loss: 0.0186 - val_loss: 0.0183\n",
            "Epoch 15/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 12ms/step - loss: 0.0185 - val_loss: 0.0181\n",
            "Epoch 16/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 8ms/step - loss: 0.0183 - val_loss: 0.0179\n",
            "Epoch 17/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0181 - val_loss: 0.0178\n",
            "Epoch 18/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 8ms/step - loss: 0.0180 - val_loss: 0.0176\n",
            "Epoch 19/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 10ms/step - loss: 0.0178 - val_loss: 0.0174\n",
            "Epoch 20/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 13ms/step - loss: 0.0177 - val_loss: 0.0173\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - loss: 0.0170\n",
            "Validation MSE: 0.017296982929110527\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3EEjkFk-yKGK"
      },
      "source": [
        "#### Model 2: Deep simple RNN\n",
        "This time we will use multiple recurrent layers in our model. The final output will be computed with a dense layer.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6YTNDegf3rTt"
      },
      "source": [
        "##### Chaining two RNN layers\n",
        "Notice that while recurrent layers expect sequence inputs with 3 dimensions (samples, time, element size), they output by default 2 D data as in `[samples, predicted element size]`.\n",
        "\n",
        "If you chain two RNN layers, the first one needs yield sequence-shaped outputs (with the 3 dimensions) so that it is compatible with the second layer. In this case, the layer needs to be declared with the keywork argument `return_sequences=True`."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "doNNziYp0F6F"
      },
      "source": [
        "##### TODO: Create a Deep RNN with 2 hidden recurrent layers\n",
        "- Use two simple RNN layers with 20 units\n",
        "- Compute the final output doing a linear read-out (use a Dense layer with no activation)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "32b1oHTQytQ1",
        "outputId": "8c103112-a721-4136-c281-5fa161a14075",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        }
      },
      "source": [
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "\n",
        "model_rnn_2hiddenlayer = keras.models.Sequential([\n",
        "    keras.layers.SimpleRNN(20, input_shape=[None, 1], return_sequences=True),\n",
        "    keras.layers.SimpleRNN(20),\n",
        "    keras.layers.Dense(1)\n",
        "])\n",
        "\n",
        "model_rnn_2hiddenlayer.summary()\n"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential_5\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_5\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ simple_rnn_3 (\u001b[38;5;33mSimpleRNN\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)            │             \u001b[38;5;34m440\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ simple_rnn_4 (\u001b[38;5;33mSimpleRNN\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)                  │             \u001b[38;5;34m820\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)                   │              \u001b[38;5;34m21\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ simple_rnn_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SimpleRNN</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">440</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ simple_rnn_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SimpleRNN</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)                  │             <span style=\"color: #00af00; text-decoration-color: #00af00\">820</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)                   │              <span style=\"color: #00af00; text-decoration-color: #00af00\">21</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m1,281\u001b[0m (5.00 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,281</span> (5.00 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m1,281\u001b[0m (5.00 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,281</span> (5.00 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NuJzOArd0dzQ"
      },
      "source": [
        "##### TODO: train the Deep RNN model\n",
        " - Use the Adam optimizer with the default learning rate\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8RxDc74T0mRL",
        "outputId": "a8bf3d96-2d04-4ce5-9127-5b218652792a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "model_rnn_2hiddenlayer.compile(optimizer=Adam(),loss='mse')\n",
        "\n",
        "history = model_rnn_2hiddenlayer.fit(\n",
        "    X_train, y_train,\n",
        "    validation_data=(X_valid, y_valid),\n",
        "    epochs=20,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss (MSE)')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "# Evaluate performance on the validation set\n",
        "val_loss = model_rnn.evaluate(X_valid, y_valid, verbose=1)\n",
        "print(f'Validation MSE: {val_loss}')"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 22ms/step - loss: 0.1107 - val_loss: 0.0058\n",
            "Epoch 2/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 20ms/step - loss: 0.0052 - val_loss: 0.0039\n",
            "Epoch 3/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 27ms/step - loss: 0.0039 - val_loss: 0.0033\n",
            "Epoch 4/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 22ms/step - loss: 0.0034 - val_loss: 0.0031\n",
            "Epoch 5/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 39ms/step - loss: 0.0032 - val_loss: 0.0030\n",
            "Epoch 6/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 27ms/step - loss: 0.0031 - val_loss: 0.0028\n",
            "Epoch 7/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 28ms/step - loss: 0.0030 - val_loss: 0.0027\n",
            "Epoch 8/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 30ms/step - loss: 0.0030 - val_loss: 0.0026\n",
            "Epoch 9/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 30ms/step - loss: 0.0029 - val_loss: 0.0026\n",
            "Epoch 10/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 20ms/step - loss: 0.0029 - val_loss: 0.0026\n",
            "Epoch 11/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 27ms/step - loss: 0.0029 - val_loss: 0.0025\n",
            "Epoch 12/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 27ms/step - loss: 0.0029 - val_loss: 0.0025\n",
