CATP/Demo_CATP_outputs_deleted.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_nEEW_yUm07_"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "*Necessary imports*"
      ],
      "metadata": {
        "id": "OLNzTQwqeoyC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense\n",
        "from tensorflow.keras.losses import MeanSquaredError\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import matplotlib\n",
        "from tqdm import tqdm\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Reshape, Flatten\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import LSTM, Reshape, Dense\n",
        "import matplotlib.pyplot as plt\n",
        "import time\n",
        "starttime = time.time()\n",
        "# matplotlib.use('TkAgg')"
      ],
      "metadata": {
        "id": "RDs7jWGoeu_e"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "dIsWJhvchJG9"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Dummy data generation\n",
        "We consider `n_sensors` sensors, each having data as a time series of sine wave with some uniform zero-mean noise added. Each sensors has a phase difference. The parameters are:\n",
        "\n",
        "* `n_sensors`: number of sensors in the dataset; 3 for this data\n",
        "*  `cycles`: How many time periods of sine wave in full data set; 300 for this data\n",
        "*  `resolution`: Total number of time stamps  in `cycles` time periods; 10K for this data\n",
        "* `phase`: phase difference between sensors; 9 for this data\n",
        "* `PERIODICITY`: computed as $$\\frac{\\text{resolution}}{\\text{cycles}}$$. This is also referred to as the daily offset in this demo script, since the periodicity is later used to compute the nearest neighbours in temporal bands at multiples of periodicity; (cf.`Figure 2`. in the paper)\n",
        "* `NOISE`: magnitude of noise in data; default 0.2\n"
      ],
      "metadata": {
        "id": "LLnwKrLterNS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "n_sensors = 3\n",
        "\n",
        "import os\n",
        "if not os.path.exists(\"plots_from_demo_data\"):\n",
        "    os.mkdir(\"plots_from_demo_data\")\n",
        "\n",
        "def generate_continuous_dataset():\n",
        "    \"\"\"\n",
        "    Generates a continuous dataset based on sine wave cycles with added noise.\n",
        "\n",
        "    The function creates a dataset of sine waves for a specified number of cycles and resolution.\n",
        "    Noise is added to simulate real-world data. The dataset is plotted and saved as an image.\n",
        "\n",
        "    Returns:\n",
        "        tuple: A tuple containing the generated dataset and the periodicity of the sine waves.\n",
        "    \"\"\"\n",
        "    cycles = 300 # how many sine cycles\n",
        "    resolution = 10000 # how many datapoints to generate\n",
        "    phase = 9\n",
        "\n",
        "    length = np.pi * 2 * cycles\n",
        "\n",
        "    sensor_list = []\n",
        "    NOISE_LEVEL = 0.1\n",
        "\n",
        "    for i in range(n_sensors):\n",
        "        values = np.sin( (np.arange(0, length, length / resolution)) - phase * i )\n",
        "        shape = values.shape\n",
        "        noise_with_mean_zero = (np.random.rand(*shape) - 0.5)\n",
        "\n",
        "        sensor_list.append ( values + noise_with_mean_zero * NOISE_LEVEL )\n",
        "\n",
        "    dataset = np.random.rand(len(sensor_list), resolution) * 0\n",
        "\n",
        "    for counter, sensor in enumerate(sensor_list):\n",
        "        dataset[counter, :] = sensor\n",
        "        plt.plot(sensor, label= \"sensor \" + str(counter+1))\n",
        "\n",
        "    PERIODICITY = int(resolution/cycles)\n",
        "    plt.plot(range(phase, PERIODICITY + phase), [1.5] * PERIODICITY, color=\"black\", linestyle=\"--\", label=\"daily offset \" + str (PERIODICITY))\n",
        "\n",
        "    plt.legend()\n",
        "    plt.ylim(-2, 2)\n",
        "    plt.xlim(0, 120)\n",
        "    plt.title(\"Raw data\")\n",
        "    plt.savefig(\"plots_from_demo_data/two_sensors_time_series.jpg\", dpi=300)\n",
        "    plt.show()\n",
        "    plt.clf()\n",
        "\n",
        "    return dataset, PERIODICITY\n",
        "large_dataset, PERIODICITY = generate_continuous_dataset()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 490
        },
        "id": "TLJfMu0dgcRL",
        "outputId": "fec368ff-c580-4955-c5ac-2423d738056d"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Convert to supervised labels:**\n",
        "The time series needs to converted to supervised labels in order to be modelled by the DL models. This is accomplished using the function `sample_blocks_for_XY`. The `i_o` value decides the number of time frames in input and ouput data. (cf. Figure 2 from the paper)\n"
      ],
      "metadata": {
        "id": "9vq2sYQohW4i"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def dataloader(large_dataset, timestamp, i_o):\n",
        "    \"\"\"\n",
        "    Loads a subset of data from a larger dataset.\n",
        "    The dataloader should be able to access the neighbours using a timestamp index.\n",
        "\n",
        "    Args:\n",
        "        large_dataset (np.ndarray): The larger dataset from which to load data.\n",
        "        i (int): The starting index for loading data.\n",
        "        i_o (int): The size of the input-output blocks.\n",
        "\n",
        "    Returns:\n",
        "        tuple: A tuple containing the X and Y blocks for the specified index.\n",
        "\n",
        "    Refer to: https://github.com/mie-lab/Complexity-Aware-Traffic-Prediction/blob/1c7c302e68276a5c8a61be7bbefca5d36b871ec6/CATP/preprocessing/ProcessRaw.py#L261C2-L261C2\n",
        "    for implementation for traffic-4-cast dataset\n",
        "    \"\"\"\n",
        "    x_i = large_dataset[:,timestamp: timestamp+i_o]\n",
        "    y_i = large_dataset[:,timestamp + i_o + 1: timestamp + 2 * i_o + 1]\n",
        "    return x_i, y_i\n",
        "\n",
        "def sample_blocks_for_XY(dataset, i_o):\n",
        "    \"\"\"\n",
        "    Converts sequential data into X,Y for time series supervised regression task\n",
        "    This function ensures that there is enough room in the dataset for the input-output blocks.\n",
        "\n",
        "    Args:\n",
        "        dataset (np.ndarray): The dataset from which to sample.\n",
        "        i_o (int): The size of the input-output blocks.\n",
        "\n",
        "    Returns:\n",
        "        tuple: A tuple containing arrays for X, Y, and a list of indices.\n",
        "    \"\"\"\n",
        "    X, Y = [], []\n",
        "    max_start_index = dataset.shape[1] - 2 * i_o - 1 # Ensure room for i_o at the ends of dataset\n",
        "    indices_list = []\n",
        "    for i in range(max_start_index):\n",
        "        x_block, y_block = dataloader(dataset, timestamp=i, i_o=i_o) # dataset[:, start_index:start_index + i_o]\n",
        "        X.append(x_block)\n",
        "        Y.append(y_block)\n",
        "        indices_list.append(i)\n",
        "    return np.array(X), np.array(Y), indices_list"
      ],
      "metadata": {
        "id": "PF7078Pwh7fZ"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Build model**\n",
        "We use keras with tensorflow backend, the input and output shapes are shown for references"
      ],
      "metadata": {
        "id": "RQFjlJ7jh8UD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def build_model_fc(i_o):\n",
        "    # Calculate the total number of elements in the input (e.g., 2*100 for a 2x100 input)\n",
        "    model = Sequential([\n",
        "        Flatten(input_shape=((n_sensors, i_o))),\n",
        "        Dense(64, activation='relu'),\n",
        "        Dense(n_sensors * i_o),  # Output layer with as many neurons as the total elements in the input\n",
        "        Reshape((n_sensors, i_o))  # Reshape the output to match the input shape\n",
        "    ])\n",
        "    model.compile(optimizer='sgd', loss='mse')\n",
        "    return model\n",
        "\n",
        "def linf_distance(a,b):\n",
        "    \"\"\"\n",
        "    Computes the L-infinity distance between two vectors.\n",
        "    Args:\n",
        "        a (np.ndarray): The first array.\n",
        "        b (np.ndarray): The second array.\n",
        "\n",
        "    Returns:\n",
        "        float: The L-infinity distance between the two arrays.\n",
        "    \"\"\"\n",
        "    return np.max(np.abs(a.flatten()-b.flatten()))\n",
        "\n",
        "dummy_model = build_model_fc(i_o = 10)\n",
        "dummy_model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 324
        },
