MomentNetwork_demo/MomentNetwork_grav_wave_demo.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MomentNetwork_grav_wave_demo.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOGnhF6/JCF8L/P9iOIw4Sl",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/NiallJeffrey/MomentNetworks/blob/master/MomentNetwork_demo/MomentNetwork_grav_wave_demo.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LNlAR4PJj22u"
      },
      "source": [
        "# This is a Moment Network example with gravitational wave signals\n",
        "## Using simulated gravitational wave data to estimate the moments of the marginal posterior for the strain $h_+$ at each time step (these would form a $>100$-dimensional posterior distribution) \n",
        "## Here we use a dense neural network as a demo, though a CNN would work well for this data and can be easily used in the Moment Network hierarchy."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3GQCvYHXIFVu",
        "outputId": "f909e2b1-2b04-4b36-da88-066567c2685e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "import sys\n",
        "!{sys.executable} -m pip install pycbc lalsuite ligo-common --no-cache-dir\n",
        "!pip install 'git+https://github.com/NiallJeffrey/MomentNetworks.git'"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: pycbc in /usr/local/lib/python3.6/dist-packages (1.16.12)\n",
            "Requirement already satisfied: lalsuite in /usr/local/lib/python3.6/dist-packages (6.78)\n",
            "Requirement already satisfied: ligo-common in /usr/local/lib/python3.6/dist-packages (1.0.3)\n",
            "Requirement already satisfied: decorator>=3.4.2 in /usr/local/lib/python3.6/dist-packages (from pycbc) (4.4.2)\n",
            "Requirement already satisfied: gwdatafind in /usr/local/lib/python3.6/dist-packages (from pycbc) (1.0.4)\n",
            "Requirement already satisfied: h5py>=2.5 in /usr/local/lib/python3.6/dist-packages (from pycbc) (2.10.0)\n",
            "Requirement already satisfied: numpy<1.19,>=1.16.0; python_version >= \"3.5\" in /usr/local/lib/python3.6/dist-packages (from pycbc) (1.18.5)\n",
            "Requirement already satisfied: scipy>=0.16.0; python_version >= \"3.5\" in /usr/local/lib/python3.6/dist-packages (from pycbc) (1.4.1)\n",
            "Requirement already satisfied: cython>=0.29 in /usr/local/lib/python3.6/dist-packages (from pycbc) (0.29.21)\n",
            "Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from pycbc) (1.15.0)\n",
            "Requirement already satisfied: ligo-segments in /usr/local/lib/python3.6/dist-packages (from pycbc) (1.2.0)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.6/dist-packages (from pycbc) (2.11.2)\n",
            "Requirement already satisfied: lscsoft-glue>=1.59.3 in /usr/local/lib/python3.6/dist-packages (from pycbc) (2.0.0)\n",
            "Requirement already satisfied: beautifulsoup4>=4.6.0 in /usr/local/lib/python3.6/dist-packages (from pycbc) (4.6.3)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.6/dist-packages (from pycbc) (4.41.1)\n",
            "Requirement already satisfied: pillow in /usr/local/lib/python3.6/dist-packages (from pycbc) (7.0.0)\n",
            "Requirement already satisfied: astropy>=2.0.3; python_version > \"3.0\" in /usr/local/lib/python3.6/dist-packages (from pycbc) (4.1)\n",
            "Requirement already satisfied: requests>=1.2.1 in /usr/local/lib/python3.6/dist-packages (from pycbc) (2.23.0)\n",
            "Requirement already satisfied: matplotlib>=1.5.1 in /usr/local/lib/python3.6/dist-packages (from pycbc) (3.2.2)\n",
            "Requirement already satisfied: Mako>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from pycbc) (1.1.3)\n",
            "Requirement already satisfied: mpld3>=0.3 in /usr/local/lib/python3.6/dist-packages (from pycbc) (0.5.1)\n",
            "Requirement already satisfied: emcee==2.2.1 in /usr/local/lib/python3.6/dist-packages (from pycbc) (2.2.1)\n",
            "Requirement already satisfied: python-dateutil in /usr/local/lib/python3.6/dist-packages (from lalsuite) (2.8.1)\n",