            "Epoch 13/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 30ms/step - loss: 0.0029 - val_loss: 0.0025\n",
            "Epoch 14/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 41ms/step - loss: 0.0028 - val_loss: 0.0025\n",
            "Epoch 15/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 22ms/step - loss: 0.0028 - val_loss: 0.0025\n",
            "Epoch 16/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 28ms/step - loss: 0.0028 - val_loss: 0.0024\n",
            "Epoch 17/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 21ms/step - loss: 0.0028 - val_loss: 0.0024\n",
            "Epoch 18/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 30ms/step - loss: 0.0028 - val_loss: 0.0024\n",
            "Epoch 19/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 25ms/step - loss: 0.0028 - val_loss: 0.0024\n",
            "Epoch 20/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 37ms/step - loss: 0.0027 - val_loss: 0.0024\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - loss: 0.0146\n",
            "Validation MSE: 0.01483958214521408\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HPpWzXdw1OuQ"
      },
      "source": [
        "### TODO: Final comparison and model choice\n",
        "Compare the compare performances of different models in validation to pick the best model.\n",
        "Apply it to the test set and evaluate it: is the performance close to what you got in the validation set?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v0kJDf-a1fNf",
        "outputId": "ccce8e9c-d687-40d0-8dfa-2213bec83963",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "test_loss = model_rnn_2hiddenlayer.evaluate(X_test,y_test,verbose=1)\n",
        "val_loss = history.history['val_loss'][-1]\n",
        "\n",
        "\n",
        "print(f'The model with the best val_loss result, is model_rnn_2hiddenlayer.\\n'\n",
        "      f'Test Loss for model_rnn_2hiddenlayer: {test_loss}\\n'\n",
        "      f'Validation Loss: {val_loss}\\n'\n",
        "      f'Test Loss: {test_loss}\\n'\n",
        "      f'Difference between Validation and Test Loss: {abs(val_loss - test_loss)}'\n",
        ")\n"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m32/32\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0027\n",
            "The model with the best val_loss result, is model_rnn_2hiddenlayer.\n",
            "Test Loss for model_rnn_2hiddenlayer: 0.0025727180764079094\n",
            "Validation Loss: 0.0024317163042724133\n",
            "Test Loss: 0.0025727180764079094\n",
            "Difference between Validation and Test Loss: 0.00014100177213549614\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "y7_2OCdW2HHR"
      },
      "source": [
        "## Prediction task 2: forecast multiple time steps\n",
        "\n",
        "This task can be formulated as a sequence-to-sequence task if we predict one output timstep at a time. But we can also formulate it as a sequence-to-vector task, in which we try to predict all time steps at once."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "U0Rpk0d023_f"
      },
      "source": [
        "### Method 1: Predicting 1 step at a time (seq2vec)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LfNZx94JFCjK"
      },
      "source": [
        "We'll use the previous model to predict the next 10 values. We first need to regenerate the sequences with 9 more time steps."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WdEYgOTyFCjL"
      },
      "source": [
        "np.random.seed(42)\n",
        "\n",
        "n_steps = 50\n",
        "series = generate_time_series(10000, n_steps + 10)\n",
        "X_train, Y_train = series[:7000, :n_steps], series[:7000, -10:, 0]\n",
        "X_valid, Y_valid = series[7000:9000, :n_steps], series[7000:9000, -10:, 0]\n",
        "X_test, Y_test = series[9000:, :n_steps], series[9000:, -10:, 0]"
      ],
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cPa-XFggVNey"
      },
      "source": [
        "Now let's predict the next 10 values one by one. We will keep using tha previous output as new input until we predict 10 values."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fOYi2JpgVNey",
        "outputId": "e343309e-3211-425d-98f4-351063f57b21",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "X = X_valid\n",
        "for step_ahead in range(10):\n",
        "    y_pred_one = model.predict(X)[:, np.newaxis, :]\n",
        "    X = np.concatenate([X, y_pred_one], axis=1)\n",
        "\n",
        "Y_pred = X[:, n_steps:, 0]"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SfaioULUVNez",
        "outputId": "dc45433f-c6c9-4554-a5b2-5c0845556d66",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "Y_pred.shape"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2000, 10)"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oENU05Zzxqf5"
      },
      "source": [
        "#### TODO: check the loss value (MSE) for this method"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import mean_squared_error\n",
        "\n",
        "mean_squared_error(Y_valid, Y_pred)"
      ],
      "metadata": {
        "id": "JZt6BxlM5su_",
        "outputId": "4fb3476e-c419-412f-9825-33cf32b44580",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.1125633"
            ]
          },
          "metadata": {},
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "98U87VyPETqC"
      },
      "source": [
        "#### Plot multiple forecasts\n",
        "Use this plotting function to visualize your input series and the output predictions.\n",
        "\n",
        "**Attention**: Reshape the outputs of your models acordingly:\n",
        "If you model outputs 2D arrays with shape [batch size, time steps], you can use `Y[..., np.newaxis]` or `np.expand_dims(Y, shape=-1) ` to make them compatible with this function."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4omE9niVVNey"
      },
      "source": [
        "def plot_multiple_forecasts(X, Y, Y_pred):\n",
        "    # X should be a rank 3 array with shapes [batch size, sequence length, 1]\n",
        "    # Y and y_pred should also be rank 3 with shape [0, sequence lenght, 1]\n",
        "    n_steps = X.shape[1]\n",
        "    ahead = Y.shape[1]\n",
        "    plot_series(X[0, :, 0])\n",
        "    plt.plot(np.arange(n_steps, n_steps + ahead), Y[0, :, 0], \"bo-\", label=\"Actual\")\n",