        "id": "xj_ivC1NimRg",
        "outputId": "7a4a3145-bf96-4f13-9cc8-4b1b03df654d"
      },
      "execution_count": 4,
      "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\"\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\"</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 (\u001b[38;5;33mFlatten\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m)                  │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense (\u001b[38;5;33mDense\u001b[0m)                        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)                  │           \u001b[38;5;34m1,984\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m)                  │           \u001b[38;5;34m1,950\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ reshape (\u001b[38;5;33mReshape\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m, \u001b[38;5;34m10\u001b[0m)               │               \u001b[38;5;34m0\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 (<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\">30</span>)                  │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense (<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\">64</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,984</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\">30</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,950</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ reshape (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)               │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m3,934\u001b[0m (15.37 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\">3,934</span> (15.37 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m3,934\u001b[0m (15.37 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\">3,934</span> (15.37 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",
      "source": [
        "# Model complexity function\n",
        "\n",
        "To make the model complexity computation efficient (and straightforward to implement), the key implementation steps are shown below:\n",
        "\n",
        "\n",
        "\n",
        "* **Custom Dataloader**:  For each input data point, find its temporal neighbours; this is a constant time operations for time series tasks, since the dataloader can be tweaked to return the input data point at a given `timestamp`."
      ],
      "metadata": {
        "id": "F48rIOEvjAcU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "def dataloader(large_dataset, timestamp, i_o):\n",
        "    \"\"\"\n",
        "    Loads a subset of data from a larger dataset.\n",
        "    The dataloader should be able to access the neighbours using a timestamp index.\n",
        "\n",
        "    Args:\n",
        "        large_dataset (np.ndarray): The larger dataset from which to load data.\n",
        "        i (int): The starting index for loading data.\n",
        "        i_o (int): The size of the input-output blocks.\n",
        "\n",
        "    Returns:\n",
        "        tuple: A tuple containing the X and Y blocks for the specified index.\n",
        "\n",
        "    Refer to: https://github.com/mie-lab/Complexity-Aware-Traffic-Prediction/blob/1c7c302e68276a5c8a61be7bbefca5d36b871ec6/CATP/preprocessing/ProcessRaw.py#L261C2-L261C2\n",
        "    for implementation for traffic-4-cast dataset\n",
        "    \"\"\"\n",
        "    x_i = large_dataset[:,timestamp: timestamp+i_o]\n",
        "    y_i = large_dataset[:,timestamp + i_o + 1: timestamp + 2 * i_o + 1]\n",
        "    return x_i, y_i"
      ],
      "metadata": {
        "id": "EmiTgrrmkJwv"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* **Clubbing all `model.predict` together**: To make the Model complexity ($MC$) efficient, it is recommended to compute all predictions using a dataloader since it is inefficient to call `model.predict` intermittently while doing other processing for neighbourhood search etc.. Once all predictions are ready, we can compute the model complexity by measuring the degree to which the model transforms the input space. As excerpted from the `model_complexity_MC` function below, since here our data set is small, all predictions can be stored in a list. For real datasets, (as was the case in our experiments in the paper), all predictions can be saved to disk and a similar dataloader can be used to extract the relevant predictions later on.\n",
        "    ```python\n",
        "    # predict for all data points so that we can process later\n",
        "    predicted = [0] * N\n",
        "    for i in tqdm(range(0, N, batch_size), \"Predicting for all data points\"):\n",
        "        X = []\n",
        "        for j in range(batch_size):\n",
        "            x,y = dataloader(large_dataset, j, i_o=i_o)\n",
        "            X.append(x)\n",
        "        X = np.array(X)\n",
        "        predicted[i:i+X.shape[0]] = [predicted for predicted in model_predict(X.reshape((-1, n_sensors, i_o)))]\n",
        "\n",
        "    ```\n",
        "\n",
        "* **Determine the neighbours**: The neighbours in temporal band of look forward and backward policy of `n_h` at multiples of periodicity are searched by:\n",
        "    ```python\n",
        "    for day in range(-n_d, n_d+1):\n",
        "        for hour in range(-n_h, n_h+1):\n",
        "            j = i + day * periodicity + hour\n",
        "    ```\n",
        "From Equation 4 in the paper, we had the set of neighbours for data tensor at time stamp `t` as  \\mathbb{U}_t, given by:\n",
        "$$\n",
        " \\mathbb{U}_t = \\left\\{ t' \\: | \\: t' \\neq t, \\: t' =~t~\\pm \\underbrace{n_d \\cdot 24 \\cdot \\frac{60}{p}}_{\\text{daily periodicity offset}} \\pm \\underbrace{~~~n_h~~~}_{h \\text{ hours look forward & backward}} \\right\\}\n",
        "$$\n",
        "Given our custom dataloader, the input data point at timestamp `j` values can be extracted in constant time as:\n",
        "```python\n",
        "x_j, y_j = dataloader(large_dataset, j, i_o=i_o)\n",
        "```\n",
        "\n",
        "* **Compute the maximum distance in input space**\n",
        "From equations 6 and 7 in the paper, we have:\n",
        "\\begin{equation}\n",
        "\\mathbb{T}_\\mathbf{x} = \\{\\mathbf{x_t} \\: | \\: t \\in \\mathbb{U}_t\\}\n",
        "\\end{equation}\n",
        "\n",
        "\\begin{equation}\n",
        "    r_\\mathbf{x}=\\text{max}(\\{||\\mathbf{x_i}-\\mathbf{x}||_{\\infty}\\}\\; | \\mathbf{x_i} \\in \\mathbb{T}_\\mathbf{x})\n",
        "\\end{equation}\n",
        "\n",
        "In the function `model_complexity_MC`, the $r_x$ is computed using:\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "```\n",
        ".\n",
        ".   # inside the loop\n",
        "    neighbour_index_list.append(j)\n",
        "    x_distance_list.append(linf_distance(x_i, x_j))\n",
        "max_dist_x = np.max(x_distance_list)\n",
        "```\n",
        "\n",
        "\n",
        "\n",
        "* **Sample complexity**:\n",
        "For each data point, we track the predictions (output of `model.predict`, which has been pre-computed and saved in the variable `predictions`), to compute the criticality defined in Equation 8 as:\n",
        "\\begin{equation}\n",
        "SC(\\mathbf{x}|f,\\mathcal{D}) = \\sum_{\\mathbf{x_j} \\in \\mathbb{T}_\\mathbf{x}} \\left(d_{f(\\mathbf{x_j})} - r_\\mathbf{x}\\right) \\cdot 1_{d_{f(\\mathbf{x_j})}>r_\\mathbf{x}}\n",
        "\\end{equation}\n",
        "```python\n",
        "        # inside loop\n",
        "        compute_criticality = [0]\n",
        "        for y_distance in y_distance_list:\n",
        "            if y_distance > max_dist_x:\n",
        "                compute_criticality.append(y_distance)\n",
        "        criticality = sum(compute_criticality)\n",
        "        list_of_criticality_values.append(criticality)\n",
        "```\n",
        "\n",
        "* **$MC$ as the mean over all $n$ sample complexity values**:\n",
        "\n",
        "```python\n",
        "# outside loop\n",
        "return np.mean(list_of_criticality_values) # computed complexity value of the model\n",
        "```\n",
        "Reproducing the Equation 9 from the paper, we have:\n",
        "\\begin{equation}\n",
        "MC(f| \\mathcal{D}) = \\frac{1}{N}\\sum_{k=1}^{N} SC(\\mathbf{x}_k|f, \\mathcal{D})\n",
        "\\end{equation}"
      ],
      "metadata": {
        "id": "nAqzIZtBkQHp"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "def model_complexity_MC(large_dataset,\n",
        "                i_o,\n",
        "                n,\n",
        "                model_predict,\n",
        "                periodicity,\n",
        "                n_d=3,\n",
        "                n_h=2,\n",
        "                batch_size=32):\n",