            "Requirement already satisfied: importlib-resources; python_version < \"3.7\" in /usr/local/lib/python3.6/dist-packages (from lalsuite) (3.3.0)\n",
            "Requirement already satisfied: pyOpenSSL in /usr/local/lib/python3.6/dist-packages (from gwdatafind->pycbc) (19.1.0)\n",
            "Requirement already satisfied: MarkupSafe>=0.23 in /usr/local/lib/python3.6/dist-packages (from jinja2->pycbc) (1.1.1)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests>=1.2.1->pycbc) (3.0.4)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests>=1.2.1->pycbc) (2020.6.20)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests>=1.2.1->pycbc) (2.10)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests>=1.2.1->pycbc) (1.24.3)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.5.1->pycbc) (0.10.0)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.5.1->pycbc) (1.3.1)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.5.1->pycbc) (2.4.7)\n",
            "Requirement already satisfied: zipp>=0.4; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from importlib-resources; python_version < \"3.7\"->lalsuite) (3.4.0)\n",
            "Requirement already satisfied: cryptography>=2.8 in /usr/local/lib/python3.6/dist-packages (from pyOpenSSL->gwdatafind->pycbc) (3.2.1)\n",
            "Requirement already satisfied: cffi!=1.11.3,>=1.8 in /usr/local/lib/python3.6/dist-packages (from cryptography>=2.8->pyOpenSSL->gwdatafind->pycbc) (1.14.3)\n",
            "Requirement already satisfied: pycparser in /usr/local/lib/python3.6/dist-packages (from cffi!=1.11.3,>=1.8->cryptography>=2.8->pyOpenSSL->gwdatafind->pycbc) (2.20)\n",
            "Collecting git+https://github.com/NiallJeffrey/MomentNetworks.git\n",
            "  Cloning https://github.com/NiallJeffrey/MomentNetworks.git to /tmp/pip-req-build-yb1wq3c2\n",
            "  Running command git clone -q https://github.com/NiallJeffrey/MomentNetworks.git /tmp/pip-req-build-yb1wq3c2\n",
            "Requirement already satisfied (use --upgrade to upgrade): momentnetworks==0.1 from git+https://github.com/NiallJeffrey/MomentNetworks.git in /usr/local/lib/python3.6/dist-packages\n",
            "Requirement already satisfied: emcee in /usr/local/lib/python3.6/dist-packages (from momentnetworks==0.1) (2.2.1)\n",
            "Requirement already satisfied: getdist in /usr/local/lib/python3.6/dist-packages (from momentnetworks==0.1) (1.1.2)\n",
            "Requirement already satisfied: chainconsumer in /usr/local/lib/python3.6/dist-packages (from momentnetworks==0.1) (0.32.0)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from momentnetworks==0.1) (1.4.1)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.6/dist-packages (from momentnetworks==0.1) (3.2.2)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from emcee->momentnetworks==0.1) (1.18.5)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.6/dist-packages (from chainconsumer->momentnetworks==0.1) (1.1.4)\n",
            "Requirement already satisfied: statsmodels>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from chainconsumer->momentnetworks==0.1) (0.10.2)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->momentnetworks==0.1) (1.3.1)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->momentnetworks==0.1) (2.8.1)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->momentnetworks==0.1) (2.4.7)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib->momentnetworks==0.1) (0.10.0)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas->chainconsumer->momentnetworks==0.1) (2018.9)\n",
            "Requirement already satisfied: patsy>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from statsmodels>=0.7.0->chainconsumer->momentnetworks==0.1) (0.5.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.6/dist-packages (from python-dateutil>=2.1->matplotlib->momentnetworks==0.1) (1.15.0)\n",
            "Building wheels for collected packages: momentnetworks\n",