        "    plt.plot(np.arange(n_steps, n_steps + ahead), Y_pred[0, :, 0], \"rx-\", label=\"Forecast\", markersize=10)\n",
        "    plt.axis([0, n_steps + ahead, -1, 1])\n",
        "    plt.legend(fontsize=14)\n"
      ],
      "execution_count": 60,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_multiple_forecasts(X_valid, Y_valid[..., np.newaxis], Y_pred[..., np.newaxis])\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "-uY7TXiC7aFH",
        "outputId": "10995bfd-80ef-4cb2-a18c-525ff137e29f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 461
        }
      },
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "x2H51keT2osf"
      },
      "source": [
        "#### TODO: Baselines: naive and linear prediction\n",
        "\n",
        "Implement the naive and the linear baselines folowing this method to predict the next 10 steps. Compare their performances to the previous RNN model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TyuiOUFx4gvf"
      },
      "source": [
        "X_naive = X_valid.copy()\n",
        "\n",
        "for step_ahead in range(10):\n",
        "    y_pred_one = X_naive[:, -1, 0]\n",
        "\n",
        "    y_pred_one = y_pred_one[:, np.newaxis, np.newaxis]\n",
        "\n",
        "    X_naive = np.concatenate([X_naive, y_pred_one], axis=1)\n",
        "\n",
        "Y_pred_naive = X_naive[:, n_steps:, 0]\n"
      ],
      "execution_count": 83,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_multiple_forecasts(X_valid, Y_valid[..., np.newaxis], Y_pred_naive[..., np.newaxis])\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "nwC35QIk_hlH",
        "outputId": "0405a592-df1a-44cf-cd7f-6629eb6c657b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 461
        }
      },
      "execution_count": 84,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Linear baseline prediction: use the linear trend of the last two values to predict the next one\n",
        "X_linear = X_valid.copy()  # Start with the validation data\n",
        "\n",
        "for step_ahead in range(10):\n",
        "    # Use the last two values (last time steps) to predict the next value\n",
        "    last_value = X_linear[:, -1, 0]  # The last observed value\n",
        "    second_last_value = X_linear[:, -2, 0]  # The second last value\n",
        "\n",
        "    # Linear extrapolation (difference between last two values)\n",
        "    y_pred_one = last_value + (last_value - second_last_value)  # Linear extrapolation\n",
        "\n",
        "    # Reshape y_pred_one to have shape (samples, 1, 1)\n",
        "    y_pred_one = y_pred_one[:, np.newaxis, np.newaxis]  # Reshaping to (samples, 1, 1)\n",
        "\n",
        "    # Append the predicted value to the sequence\n",
        "    X_linear = np.concatenate([X_linear, y_pred_one], axis=1)  # Concatenate along the time axis (axis=1)\n",
        "\n",
        "# Extract predicted values for the next 10 steps\n",
        "Y_pred_linear = X_linear[:, n_steps:, 0]  # Get the predicted next 10 values\n"
      ],
      "metadata": {
        "id": "TlMTCLgHCBTD"
      },
      "execution_count": 90,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_multiple_forecasts(X_valid, Y_valid[..., np.newaxis], Y_pred_linear[..., np.newaxis])\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "W6em0NvGALkb",
        "outputId": "589ed450-a7de-4957-901e-648cb7f38454",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 461
        }
      },
      "execution_count": 91,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fzFdKKHG_2fn"
      },
      "source": [
        "### Method 2A: Predicting 10 steps at once at the end (seq2vec)\n",
        "\n",
        "In this section we wil predict all 10 steps at once as a vector output at the end of the input sequence processing. Intermediate network outputs will be ignored."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7rdtEMSv4y17"
      },
      "source": [
        "#### Model 1: Deep simple RNN with output prediction at the end\n",
        "\n",
        "Let's create an RNN that predicts all 10 next values at once. To do that, all you need is to change the size of the final `Dense` layer to 10."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OWRGy4zhVNez"
      },
      "source": [
        "##### TODO: build the new RNN model with output size = 10\n",
        "- Use two simple RNN layers with 20 units\n",
        "- Compute the final output doing a linear read-out (use a Dense layer with no activation)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RYn3ggzg41dV",
        "outputId": "0c243623-8eb4-45c0-af11-7e302879c5ca",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        }
      },
      "source": [
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "model_rnn = keras.models.Sequential([\n",
        "    keras.layers.SimpleRNN(20, input_shape=[None,1], return_sequences=True),\n",
        "    keras.layers.SimpleRNN(20),\n",
        "    keras.layers.Dense(10)\n",
        "])\n",
        "\n",
        "model_rnn.summary()"
      ],
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential_6\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_6\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ simple_rnn_5 (\u001b[38;5;33mSimpleRNN\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)            │             \u001b[38;5;34m440\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ simple_rnn_6 (\u001b[38;5;33mSimpleRNN\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)                  │             \u001b[38;5;34m820\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_3 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)                  │             \u001b[38;5;34m210\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ simple_rnn_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SimpleRNN</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">440</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ simple_rnn_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SimpleRNN</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)                  │             <span style=\"color: #00af00; text-decoration-color: #00af00\">820</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)                  │             <span style=\"color: #00af00; text-decoration-color: #00af00\">210</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m1,470\u001b[0m (5.74 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,470</span> (5.74 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m1,470\u001b[0m (5.74 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,470</span> (5.74 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "63luki0pBm6g"