        "    \"\"\"\n",
        "    Computes the model complexity metric using a given DL model\n",
        "\n",
        "    This function is similar to `model_complexity_MC` but uses the ground truth data as the prediction\n",
        "    from the perfect model. It calculates the intrinsic complexity based on input-output distances.\n",
        "\n",
        "    Args:\n",
        "        large_dataset (np.ndarray): The dataset to compute the complexity on.\n",
        "        i_o (int): The number of frames in input and output.\n",
        "        n (int): The number of data points to consider in the complexity calculation.\n",
        "        periodicity refers to the offset required for 1 day;\n",
        "        n_d=3 corresponds to 1 week of neighbours (3 days look ahead and back; and the current day)\n",
        "        n_h=2 corresponds to 2 hours of neighbours (1 hour look ahead and back)\n",
        "        model_predict (function): The prediction function of the model\n",
        "    \"\"\"\n",
        "\n",
        "\n",
        "    list_of_criticality_values = []\n",
        "\n",
        "    N =  large_dataset.shape[1]\n",
        "\n",
        "    # predict for all data points so that we can process later\n",
        "    predicted = [0] * N\n",
        "    for i in tqdm(range(0, N, batch_size), \"Predicting for all data points\"):\n",
        "        X = []\n",
        "        for j in range(batch_size):\n",
        "            x,y = dataloader(large_dataset, j, i_o=i_o)\n",
        "            X.append(x)\n",
        "        X = np.array(X)\n",
        "\n",
        "        predicted[i:i+X.shape[0]] = [predicted for predicted in model_predict(X.reshape((-1, n_sensors, i_o)))]\n",
        "\n",
        "\n",
        "    for i in tqdm(range(i_o, n), \"Iterating over all \" + str(n) + \" data points\"):\n",
        "\n",
        "        # create list of all neighbours in temporal bands\n",
        "        # at multiples of periodicity\n",
        "        neighbour_list = []\n",
        "        x_distance_list = []\n",
        "        neighbour_index_list = []\n",
        "        x_i, y_i = dataloader(large_dataset, i, i_o=i_o)\n",
        "\n",
        "        f_x_i = predicted[i]             # f(x_i)\n",
        "\n",
        "        for day in range(-n_d, n_d+1):\n",
        "            for hour in range(-n_h, n_h+1):\n",
        "\n",
        "                j = i + day * periodicity + hour\n",
        "\n",
        "                # ignore the tensors which are at the boundaries of the dataset\n",
        "                if j + i_o < 0 or j < 0 or \\\n",
        "                        j >= large_dataset.shape[1] or j+i_o >= large_dataset.shape[1]:\n",
        "                    continue\n",
        "\n",
        "                if j != i:\n",
        "                    x_j, y_j = dataloader(large_dataset, j, i_o=i_o)\n",
        "                    assert (x_j.shape == x_i.shape)\n",
        "                    neighbour_list.append(x_j)\n",
        "                    neighbour_index_list.append(j)\n",
        "                    x_distance_list.append(linf_distance(x_i, x_j))\n",
        "\n",
        "        max_dist_x = np.max(x_distance_list)\n",
        "\n",
        "        y_distance_list = []\n",
        "        for neighbour_index in neighbour_index_list:\n",
        "            # get f(x)\n",
        "            f_x_j = predicted[neighbour_index] # model_predict(neighbour.reshape((-1,2,i_o)))\n",
        "            y_distance_list.append(linf_distance(f_x_j, f_x_i))\n",
        "\n",
        "        compute_criticality = [0]\n",
        "        for y_distance in y_distance_list:\n",
        "            if y_distance > max_dist_x:\n",
        "                compute_criticality.append(y_distance)\n",
        "\n",
        "        criticality = sum(compute_criticality)\n",
        "\n",
        "        list_of_criticality_values.append(criticality)\n",
        "    return np.mean(list_of_criticality_values)\n"
      ],
      "metadata": {
        "id": "ozeyQG5nsG2z"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* **The Intrinsic complexity ($IC$)** is implemented similarly as $MC$ with the only difference that $f_{PM} (x_j) = y_j$ instead of $f(x_j)$ = `model.predict`($x_j$). In the following function, this is marked as:         \n",
        "`# THE ONLY DIFFERENCE FROM MODEL COMPLEXITY COMPUTATION`\n"
      ],
      "metadata": {
        "id": "oZwdc3LQsH_q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "def intrinsic_complexity_IC(large_dataset,\n",
        "                i_o,\n",
        "                n,\n",
        "                model_predict,\n",
        "                periodicity,\n",
        "                n_d=3,\n",
        "                n_h=2,\n",
        "                batch_size=32):\n",
        "    \"\"\"\n",
        "    Computes the intrinsic complexity metric using ground truth data.\n",
        "\n",
        "    This function is similar to `model_complexity_MC` but uses the ground truth data as the prediction\n",
        "    from the perfect model. It calculates the intrinsic complexity based on input-output distances.\n",
        "\n",
        "    Args:\n",
        "        large_dataset (np.ndarray): The dataset to compute the complexity on.\n",
        "        i_o (int): The number of frames in input and output.\n",
        "        n (int): The number of data points to consider in the complexity calculation.\n",
        "        periodicity refers to the offset required for 1 day;\n",
        "        n_d=3 corresponds to 1 week of neighbours (3 days look ahead and back; and the current day)\n",
        "        n_h=2 corresponds to 2 hours of neighbours (1 hour look ahead and back)\n",
        "        model_predict (function): The prediction function of the model\n",
        "    \"\"\"\n",
        "\n",
        "\n",
        "    list_of_criticality_values = []\n",
        "\n",
        "    N =  large_dataset.shape[1]\n",
        "\n",
        "    # predict for all data points so that we can process later\n",
        "\n",
        "    predicted = [0] * N\n",
        "    for i in tqdm(range(0, N, batch_size), \"Predicting for all data points\"):\n",
        "        X = []\n",
        "        Y = []  # only differences from MC function. Here we use the ground truth as the prediction\n",
        "                # from the perfect model\n",
        "        for j in range(batch_size):\n",
        "            x,y = dataloader(large_dataset, j, i_o=i_o)\n",
        "            X.append(x)\n",
        "            Y.append(y)\n",
        "        X = np.array(X)\n",
        "\n",
        "\n",
        "        # THE ONLY DIFFERENCE FROM MODEL COMPLEXITY COMPUTATION\n",
        "        predicted[i:i+X.shape[0]] = [predicted for predicted in Y]\n",
        "\n",
        "\n",
        "\n",
        "    for i in tqdm(range(i_o, n), \"Iterating over all \" + str(n) + \" data points\"):\n",
        "\n",
        "        # create list of all neighbours in temporal bands\n",
        "        # at multiples of periodicity\n",
        "        neighbour_list = []\n",
        "        x_distance_list = []\n",
        "        neighbour_index_list = []\n",
        "        x_i, y_i = dataloader(large_dataset, i, i_o=i_o)\n",
        "\n",
        "        f_x_i = predicted[i]   # f(x_i)\n",
        "\n",
        "        for day in range(-n_d, n_d+1):\n",
        "            for hour in range(-n_h, n_h+1):\n",
        "\n",
        "                j = i + day * periodicity + hour\n",
        "\n",
        "                # ignore the tensors which are at the boundaries of the dataset\n",
        "                if j + i_o < 0 or j < 0 or \\\n",
        "                        j >= large_dataset.shape[1] or j+i_o >= large_dataset.shape[1]:\n",
        "                    continue\n",
        "\n",
        "                if j != i:\n",
        "                    x_j, y_j = dataloader(large_dataset, j, i_o=i_o)\n",
        "                    assert (x_j.shape == x_i.shape)\n",
        "                    neighbour_list.append(x_j)\n",
        "                    neighbour_index_list.append(j)\n",
        "                    x_distance_list.append(linf_distance(x_i, x_j))\n",
        "\n",
        "        max_dist_x = np.max(x_distance_list)\n",
        "\n",
        "        y_distance_list = []\n",
        "        for neighbour_index in neighbour_index_list:\n",
        "            f_x_j = predicted[neighbour_index] # model_predict(neighbour.reshape((-1,2,i_o)))\n",
        "            y_distance_list.append(linf_distance(f_x_j, f_x_i))\n",
        "\n",
        "        compute_criticality = [0]\n",