            "  Building wheel for momentnetworks (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for momentnetworks: filename=momentnetworks-0.1-cp36-none-any.whl size=4190 sha256=45c94e377e88599cf3be2852a88b2f5f63d822d0d2d310769075508b33d1d7e4\n",
            "  Stored in directory: /tmp/pip-ephem-wheel-cache-oa5e_t1v/wheels/d5/90/42/833efe39b17eb037678ff241c8946e912987034b18c69e2813\n",
            "Successfully built momentnetworks\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mu9TTIBUIG_q"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import time\n",
        "from scipy import signal\n",
        "from google.colab import files\n",
        "from pycbc.waveform import get_td_waveform\n",
        "import pycbc.noise\n",
        "import pycbc.psd\n",
        "import random\n",
        "from momentnetworks import demo\n",
        "import gc"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iS8237oGdeTm"
      },
      "source": [
        "# Generate simulated waveforms and noisy data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3u6MvpXnIIzm"
      },
      "source": [
        "%matplotlib inline"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tj28euIwIP6f",
        "outputId": "84051ade-5d3f-4162-d70a-ffce6b7b445d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "n_training = 5000\n",
        "n_training_samples = n_training*16\n",
        "print('Number of training samples = '+str(n_training_samples))"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of training samples = 80000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9S0U1AbRPL3s",
        "outputId": "5870c201-4cb7-487d-efd2-7ba990e7356e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "mass_1_samples = np.random.uniform(10,30,n_training)\n",
        "mass_2_samples = np.random.uniform(10,30,n_training)\n",
        "distance_samples = np.random.uniform(500,1500,n_training)\n",
        "sim_data, sim_data_noisy = demo.grav_wave_series(n_training,\n",
        "                                                 mass_1_samples=mass_1_samples,\n",
        "                                                 mass_2_samples=mass_2_samples,\n",
        "                                                 distance_samples=distance_samples)\n",
        "noise_alone = sim_data_noisy - sim_data"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0/5000\n",
            "200/5000\n",
            "400/5000\n",
            "600/5000\n",
            "800/5000\n",
            "1000/5000\n",
            "1200/5000\n",
            "1400/5000\n",
            "1600/5000\n",
            "1800/5000\n",
            "2000/5000\n",
            "2200/5000\n",
            "2400/5000\n",
            "2600/5000\n",
            "2800/5000\n",
            "3000/5000\n",
            "3200/5000\n",
            "3400/5000\n",
            "3600/5000\n",
            "3800/5000\n",
            "4000/5000\n",
            "4200/5000\n",
            "4400/5000\n",
            "4600/5000\n",
            "4800/5000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NUVEjB4kIY7u",
        "outputId": "37d9736f-f93f-472d-da70-d08d3a466b89",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 288
        }
      },
      "source": [
        "_ =plt.figure(figsize=(14,4))\n",
        "_ = plt.subplot(1,2,1)\n",
        "_ = plt.ylabel(r'$h_+$ strain',fontsize=14)\n",
        "_ =plt.plot(np.linspace(0,1/16.,128)-1/8.,sim_data_noisy[63,:], alpha=0.5), plt.ylim(-.8,0.8)\n",
        "_ =plt.plot(np.linspace(0,1/16.,128)-1/8.,sim_data[63,:], 'C1--'), plt.xlabel(r'$t\\ /$ seconds',fontsize=14)\n",
        "_ = plt.subplot(1,2,2)\n",
        "_ = plt.ylabel(r'$h_+$ strain',fontsize=14)\n",
        "_ =plt.plot(np.linspace(0,1/16.,128)-1/16.,sim_data_noisy[63+int(n_training*8),:], alpha=0.5), plt.ylim(-0.8,0.8)\n",
        "_ =plt.plot(np.linspace(0,1/16.,128)-1/16.,sim_data[63+int(n_training*8),:], 'C1--'), plt.xlabel(r'$t\\ /$ seconds',fontsize=14)\n"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "veadT_Gndkcy"
      },
      "source": [
        "# shuffling is important, otherwise the mergers are all in the second half of the array\n",
        "shuffled_indices = np.arange(sim_data_noisy.shape[0])\n",