      },
      "source": [
        "##### TODO: train the Deep RNN model\n",
        " - Use the Adam optimizer with the default learning rate\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VJgMX2LE9va9",
        "outputId": "6dc89e8f-c937-4c38-de60-4af68ea3658e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# Training the model\n",
        "\n",
        "model_rnn.compile(optimizer=Adam(), loss='mse')\n",
        "\n",
        "history = model_rnn.fit(\n",
        "    X_train, y_train,\n",
        "    validation_data=(X_valid, y_valid),\n",
        "    epochs=20,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss (MSE)')"
      ],
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 29ms/step - loss: 0.1117 - val_loss: 0.0535\n",
            "Epoch 2/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 20ms/step - loss: 0.0513 - val_loss: 0.0419\n",
            "Epoch 3/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 29ms/step - loss: 0.0405 - val_loss: 0.0325\n",
            "Epoch 4/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 20ms/step - loss: 0.0336 - val_loss: 0.0296\n",
            "Epoch 5/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 25ms/step - loss: 0.0309 - val_loss: 0.0282\n",
            "Epoch 6/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 23ms/step - loss: 0.0293 - val_loss: 0.0273\n",
            "Epoch 7/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 20ms/step - loss: 0.0282 - val_loss: 0.0265\n",
            "Epoch 8/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 29ms/step - loss: 0.0272 - val_loss: 0.0257\n",
            "Epoch 9/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 21ms/step - loss: 0.0262 - val_loss: 0.0250\n",
            "Epoch 10/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 28ms/step - loss: 0.0254 - val_loss: 0.0246\n",
            "Epoch 11/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 21ms/step - loss: 0.0243 - val_loss: 0.0229\n",
            "Epoch 12/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 29ms/step - loss: 0.0228 - val_loss: 0.0204\n",
            "Epoch 13/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 21ms/step - loss: 0.0215 - val_loss: 0.0189\n",
            "Epoch 14/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 22ms/step - loss: 0.0201 - val_loss: 0.0179\n",
            "Epoch 15/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 26ms/step - loss: 0.0189 - val_loss: 0.0172\n",
            "Epoch 16/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 20ms/step - loss: 0.0178 - val_loss: 0.0161\n",
            "Epoch 17/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 27ms/step - loss: 0.0167 - val_loss: 0.0146\n",
            "Epoch 18/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 21ms/step - loss: 0.0156 - val_loss: 0.0132\n",
            "Epoch 19/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 29ms/step - loss: 0.0145 - val_loss: 0.0120\n",
            "Epoch 20/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 20ms/step - loss: 0.0135 - val_loss: 0.0110\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0, 0.5, 'Loss (MSE)')"
            ]
          },
          "metadata": {},
          "execution_count": 93
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "Y_pred = model_rnn.predict(X_valid)\n",
        "\n",
        "Y_valid = Y_valid[..., np.newaxis]\n",
        "Y_pred = Y_pred[..., np.newaxis]\n",
        "\n",
        "plot_multiple_forecasts(X_valid, Y_valid, Y_pred)"
      ],
      "metadata": {
        "id": "p3n0BGkbEN2l",
        "outputId": "f01c53f7-baca-4bc5-d3b7-0ab65c3d3189",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 479
        }
      },
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 10ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B-QP6QRpAOTx"
      },
      "source": [
        "### Method 2B: Predicting 10 steps at once at every time step (seq2seq)\n",
        "\n",
        "Now let's create RNN models that predict the next 10 steps at each time step.\n",
        "\n",
        "This means that at t=0, the model predicts a vector with time steps t=1 to 1=10. Then as t=1, it predcits another vector with t=2 to t=11, and so on, until the end of the series.\n",
        "\n",
        "A crucial difference here is that for each time step we expect a prediction, while before we only cared about the last output. This implies that the loss will contain a term for each time step, not just the last. This way, during backpropagation, each time-step will contribute with a gradient term flowing back, not only though time but also within the same time step. In practice, this is likely to stabilize AND speed-up training.\n",
        "\n",
        "\n",
        "**Note:** In this model and all others used here, predictions for time $T$ only take into account past series values, i.e. where $t < T$. This defines them as a *causal* models."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sxWM2UTTGnQb"
      },
      "source": [
        "#### Generating appropriate data and prediciton targets\n",
        "\n",
        "For this method we'll \"collect\" more data, generating longer series with extra 10 steps.\n",
        "\n",
        "We also need to generate the corresponting target array, which should contain 10-D vectors for all time steps $t>0$."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3jqkemjPVNe0"
      },
      "source": [
        "np.random.seed(42)\n",
        "\n",
        "n_steps = 50\n",
        "# generate data with extra 10 steps\n",
        "series = generate_time_series(10000, n_steps + 10)\n",
        "\n",
        "# prep input arrays\n",
        "X_train = series[:7000, :n_steps]\n",
        "X_valid = series[7000:9000, :n_steps]\n",
        "X_test = series[9000:, :n_steps]\n",
        "\n",
        "# generate target arrays using the last 10 steps\n",
        "Y = np.empty((10000, n_steps, 10))\n",
        "for step_ahead in range(1, 10 + 1):\n",
        "    Y[..., step_ahead - 1] = series[..., step_ahead:step_ahead + n_steps, 0]\n",
        "Y_train = Y[:7000]\n",
        "Y_valid = Y[7000:9000]\n",
        "Y_test = Y[9000:]"
      ],
      "execution_count": 95,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9bfUkIg7VNe0",