        "        for y_distance in y_distance_list:\n",
        "            if y_distance > max_dist_x:\n",
        "                compute_criticality.append(y_distance)\n",
        "\n",
        "        criticality = sum(compute_criticality)\n",
        "\n",
        "        list_of_criticality_values.append(criticality)\n",
        "    return np.mean(list_of_criticality_values) # computed complexity value of the model\n"
      ],
      "metadata": {
        "id": "3LsI_jJds3EA"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## `Vanilla codes for plotting training and visualising predictions of time series.`"
      ],
      "metadata": {
        "id": "4CXSoEiNs3uM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "def plot_training_curves(model_identifer, history):\n",
        "    \"\"\"\n",
        "    plotting the training and validation loss with time\n",
        "    \"\"\"\n",
        "    loss = history.history[\"loss\"]\n",
        "    val_loss = history.history[\"val_loss\"]\n",
        "\n",
        "    epochs = range(1, len(loss) + 1)\n",
        "\n",
        "    # Plotting\n",
        "    plt.figure(figsize=(8, 4))\n",
        "    plt.plot(epochs, loss, 'tab:blue', label='Training loss')\n",
        "    plt.plot(epochs, val_loss, 'tab:orange', label='Validation loss')\n",
        "    plt.title('Training Loss' + model_identifer)\n",
        "    plt.xlabel('Epochs')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.legend()\n",
        "    plt.savefig(\"plots_from_demo_data/training_curve_\"+ model_identifer +\".jpg\", dpi=300)\n",
        "    plt.show()\n",
        "    plt.clf()\n",
        "\n",
        "def plot_selected_predictions_val_data(val_data, i_o, model_predict, model_identifier):\n",
        "    \"\"\"\n",
        "    plotting selected data points from the validation data to show the prediction performance\n",
        "    \"\"\"\n",
        "    indices = [0, 50, 1400]\n",
        "    ax = [0] * 3\n",
        "    plt.clf()\n",
        "    fig, (ax[0], ax[1], ax[2]) = plt.subplots(3)\n",
        "\n",
        "    for counter, i in enumerate (indices):\n",
        "        x,y = dataloader(val_data, timestamp=i, i_o=i_o)\n",
        "        y_predict = model_predict(x.reshape((-1, n_sensors, i_o)))\n",
        "        ax[counter].plot(x[0, :].flatten().tolist() + y[0, :].flatten().tolist(), label=\"sensor 1 GT\", color=\"tab:blue\",\n",
        "                         linewidth=2)\n",
        "        ax[counter].plot(x[1, :].flatten().tolist() + y[1, :].flatten().tolist(), label=\"sensor 2 GT\", color=\"tab:orange\",\n",
        "                         linewidth=2)\n",
        "\n",
        "        ax[counter].plot(x[0, :].flatten().tolist() + y_predict[0, 0, :].flatten().tolist(), label=\"sensor 1 pred\", color=\"tab:blue\",\n",
        "                         linewidth=0.6)\n",
        "        ax[counter].plot(x[1, :].flatten().tolist() + y_predict[0, 1, :].flatten().tolist(), label=\"sensor 2 pred\", color=\"tab:orange\",\n",
        "                         linewidth=0.6)\n",
        "        ax[counter].set_title(model_identifier)\n",
        "\n",
        "    plt.legend(fontsize=6, loc=\"upper left\")\n",
        "    # plt.title(model_identifier)\n",
        "    plt.tight_layout()\n",
        "    plt.savefig(\"plots_from_demo_data/Predictions_\" + model_identifier + \".jpg\", dpi=300)\n",
        "    plt.show()\n",
        "    plt.clf()"
      ],
      "metadata": {
        "id": "yParFJWatWRW"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "2wjpUC8mekMf"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Driver function (\"`__main__`\")"
      ],
      "metadata": {
        "id": "gx5nqUxOtXrI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "\n",
        "large_dataset, PERIODICITY = generate_continuous_dataset()\n",
        "print (large_dataset.shape)\n",
        "\n",
        "i_o = 7  # Length of Input and output sequences\n",
        "EPOCH = 20\n",
        "n_for_complexity_calculation = 5000\n",
        "\n",
        "TrainX, TrainY, indices_list = sample_blocks_for_XY(large_dataset[:, :-2000], i_o)\n",
        "ValX, ValY,_ = sample_blocks_for_XY(large_dataset[:,-2000:], i_o)\n",
        "\n",
        "\n",
        "\n",
        "model_fc = build_model_fc(i_o=i_o)\n",
        "model_fc.summary()\n",
        "\n",
        "history = model_fc.fit(TrainX, TrainY, epochs=EPOCH, verbose=0, validation_data=\n",
        "                                        [ValX, ValY], batch_size=32)\n",
        "plot_training_curves(model_identifer=\"_fc_\", history=history)\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "metric_value_IC = intrinsic_complexity_IC(large_dataset[:, :8000],\n",
        "                                   i_o,\n",
        "                                   n_for_complexity_calculation,\n",
        "                                   model_fc.predict,\n",
        "                                   periodicity=PERIODICITY,\n",
        "                                   n_d=2,\n",
        "                                   n_h=20,\n",
        "                                   batch_size=32\n",
        "                                   )\n",
        "\n",
        "\n",
        "# Compute the custom metric for one example\n",
        "metric_value_fc = model_complexity_MC(large_dataset[:, :8000],\n",
        "                                   i_o,\n",
        "                                   n_for_complexity_calculation,\n",
        "                                   model_fc.predict,\n",
        "                                   periodicity=PERIODICITY,\n",
        "                                   n_d=2,\n",
        "                                   n_h=20,\n",
        "                                   batch_size=32\n",
        "                                   )\n",
        "\n",
        "\n",
        "\n",
        "predicted_val_Y = model_fc.predict(ValX)\n",
        "assert (predicted_val_Y.shape == ValY.shape)\n",
        "print (\"MSE: (FC) \",  np.mean( (ValY - predicted_val_Y) ** 2 ))\n",
        "\n",
        "plot_selected_predictions_val_data(large_dataset[:, -2000:], i_o=i_o, model_predict=model_fc.predict, model_identifier=\"_FC_\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "DeAU6-H1th-D",
        "outputId": "d27aea9a-283b-4cd6-bc13-cf62b58a1e35"
      },
      "execution_count": 9,
      "outputs": [
        {
          "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": [
            "(3, 10000)\n"
          ]
        },
        {
          "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;34m21\u001b[0m)                  │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)                  │           \u001b[38;5;34m1,408\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_3 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m21\u001b[0m)                  │           \u001b[38;5;34m1,365\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ reshape_1 (\u001b[38;5;33mReshape\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m, \u001b[38;5;34m7\u001b[0m)                │               \u001b[38;5;34m0\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\">21</span>)                  │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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\">64</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,408</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\">21</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,365</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ reshape_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>)                │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m2,773\u001b[0m (10.83 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\">2,773</span> (10.83 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m2,773\u001b[0m (10.83 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\">2,773</span> (10.83 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": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x400 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Predicting for all data points: 100%|██████████| 250/250 [00:00<00:00, 13998.93it/s]\n",
            "Iterating over all 5000 data points: 100%|██████████| 4993/4993 [00:24<00:00, 207.81it/s]\n",