        "outputId": "f2bf1b2e-3b73-42ad-d17e-c66385a254b5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "X_train.shape, Y_train.shape"
      ],
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((7000, 50, 1), (7000, 50, 10))"
            ]
          },
          "metadata": {},
          "execution_count": 96
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3EKy30KcChpp"
      },
      "source": [
        "#### Custom loss function\n",
        "\n",
        "All outputs are needed during training, but only the output at the last time step is useful at inference time for predictions and evaluation.\n",
        "So eve though we will rely on the MSE over all the outputs for training, we will use a custom metric for evaluation, to compute MSE only over the output at the last time step:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m5xQ6L8yDFIz"
      },
      "source": [
        "def last_time_step_mse(Y_true, Y_pred):\n",
        "    return keras.metrics.mean_squared_error(Y_true[:, -1], Y_pred[:, -1])"
      ],
      "execution_count": 97,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mVqbmKMrAOTy"
      },
      "source": [
        "#### TODO Model 2: Deep simple RNN with output prediction at every timestep\n",
        "- Use two simple RNN layers with 20 units\n",
        "- Compute the final output doing a linear read-out (use a Dense layer with no activation)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vbzEv7m6DHiW",
        "outputId": "6cb9fc8e-c6cd-40e8-e3d8-551ac7f68da7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        }
      },
      "source": [
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "model_rnn_output_each_timestep = Sequential([\n",
        "    keras.layers.SimpleRNN(20, input_shape=[None, 1], return_sequences=True),\n",
        "    keras.layers.SimpleRNN(20, return_sequences=True),\n",
        "    keras.layers.TimeDistributed(Dense(1))\n",
        "])\n",
        "\n",
        "model_rnn_output_each_timestep.summary()"
      ],
      "execution_count": 100,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential_7\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_7\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ simple_rnn_9 (\u001b[38;5;33mSimpleRNN\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)            │             \u001b[38;5;34m440\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ simple_rnn_10 (\u001b[38;5;33mSimpleRNN\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)            │             \u001b[38;5;34m820\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ time_distributed (\u001b[38;5;33mTimeDistributed\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)             │              \u001b[38;5;34m21\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ simple_rnn_9 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SimpleRNN</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">440</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ simple_rnn_10 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SimpleRNN</span>)            │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">820</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ time_distributed (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">TimeDistributed</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)             │              <span style=\"color: #00af00; text-decoration-color: #00af00\">21</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m1,281\u001b[0m (5.00 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,281</span> (5.00 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m1,281\u001b[0m (5.00 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,281</span> (5.00 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TdyAq1bSKLKX"
      },
      "source": [
        "##### TODO: train the Deep RNN model\n",
        " - Use the Adam optimizer with the default learning rate\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "model_rnn_output_each_timestep.compile(optimizer=Adam(), loss=\"mse\")\n",
        "\n",
        "history = model_rnn_output_each_timestep.fit(\n",
        "    X_train, Y_train,\n",
        "    validation_data=(X_valid, Y_valid),\n",
        "    epochs=20,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss (MSE)')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "# Evaluate performance on the validation set\n",
        "val_loss = model_rnn_output_each_timestep.evaluate(X_valid, Y_valid, verbose=1)\n",
        "print(f'Validation MSE: {val_loss}')\n",
        "\n",
        "# Generate predictions for the validation set\n",
        "Y_pred = model_rnn_output_each_timestep.predict(X_valid)\n",
        "\n",
        "# Call the plot_multiple_forecasts function\n",
        "plot_multiple_forecasts(X_valid, Y_valid, Y_pred)\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "i359NIRIFES0",
        "outputId": "71ed0d4d-c4ff-497a-bff6-5e363c5261aa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "execution_count": 103,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 32ms/step - loss: 0.0855 - val_loss: 0.0847\n",
            "Epoch 2/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 23ms/step - loss: 0.0850 - val_loss: 0.0845\n",
            "Epoch 3/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 32ms/step - loss: 0.0850 - val_loss: 0.0844\n",
            "Epoch 4/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 24ms/step - loss: 0.0849 - val_loss: 0.0844\n",
            "Epoch 5/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 32ms/step - loss: 0.0848 - val_loss: 0.0843\n",
            "Epoch 6/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 25ms/step - loss: 0.0847 - val_loss: 0.0843\n",
            "Epoch 7/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 30ms/step - loss: 0.0846 - val_loss: 0.0842\n",
            "Epoch 8/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 24ms/step - loss: 0.0845 - val_loss: 0.0841\n",
            "Epoch 9/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 33ms/step - loss: 0.0845 - val_loss: 0.0841\n",
            "Epoch 10/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 24ms/step - loss: 0.0843 - val_loss: 0.0840\n",
            "Epoch 11/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 29ms/step - loss: 0.0842 - val_loss: 0.0839\n",
            "Epoch 12/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 26ms/step - loss: 0.0841 - val_loss: 0.0838\n",
            "Epoch 13/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 25ms/step - loss: 0.0840 - val_loss: 0.0837\n",
            "Epoch 14/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 24ms/step - loss: 0.0839 - val_loss: 0.0836\n",
            "Epoch 15/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 28ms/step - loss: 0.0838 - val_loss: 0.0835\n",
            "Epoch 16/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 24ms/step - loss: 0.0838 - val_loss: 0.0834\n",
            "Epoch 17/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 33ms/step - loss: 0.0837 - val_loss: 0.0833\n",
            "Epoch 18/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 24ms/step - loss: 0.0836 - val_loss: 0.0832\n",
            "Epoch 19/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 32ms/step - loss: 0.0835 - val_loss: 0.0832\n",
            "Epoch 20/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 24ms/step - loss: 0.0835 - val_loss: 0.0831\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0834\n",
            "Validation MSE: 0.08307483792304993\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 17ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "psYQ8rPe43_s"
      },
      "source": [
        "#### TODO Model 3: Deep LSTM with output prediction at every timestep\n",
        "\n",
        "- Use two LSTM layers with 20 units\n",
        "- Compute the final output doing a linear read-out (use a Dense layer with no activation)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rUjDSmje5ktJ",
        "outputId": "dde4965a-1d30-4d21-ce94-6fbfa9ce9e65",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        }
      },
      "source": [
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "model_lstm = Sequential([\n",
        "    keras.layers.LSTM(20, input_shape=[None, 1], return_sequences=True),\n",
        "    keras.layers.LSTM(20, return_sequences=True),\n",
        "    keras.layers.TimeDistributed(Dense(1))\n",
        "])\n",
        "\n",
        "model_lstm.summary()\n"
      ],
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential_8\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_8\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ lstm (\u001b[38;5;33mLSTM\u001b[0m)                          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)            │           \u001b[38;5;34m1,760\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ lstm_1 (\u001b[38;5;33mLSTM\u001b[0m)                        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)            │           \u001b[38;5;34m3,280\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ time_distributed_1 (\u001b[38;5;33mTimeDistributed\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)             │              \u001b[38;5;34m21\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ lstm (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>)                          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)            │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,760</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ lstm_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>)                        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)            │           <span style=\"color: #00af00; text-decoration-color: #00af00\">3,280</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ time_distributed_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">TimeDistributed</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)             │              <span style=\"color: #00af00; text-decoration-color: #00af00\">21</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m5,061\u001b[0m (19.77 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">5,061</span> (19.77 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m5,061\u001b[0m (19.77 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">5,061</span> (19.77 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4KZhGpAjBo5F"
      },
      "source": [
        "##### TODO: train the Deep RNN model\n",
        " - Use the Adam optimizer with the default learning rate\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qd_1bYjZ9zns",
        "outputId": "471bf055-fe84-445a-c6eb-df4228e17175",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# Training the model\n",
        "\n",
        "model_lstm.compile(optimizer=Adam(), loss=\"mse\")\n",
        "\n",
        "history = model_lstm.fit(\n",
        "    X_train, Y_train,\n",
        "    validation_data=(X_valid, Y_valid),\n",
        "    epochs=20,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss (MSE)')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.legend()\n",
        "plt.show()\n"
      ],
      "execution_count": 107,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 43ms/step - loss: 0.1110 - val_loss: 0.0961\n",
            "Epoch 2/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 45ms/step - loss: 0.0957 - val_loss: 0.0935\n",
            "Epoch 3/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 47ms/step - loss: 0.0937 - val_loss: 0.0921\n",
            "Epoch 4/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 39ms/step - loss: 0.0915 - val_loss: 0.0898\n",
            "Epoch 5/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 42ms/step - loss: 0.0895 - val_loss: 0.0880\n",
            "Epoch 6/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 46ms/step - loss: 0.0884 - val_loss: 0.0872\n",
            "Epoch 7/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 47ms/step - loss: 0.0877 - val_loss: 0.0866\n",
            "Epoch 8/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 39ms/step - loss: 0.0871 - val_loss: 0.0861\n",
            "Epoch 9/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 45ms/step - loss: 0.0866 - val_loss: 0.0858\n",
            "Epoch 10/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 47ms/step - loss: 0.0863 - val_loss: 0.0854\n",
            "Epoch 11/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 42ms/step - loss: 0.0859 - val_loss: 0.0851\n",
            "Epoch 12/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 39ms/step - loss: 0.0856 - val_loss: 0.0849\n",
            "Epoch 13/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 45ms/step - loss: 0.0854 - val_loss: 0.0846\n",
            "Epoch 14/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 47ms/step - loss: 0.0851 - val_loss: 0.0844\n",