            "Predicting for all data points:   0%|          | 0/250 [00:00<?, ?it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 59ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   0%|          | 1/250 [00:00<00:33,  7.36it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   1%|          | 3/250 [00:00<00:24, 10.22it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   2%|▏         | 4/250 [00:00<00:25,  9.79it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   2%|▏         | 6/250 [00:00<00:22, 10.75it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   3%|▎         | 8/250 [00:00<00:21, 11.45it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   4%|▍         | 10/250 [00:00<00:20, 11.80it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   5%|▍         | 12/250 [00:01<00:20, 11.90it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   6%|▌         | 14/250 [00:01<00:20, 11.63it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   6%|▋         | 16/250 [00:01<00:20, 11.20it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   7%|▋         | 18/250 [00:01<00:20, 11.33it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   8%|▊         | 20/250 [00:01<00:19, 11.73it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   9%|▉         | 22/250 [00:01<00:19, 11.57it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  10%|▉         | 24/250 [00:02<00:19, 11.54it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  10%|█         | 26/250 [00:02<00:19, 11.34it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  11%|█         | 28/250 [00:02<00:21, 10.49it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  12%|█▏        | 30/250 [00:02<00:21, 10.47it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  13%|█▎        | 32/250 [00:02<00:20, 10.71it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  14%|█▎        | 34/250 [00:03<00:19, 10.89it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  14%|█▍        | 36/250 [00:03<00:19, 10.84it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  15%|█▌        | 38/250 [00:03<00:20, 10.38it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  16%|█▌        | 40/250 [00:03<00:19, 10.68it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  17%|█▋        | 42/250 [00:03<00:18, 11.40it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  18%|█▊        | 44/250 [00:03<00:17, 11.60it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  18%|█▊        | 46/250 [00:04<00:17, 11.60it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  19%|█▉        | 48/250 [00:04<00:17, 11.44it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 50ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  20%|██        | 50/250 [00:04<00:18, 10.54it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  21%|██        | 52/250 [00:04<00:18, 10.75it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  22%|██▏       | 54/250 [00:04<00:18, 10.88it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  22%|██▏       | 56/250 [00:05<00:17, 11.03it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  23%|██▎       | 58/250 [00:05<00:17, 10.97it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  24%|██▍       | 60/250 [00:05<00:18, 10.45it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  25%|██▍       | 62/250 [00:05<00:18, 10.19it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  26%|██▌       | 64/250 [00:05<00:17, 10.64it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  26%|██▋       | 66/250 [00:06<00:16, 10.96it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  27%|██▋       | 68/250 [00:06<00:16, 11.09it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  28%|██▊       | 70/250 [00:06<00:16, 11.03it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  29%|██▉       | 72/250 [00:06<00:16, 10.70it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  30%|██▉       | 74/250 [00:06<00:16, 10.97it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  30%|███       | 76/250 [00:06<00:15, 11.23it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  31%|███       | 78/250 [00:07<00:15, 11.29it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  32%|███▏      | 80/250 [00:07<00:14, 11.72it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  33%|███▎      | 82/250 [00:07<00:14, 11.34it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  34%|███▎      | 84/250 [00:07<00:15, 11.01it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  34%|███▍      | 86/250 [00:07<00:14, 11.34it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  35%|███▌      | 88/250 [00:07<00:13, 11.62it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  36%|███▌      | 90/250 [00:08<00:13, 11.82it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  37%|███▋      | 92/250 [00:08<00:13, 12.05it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  38%|███▊      | 94/250 [00:08<00:13, 11.77it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  38%|███▊      | 96/250 [00:08<00:13, 11.82it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  39%|███▉      | 98/250 [00:08<00:12, 12.22it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  40%|████      | 100/250 [00:08<00:12, 12.30it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  41%|████      | 102/250 [00:09<00:12, 12.24it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  42%|████▏     | 104/250 [00:09<00:11, 12.44it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  42%|████▏     | 106/250 [00:09<00:12, 11.83it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  43%|████▎     | 108/250 [00:09<00:12, 11.63it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  44%|████▍     | 110/250 [00:09<00:13, 10.32it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  45%|████▍     | 112/250 [00:10<00:13,  9.92it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  46%|████▌     | 114/250 [00:10<00:13,  9.83it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  46%|████▋     | 116/250 [00:10<00:14,  9.17it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  47%|████▋     | 117/250 [00:10<00:14,  8.98it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  47%|████▋     | 118/250 [00:10<00:14,  8.91it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  48%|████▊     | 119/250 [00:10<00:15,  8.65it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  48%|████▊     | 120/250 [00:11<00:14,  8.80it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  48%|████▊     | 121/250 [00:11<00:14,  8.93it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  49%|████▉     | 122/250 [00:11<00:14,  8.68it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 58ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  49%|████▉     | 123/250 [00:11<00:15,  8.09it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  50%|████▉     | 124/250 [00:11<00:15,  7.91it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  50%|█████     | 125/250 [00:11<00:15,  8.06it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  50%|█████     | 126/250 [00:11<00:16,  7.67it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  51%|█████     | 127/250 [00:11<00:16,  7.69it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  51%|█████     | 128/250 [00:12<00:15,  7.88it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  52%|█████▏    | 129/250 [00:12<00:15,  7.79it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  52%|█████▏    | 130/250 [00:12<00:14,  8.03it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  52%|█████▏    | 131/250 [00:12<00:15,  7.79it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  53%|█████▎    | 132/250 [00:12<00:15,  7.68it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  53%|█████▎    | 133/250 [00:12<00:15,  7.72it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  54%|█████▎    | 134/250 [00:12<00:15,  7.73it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  54%|█████▍    | 135/250 [00:12<00:15,  7.49it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  55%|█████▍    | 137/250 [00:13<00:12,  8.98it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  56%|█████▌    | 139/250 [00:13<00:11,  9.95it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  56%|█████▌    | 140/250 [00:13<00:11,  9.73it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  57%|█████▋    | 142/250 [00:13<00:10, 10.29it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  58%|█████▊    | 144/250 [00:13<00:09, 10.70it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  58%|█████▊    | 146/250 [00:13<00:09, 11.08it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  59%|█████▉    | 148/250 [00:14<00:08, 11.49it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  60%|██████    | 150/250 [00:14<00:08, 11.90it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  61%|██████    | 152/250 [00:14<00:08, 11.42it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  62%|██████▏   | 154/250 [00:14<00:08, 11.24it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  62%|██████▏   | 156/250 [00:14<00:08, 11.47it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  63%|██████▎   | 158/250 [00:14<00:08, 11.14it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  64%|██████▍   | 160/250 [00:15<00:07, 11.57it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  65%|██████▍   | 162/250 [00:15<00:07, 11.57it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 17ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  66%|██████▌   | 164/250 [00:15<00:07, 11.68it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  66%|██████▋   | 166/250 [00:15<00:07, 11.64it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  67%|██████▋   | 168/250 [00:15<00:06, 11.94it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  68%|██████▊   | 170/250 [00:15<00:06, 12.07it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  69%|██████▉   | 172/250 [00:16<00:06, 11.88it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  70%|██████▉   | 174/250 [00:16<00:06, 12.00it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  70%|███████   | 176/250 [00:16<00:06, 11.56it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  71%|███████   | 178/250 [00:16<00:06, 12.00it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  72%|███████▏  | 180/250 [00:16<00:05, 12.21it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  73%|███████▎  | 182/250 [00:16<00:05, 12.05it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  74%|███████▎  | 184/250 [00:17<00:05, 11.86it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  74%|███████▍  | 186/250 [00:17<00:05, 11.80it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  75%|███████▌  | 188/250 [00:17<00:05, 11.34it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  76%|███████▌  | 190/250 [00:17<00:05, 11.51it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  77%|███████▋  | 192/250 [00:17<00:05, 11.30it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  78%|███████▊  | 194/250 [00:18<00:04, 11.26it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  78%|███████▊  | 196/250 [00:18<00:04, 11.38it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  79%|███████▉  | 198/250 [00:18<00:04, 10.92it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  80%|████████  | 200/250 [00:18<00:04, 11.06it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  81%|████████  | 202/250 [00:18<00:04, 11.40it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  82%|████████▏ | 204/250 [00:18<00:04, 11.09it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  82%|████████▏ | 206/250 [00:19<00:04, 10.91it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  83%|████████▎ | 208/250 [00:19<00:03, 11.31it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  84%|████████▍ | 210/250 [00:19<00:03, 10.97it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  85%|████████▍ | 212/250 [00:19<00:03, 11.11it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  86%|████████▌ | 214/250 [00:19<00:03, 11.48it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  86%|████████▋ | 216/250 [00:20<00:03, 11.31it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  87%|████████▋ | 218/250 [00:20<00:02, 11.39it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  88%|████████▊ | 220/250 [00:20<00:02, 11.15it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  89%|████████▉ | 222/250 [00:20<00:02, 10.88it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  90%|████████▉ | 224/250 [00:20<00:02, 10.98it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  90%|█████████ | 226/250 [00:20<00:02, 11.19it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  91%|█████████ | 228/250 [00:21<00:01, 11.14it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  92%|█████████▏| 230/250 [00:21<00:01, 11.17it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  93%|█████████▎| 232/250 [00:21<00:01, 10.78it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  94%|█████████▎| 234/250 [00:21<00:01, 10.72it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  94%|█████████▍| 236/250 [00:21<00:01, 11.12it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  95%|█████████▌| 238/250 [00:22<00:01, 11.10it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  96%|█████████▌| 240/250 [00:22<00:00, 10.75it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  97%|█████████▋| 242/250 [00:22<00:00, 10.66it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  98%|█████████▊| 244/250 [00:22<00:00, 10.58it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  98%|█████████▊| 246/250 [00:22<00:00, 10.96it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  99%|█████████▉| 248/250 [00:22<00:00, 11.32it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Predicting for all data points: 100%|██████████| 250/250 [00:23<00:00, 10.79it/s]\n",
            "Iterating over all 5000 data points: 100%|██████████| 4993/4993 [00:24<00:00, 201.67it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step\n",
            "MSE: (FC)  0.0020682337175226785\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Intrinsic Complexity: \", metric_value_IC)\n",
        "print(\"Model complexity Fully Connected: \", metric_value_fc)\n",
        "print (\"End-to-end Run time of script: \", round(time.time() - starttime, 2))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5-XZO64SwJzS",
        "outputId": "da556dc9-ad43-4f88-861f-5c4db1d716dc"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Intrinsic Complexity:  23.860569139086536\n",
            "Model complexity Fully Connected:  20.978195152539612\n",
            "End-to-end Run time of script:  265.12\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "bHUPPwwVekyf"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "9AQi1mM2THSc"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "RSASh3rNRLTX"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "JKkQ3wfkm2e-"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Same thing for LSTM model"
      ],
      "metadata": {
        "id": "Te_yOY0FtnoD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def build_model_lstm(i_o):\n",
        "    # Calculate the total number of elements in the input (e.g., 2*100 for a 2x100 input)\n",
        "    output_shape = (n_sensors, i_o)  # Adjust this to your desired output shape\n",
        "    total_output_elements = np.prod(output_shape)\n",
        "\n",
        "    model = Sequential([\n",
        "        LSTM(64, input_shape=(n_sensors, i_o), return_sequences=False),\n",
        "        Dense(64, activation='relu'),\n",
        "        Dense(total_output_elements),  # Ensure this matches the total number of elements in the output shape\n",
        "        Reshape(output_shape)  # Reshape the output to the desired shape\n",
        "    ])\n",
        "    model.compile(optimizer='sgd', loss='mse')\n",
        "    return model\n",
        "\n",
        "model_lstm = build_model_lstm(i_o=i_o)\n",
        "model_lstm.summary()\n",
        "history = model_lstm.fit(TrainX, TrainY, epochs=EPOCH, verbose=0, validation_data=\n",
        "                                        [ValX, ValY], batch_size=32)\n",
        "plot_training_curves(model_identifer=\"_lstm_\", history=history)\n",
        "\n",
        "metric_value_lstm = model_complexity_MC(large_dataset[:, :8000],\n",
        "                                   i_o,\n",
        "                                   5000,\n",
        "                                   model_lstm.predict,\n",
        "                                   periodicity=PERIODICITY,\n",
        "                                   n_d=2,\n",
        "                                   n_h=20,\n",
        "                                   batch_size = 32\n",
        "                                    )\n",
        "predicted_val_Y = model_lstm.predict(ValX)\n",
        "assert (predicted_val_Y.shape == ValY.shape)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "apBPCyY5tq0q",
        "outputId": "e539fb91-f4fd-473e-9a41-9278535a5e84"
      },
      "execution_count": 11,