            "Epoch 15/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 42ms/step - loss: 0.0849 - val_loss: 0.0842\n",
            "Epoch 16/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 44ms/step - loss: 0.0847 - val_loss: 0.0840\n",
            "Epoch 17/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 46ms/step - loss: 0.0845 - val_loss: 0.0839\n",
            "Epoch 18/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 46ms/step - loss: 0.0844 - val_loss: 0.0837\n",
            "Epoch 19/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 40ms/step - loss: 0.0842 - val_loss: 0.0835\n",
            "Epoch 20/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 47ms/step - loss: 0.0840 - val_loss: 0.0834\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate performance on the validation set\n",
        "val_loss = model_lstm.evaluate(X_valid, Y_valid, verbose=1)\n",
        "print(f'Validation MSE: {val_loss}')\n",
        "\n",
        "Y_pred = model_lstm.predict(X_valid)\n",
        "plot_multiple_forecasts(X_valid, Y_valid, Y_pred)\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "_MJTpO2bHfBq",
        "outputId": "37652c22-f485-4290-ffee-ec88703d04d2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 515
        }
      },
      "execution_count": 109,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 11ms/step - loss: 0.0837\n",
            "Validation MSE: 0.08338797092437744\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 14ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cY2whzZD5l47"
      },
      "source": [
        "#### TODO Model 4: Deep GRU with output prediction at every timestep\n",
        "- Use two GRU layers with 20 units\n",
        "- Compute the final output doing a linear read-out (use a Dense layer with no activation)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9b6nqVxO5lX1",
        "outputId": "e31ae028-7267-440d-a2c2-7f082e6d3d3a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        }
      },
      "source": [
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "\n",
        "model_gru = Sequential([\n",
        "    keras.layers.GRU(20, input_shape=[None, 1], return_sequences=True),\n",
        "    keras.layers.GRU(20, return_sequences=True),\n",
        "    keras.layers.TimeDistributed(Dense(1))\n",
        "])\n",
        "\n",
        "model_gru.summary()"
      ],
      "execution_count": 110,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential_9\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_9\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ gru (\u001b[38;5;33mGRU\u001b[0m)                            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)            │           \u001b[38;5;34m1,380\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ gru_1 (\u001b[38;5;33mGRU\u001b[0m)                          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m)            │           \u001b[38;5;34m2,520\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ time_distributed_2 (\u001b[38;5;33mTimeDistributed\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)             │              \u001b[38;5;34m21\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ gru (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GRU</span>)                            │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)            │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,380</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ gru_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GRU</span>)                          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)            │           <span style=\"color: #00af00; text-decoration-color: #00af00\">2,520</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ time_distributed_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">TimeDistributed</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)             │              <span style=\"color: #00af00; text-decoration-color: #00af00\">21</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m3,921\u001b[0m (15.32 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,921</span> (15.32 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m3,921\u001b[0m (15.32 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,921</span> (15.32 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "L2PXnUNeBptl"
      },
      "source": [
        "##### TODO: train the Deep RNN model\n",
        " - Use the Adam optimizer with the default learning rate\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ezffRGX290Qr",
        "outputId": "55fee012-f5a6-4083-f16a-e84b41c4d6d5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# Training the model\n",
        "\n",
        "model_gru.compile(optimizer=Adam(), loss=\"mse\")\n",
        "\n",
        "history = model_gru.fit(\n",
        "    X_train, Y_train,\n",
        "    validation_data=(X_valid, Y_valid),\n",
        "    epochs=20,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss (MSE)')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": 111,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 61ms/step - loss: 0.1146 - val_loss: 0.0981\n",
            "Epoch 2/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 58ms/step - loss: 0.0987 - val_loss: 0.0970\n",
            "Epoch 3/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 56ms/step - loss: 0.0972 - val_loss: 0.0935\n",
            "Epoch 4/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 57ms/step - loss: 0.0929 - val_loss: 0.0900\n",
            "Epoch 5/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 59ms/step - loss: 0.0898 - val_loss: 0.0882\n",
            "Epoch 6/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 58ms/step - loss: 0.0881 - val_loss: 0.0873\n",
            "Epoch 7/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 59ms/step - loss: 0.0869 - val_loss: 0.0861\n",
            "Epoch 8/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 58ms/step - loss: 0.0862 - val_loss: 0.0855\n",
            "Epoch 9/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 59ms/step - loss: 0.0858 - val_loss: 0.0852\n",
            "Epoch 10/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 59ms/step - loss: 0.0855 - val_loss: 0.0849\n",
            "Epoch 11/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 57ms/step - loss: 0.0852 - val_loss: 0.0847\n",