      "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_2\"\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_2\"</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;34m64\u001b[0m)                  │          \u001b[38;5;34m18,432\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_4 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)                  │           \u001b[38;5;34m4,160\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_5 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m21\u001b[0m)                  │           \u001b[38;5;34m1,365\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ reshape_2 (\u001b[38;5;33mReshape\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m, \u001b[38;5;34m7\u001b[0m)                │               \u001b[38;5;34m0\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: #00af00; text-decoration-color: #00af00\">64</span>)                  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,432</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_4 (<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\">64</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">4,160</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_5 (<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\">21</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,365</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ reshape_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>)                │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m23,957\u001b[0m (93.58 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\">23,957</span> (93.58 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m23,957\u001b[0m (93.58 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\">23,957</span> (93.58 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": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x400 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   0%|          | 0/250 [00:00<?, ?it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 352ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   0%|          | 1/250 [00:00<01:49,  2.27it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   1%|          | 2/250 [00:00<01:04,  3.87it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   1%|          | 3/250 [00:00<00:50,  4.92it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   2%|▏         | 4/250 [00:00<00:42,  5.76it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   2%|▏         | 5/250 [00:00<00:38,  6.41it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   2%|▏         | 6/250 [00:01<00:36,  6.77it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   3%|▎         | 7/250 [00:01<00:37,  6.45it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   3%|▎         | 8/250 [00:01<00:35,  6.90it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   4%|▎         | 9/250 [00:01<00:32,  7.48it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   4%|▍         | 10/250 [00:01<00:31,  7.72it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   4%|▍         | 11/250 [00:01<00:30,  7.76it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   5%|▍         | 12/250 [00:01<00:31,  7.49it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   5%|▌         | 13/250 [00:02<00:30,  7.70it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 55ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   6%|▌         | 14/250 [00:02<00:32,  7.18it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   6%|▌         | 15/250 [00:02<00:32,  7.30it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   6%|▋         | 16/250 [00:02<00:31,  7.37it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   7%|▋         | 17/250 [00:02<00:31,  7.46it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   7%|▋         | 18/250 [00:02<00:30,  7.51it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   8%|▊         | 19/250 [00:02<00:30,  7.70it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   8%|▊         | 20/250 [00:02<00:29,  7.83it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   8%|▊         | 21/250 [00:03<00:29,  7.74it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   9%|▉         | 22/250 [00:03<00:31,  7.30it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:   9%|▉         | 23/250 [00:03<00:30,  7.37it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  10%|▉         | 24/250 [00:03<00:31,  7.27it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  10%|█         | 25/250 [00:03<00:30,  7.35it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  10%|█         | 26/250 [00:03<00:28,  7.95it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  11%|█         | 28/250 [00:03<00:23,  9.28it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  12%|█▏        | 30/250 [00:04<00:22,  9.80it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  12%|█▏        | 31/250 [00:04<00:23,  9.36it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  13%|█▎        | 33/250 [00:04<00:21, 10.03it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  14%|█▍        | 35/250 [00:04<00:21, 10.14it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  15%|█▍        | 37/250 [00:04<00:20, 10.47it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  16%|█▌        | 39/250 [00:04<00:19, 10.60it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  16%|█▋        | 41/250 [00:05<00:20, 10.30it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  17%|█▋        | 43/250 [00:05<00:20, 10.09it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  18%|█▊        | 45/250 [00:05<00:19, 10.44it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  19%|█▉        | 47/250 [00:05<00:18, 10.73it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  20%|█▉        | 49/250 [00:05<00:18, 10.98it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  20%|██        | 51/250 [00:06<00:18, 11.02it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  21%|██        | 53/250 [00:06<00:19, 10.14it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  22%|██▏       | 55/250 [00:06<00:18, 10.28it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  23%|██▎       | 57/250 [00:06<00:18, 10.70it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  24%|██▎       | 59/250 [00:06<00:17, 10.86it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  24%|██▍       | 61/250 [00:07<00:17, 10.91it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  25%|██▌       | 63/250 [00:07<00:17, 10.60it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  26%|██▌       | 65/250 [00:07<00:18,  9.98it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  27%|██▋       | 67/250 [00:07<00:17, 10.26it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  28%|██▊       | 69/250 [00:07<00:17, 10.32it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  28%|██▊       | 71/250 [00:08<00:17, 10.25it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  29%|██▉       | 73/250 [00:08<00:17,  9.95it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  30%|███       | 75/250 [00:08<00:17, 10.11it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  31%|███       | 77/250 [00:08<00:17, 10.12it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  32%|███▏      | 79/250 [00:08<00:16, 10.10it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  32%|███▏      | 81/250 [00:09<00:16, 10.33it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  33%|███▎      | 83/250 [00:09<00:16, 10.35it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  34%|███▍      | 85/250 [00:09<00:16, 10.24it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  35%|███▍      | 87/250 [00:09<00:15, 10.67it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  36%|███▌      | 89/250 [00:09<00:15, 10.69it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  36%|███▋      | 91/250 [00:09<00:14, 10.80it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  37%|███▋      | 93/250 [00:10<00:14, 10.68it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  38%|███▊      | 95/250 [00:10<00:14, 10.50it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  39%|███▉      | 97/250 [00:10<00:14, 10.74it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  40%|███▉      | 99/250 [00:10<00:13, 11.11it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  40%|████      | 101/250 [00:10<00:13, 11.39it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  41%|████      | 103/250 [00:11<00:12, 11.36it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  42%|████▏     | 105/250 [00:11<00:13, 11.01it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  43%|████▎     | 107/250 [00:11<00:12, 11.09it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  44%|████▎     | 109/250 [00:11<00:12, 10.96it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  44%|████▍     | 111/250 [00:11<00:12, 11.18it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  45%|████▌     | 113/250 [00:11<00:12, 11.02it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  46%|████▌     | 115/250 [00:12<00:12, 11.13it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  47%|████▋     | 117/250 [00:12<00:12, 10.58it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  48%|████▊     | 119/250 [00:12<00:12, 10.51it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  48%|████▊     | 121/250 [00:12<00:11, 10.94it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  49%|████▉     | 123/250 [00:12<00:11, 11.33it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  50%|█████     | 125/250 [00:13<00:11, 11.12it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  51%|█████     | 127/250 [00:13<00:11, 10.77it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  52%|█████▏    | 129/250 [00:13<00:11, 10.60it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  52%|█████▏    | 131/250 [00:13<00:11, 10.63it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  53%|█████▎    | 133/250 [00:13<00:11, 10.29it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  54%|█████▍    | 135/250 [00:14<00:11,  9.62it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  54%|█████▍    | 136/250 [00:14<00:12,  9.19it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  55%|█████▍    | 137/250 [00:14<00:13,  8.50it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  55%|█████▌    | 138/250 [00:14<00:13,  8.51it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  56%|█████▌    | 139/250 [00:14<00:13,  8.31it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  56%|█████▌    | 140/250 [00:14<00:13,  8.41it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  56%|█████▋    | 141/250 [00:14<00:13,  8.22it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  57%|█████▋    | 142/250 [00:14<00:13,  8.25it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  57%|█████▋    | 143/250 [00:15<00:12,  8.55it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  58%|█████▊    | 144/250 [00:15<00:13,  8.03it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  58%|█████▊    | 145/250 [00:15<00:12,  8.14it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  58%|█████▊    | 146/250 [00:15<00:13,  7.99it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  59%|█████▉    | 147/250 [00:15<00:12,  8.13it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  59%|█████▉    | 148/250 [00:15<00:13,  7.79it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  60%|█████▉    | 149/250 [00:15<00:12,  7.79it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  60%|██████    | 150/250 [00:15<00:13,  7.68it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  60%|██████    | 151/250 [00:16<00:12,  7.82it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  61%|██████    | 152/250 [00:16<00:12,  7.67it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  61%|██████    | 153/250 [00:16<00:12,  7.79it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  62%|██████▏   | 154/250 [00:16<00:12,  7.98it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  62%|██████▏   | 155/250 [00:16<00:11,  8.04it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  62%|██████▏   | 156/250 [00:16<00:11,  7.87it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  63%|██████▎   | 157/250 [00:16<00:11,  8.08it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  63%|██████▎   | 158/250 [00:16<00:11,  7.98it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  64%|██████▎   | 159/250 [00:17<00:11,  8.19it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  64%|██████▍   | 160/250 [00:17<00:11,  7.82it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  64%|██████▍   | 161/250 [00:17<00:10,  8.22it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  65%|██████▌   | 163/250 [00:17<00:09,  9.21it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  66%|██████▌   | 165/250 [00:17<00:08, 10.13it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  67%|██████▋   | 167/250 [00:17<00:07, 10.57it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  68%|██████▊   | 169/250 [00:18<00:07, 10.80it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  68%|██████▊   | 171/250 [00:18<00:07, 10.66it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  69%|██████▉   | 173/250 [00:18<00:07, 10.52it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 20ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  70%|███████   | 175/250 [00:18<00:07, 10.59it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  71%|███████   | 177/250 [00:18<00:06, 10.61it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  72%|███████▏  | 179/250 [00:18<00:06, 10.92it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  72%|███████▏  | 181/250 [00:19<00:06, 10.99it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  73%|███████▎  | 183/250 [00:19<00:05, 11.18it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  74%|███████▍  | 185/250 [00:19<00:05, 11.07it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  75%|███████▍  | 187/250 [00:19<00:05, 11.04it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  76%|███████▌  | 189/250 [00:19<00:05, 11.05it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  76%|███████▋  | 191/250 [00:20<00:05, 10.83it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  77%|███████▋  | 193/250 [00:20<00:05, 10.46it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  78%|███████▊  | 195/250 [00:20<00:05, 10.41it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  79%|███████▉  | 197/250 [00:20<00:07,  7.21it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  79%|███████▉  | 198/250 [00:21<00:06,  7.56it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  80%|████████  | 200/250 [00:21<00:06,  8.20it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  81%|████████  | 202/250 [00:21<00:05,  8.90it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  82%|████████▏ | 204/250 [00:21<00:04,  9.47it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  82%|████████▏ | 206/250 [00:21<00:04,  9.90it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  83%|████████▎ | 208/250 [00:21<00:04,  9.84it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  84%|████████▍ | 210/250 [00:22<00:04,  9.55it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  85%|████████▍ | 212/250 [00:22<00:03,  9.79it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  85%|████████▌ | 213/250 [00:22<00:03,  9.82it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  86%|████████▌ | 215/250 [00:22<00:03, 10.31it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  87%|████████▋ | 217/250 [00:22<00:03, 10.35it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  88%|████████▊ | 219/250 [00:23<00:02, 10.86it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  88%|████████▊ | 221/250 [00:23<00:02, 10.47it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  89%|████████▉ | 223/250 [00:23<00:02, 10.48it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  90%|█████████ | 225/250 [00:23<00:02, 10.49it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  91%|█████████ | 227/250 [00:23<00:02, 10.71it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  92%|█████████▏| 229/250 [00:23<00:01, 10.74it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  92%|█████████▏| 231/250 [00:24<00:01, 10.56it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  93%|█████████▎| 233/250 [00:24<00:01, 10.42it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  94%|█████████▍| 235/250 [00:24<00:01, 10.70it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  95%|█████████▍| 237/250 [00:24<00:01, 10.95it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  96%|█████████▌| 239/250 [00:24<00:01, 10.87it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  96%|█████████▋| 241/250 [00:25<00:00, 11.08it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  97%|█████████▋| 243/250 [00:25<00:00, 10.56it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  98%|█████████▊| 245/250 [00:25<00:00, 10.67it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points:  99%|█████████▉| 247/250 [00:25<00:00, 10.76it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 21ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\rPredicting for all data points: 100%|█████████▉| 249/250 [00:25<00:00, 10.86it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Predicting for all data points: 100%|██████████| 250/250 [00:25<00:00,  9.63it/s]\n",
            "Iterating over all 5000 data points: 100%|██████████| 4993/4993 [00:25<00:00, 193.78it/s]\n"
          ]
        },
        {
          "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"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_selected_predictions_val_data(large_dataset[:, -2000:], i_o=i_o, model_predict=model_lstm.predict, model_identifier=\"_LSTM_\")\n",
        "\n",
        "print(\"Intrinsic Complexity: \", metric_value_IC)\n",
        "print (\"MSE: (LSTM) \",  np.mean( (ValY - predicted_val_Y) ** 2 ))\n",
        "print(\"Model complexity LSTM: \", metric_value_lstm)\n"
      ],
      "metadata": {
        "id": "jlMSdplUvj8L",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 655
        },
        "outputId": "3f6bd50c-0660-4032-e29c-98c2c5f97132"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Intrinsic Complexity:  23.860569139086536\n",
            "MSE: (LSTM)  0.004078882460478604\n",
            "Model complexity LSTM:  16.47713051824992\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "p3WbItlNZH5Y"
      },
      "execution_count": 12,
      "outputs": []
    }
  ]
}