            "Epoch 12/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 58ms/step - loss: 0.0850 - val_loss: 0.0846\n",
            "Epoch 13/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 59ms/step - loss: 0.0848 - val_loss: 0.0844\n",
            "Epoch 14/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 58ms/step - loss: 0.0846 - val_loss: 0.0842\n",
            "Epoch 15/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 58ms/step - loss: 0.0844 - val_loss: 0.0840\n",
            "Epoch 16/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 58ms/step - loss: 0.0843 - val_loss: 0.0838\n",
            "Epoch 17/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 57ms/step - loss: 0.0842 - val_loss: 0.0837\n",
            "Epoch 18/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 56ms/step - loss: 0.0841 - val_loss: 0.0835\n",
            "Epoch 19/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 59ms/step - loss: 0.0840 - val_loss: 0.0834\n",
            "Epoch 20/20\n",
            "\u001b[1m219/219\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 59ms/step - loss: 0.0838 - val_loss: 0.0833\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate performance on the validation set\n",
        "val_loss = model_lstm.evaluate(X_valid, Y_valid, verbose=1)\n",
        "print(f'Validation MSE: {val_loss}')\n",
        "\n",
        "Y_pred = model_lstm.predict(X_valid)\n",
        "plot_multiple_forecasts(X_valid, Y_valid, Y_pred)\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "Hlr1FczYHoCE",
        "outputId": "d8726ae6-6d0f-4b43-eaea-51548d1dba5a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 515
        }
      },
      "execution_count": 112,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 13ms/step - loss: 0.0837\n",
            "Validation MSE: 0.08338797092437744\n",
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aUm5-9itHWX0"
      },
      "source": [
        "### TODO: Model comparison for prediction task 2\n",
        "\n",
        "Create a summary table with the validation performaces of all models used in this secont task. Choose the best of them and evaluate it on the test set. Did it generalize as well as predicted by the validation score?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3UC2w7btHTep"
      },
      "source": [
        "## Extra models\n",
        "\n",
        "Sequences can be treated with convolutional layers as we'll see in class soon. Here are two example models applied to the time series data we have used in this notebook."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Bb3-HpgkIRBJ"
      },
      "source": [
        "#### Generating appropriate data and prediciton targets"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fb_8S82_IRBK"
      },
      "source": [
        "np.random.seed(42)\n",
        "\n",
        "n_steps = 50\n",
        "series = generate_time_series(10000, n_steps + 10)\n",
        "X_train = series[:7000, :n_steps]\n",
        "X_valid = series[7000:9000, :n_steps]\n",
        "X_test = series[9000:, :n_steps]\n",
        "Y = np.empty((10000, n_steps, 10))\n",
        "for step_ahead in range(1, 10 + 1):\n",
        "    Y[..., step_ahead - 1] = series[..., step_ahead:step_ahead + n_steps, 0]\n",
        "Y_train = Y[:7000]\n",
        "Y_valid = Y[7000:9000]\n",
        "Y_test = Y[9000:]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nUviOPLDIRBL"
      },
      "source": [
        "X_train.shape, Y_train.shape"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AfVfbrac8N-F"
      },
      "source": [
        "#### Model 5: 1D convolution + GRUs\n",
        "**Attention** Due to using a convolution stride of 2, the convolutional layer subsamples the input series by a factor of 2. That means that the predicted outputs will be related to the prediction targets with a similar subsampling factor.\n",
        "\n",
        "Additionally, the kernel size of 4 in combination with using a valid convolution makes that the first predicted output corresponds to the 4th time-steps.\n",
        "\n",
        "This means you should compare your predictions to a cropped and subsampled version of the targets array, as follows:\n",
        "```\n",
        "subsampled_target = target[:, 3::2]\n",
        "```"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aZ9VYQJ28Nl0"
      },
      "source": [
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "model = keras.models.Sequential([\n",
        "    keras.layers.Conv1D(filters=20, kernel_size=4, strides=2, padding=\"valid\",\n",
        "                        input_shape=[None, 1]),\n",
        "    keras.layers.GRU(20, return_sequences=True),\n",
        "    keras.layers.GRU(20, return_sequences=True),\n",
        "    keras.layers.TimeDistributed(keras.layers.Dense(10))\n",
        "])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a3JoEZ5DBrtO"
      },
      "source": [
        "##### TODO: train the conv + GRU model\n",
        " - Use the Adam optimizer with the default learning rate\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_dcu6mqzBsfT"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4Tiilsvt8UD8"
      },
      "source": [
        "#### Model 6: WaveNet-like model\n",
        "![Wavenet diagram](http://benanne.github.io/images/wavenet.png)\n",
        "[From Oord et al., 2016](https://arxiv.org/abs/1609.03499)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e_ZlhU-mVNe5"
      },
      "source": [
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "model = keras.models.Sequential()\n",
        "model.add(keras.layers.InputLayer(input_shape=[None, 1]))\n",
        "for rate in (1, 2, 4, 8) * 2:\n",
        "    model.add(keras.layers.Conv1D(filters=20, kernel_size=2, padding=\"causal\",\n",
        "                                  activation=\"relu\", dilation_rate=rate))\n",
        "model.add(keras.layers.Conv1D(filters=10, kernel_size=1))\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eUAOLoDdIIxt"
      },
      "source": [
        "##### TODO: train the WaveNet model\n",
        " - Use the Adam optimizer with the default learning rate\n",
        " - Train it it for 20 epochs\n",
        " - Don't forget to include your validation data\n",
        " - Check the learning curves and the performance on the validation set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ox-NYduk8Ul2"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    }
  ]
}