        "np.random.shuffle(shuffled_indices)\n",
        "sim_data_noisy= sim_data_noisy[shuffled_indices]\n",
        "sim_data= sim_data[shuffled_indices]"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WzApqub6httq",
        "outputId": "07209974-8111-4fcf-dd75-a588df60fd04",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 288
        }
      },
      "source": [
        "_ =plt.figure(figsize=(14,4))\n",
        "_ = plt.subplot(1,2,1)\n",
        "_ = plt.ylabel(r'$h_+$ strain',fontsize=14)\n",
        "_ =plt.plot(np.linspace(0,1/16.,128)-1/8.,sim_data_noisy[63,:], alpha=0.5), plt.ylim(-.8,0.8)\n",
        "_ =plt.plot(np.linspace(0,1/16.,128)-1/8.,sim_data[63,:], 'C1--'), plt.xlabel(r'$t\\ /$ seconds',fontsize=14)\n",
        "_ = plt.subplot(1,2,2)\n",
        "_ = plt.ylabel(r'$h_+$ strain',fontsize=14)\n",
        "_ =plt.plot(np.linspace(0,1/16.,128)-1/16.,sim_data_noisy[63+int(n_training*8),:], alpha=0.5), plt.ylim(-0.8,0.8)\n",
        "_ =plt.plot(np.linspace(0,1/16.,128)-1/16.,sim_data[63+int(n_training*8),:], 'C1--'), plt.xlabel(r'$t\\ /$ seconds',fontsize=14)\n"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H_d2FftXdnLp"
      },
      "source": [
        "# Train moment network\n",
        "## As long as the validation loss is decreasing, the estimate of the moment of the marginal posterior is improving"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mczQ-AcrL7HN",
        "outputId": "0795d95c-4edf-462f-a375-4bea021c6708",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "gc.collect()\n",
        "validation_index = np.int(n_training * 7)\n",
        "model_instance = demo.simple_leaky(128,128, learning_rate=1e-4)\n",
        "regression = model_instance.model()\n",
        "\n",
        "history = regression.fit(sim_data_noisy[:validation_index]+0.5,\n",
        "                               sim_data[:validation_index]+0.5,\n",
        "                               epochs =500,batch_size =512, shuffle=True,\n",
        "                               validation_data = (sim_data_noisy[validation_index:]+0.5,\n",
        "                               sim_data[validation_index:]+0.5))"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "128\n",
            "Model: \"functional_11\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_6 (InputLayer)         [(None, 128)]             0         \n",
            "_________________________________________________________________\n",
            "dense_16 (Dense)             (None, 128)               16512     \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_11 (LeakyReLU)   (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "dense_17 (Dense)             (None, 256)               33024     \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_12 (LeakyReLU)   (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "dense_18 (Dense)             (None, 128)               32896     \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_13 (LeakyReLU)   (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "dense_19 (Dense)             (None, 128)               16512     \n",
            "=================================================================\n",
            "Total params: 98,944\n",
            "Trainable params: 98,944\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 0.1735 - val_loss: 0.0497\n",
            "Epoch 2/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0225 - val_loss: 0.0152\n",
            "Epoch 3/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0149 - val_loss: 0.0148\n",
            "Epoch 4/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0146 - val_loss: 0.0145\n",
            "Epoch 5/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0142 - val_loss: 0.0139\n",
            "Epoch 6/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0133 - val_loss: 0.0128\n",
            "Epoch 7/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0120 - val_loss: 0.0113\n",
            "Epoch 8/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0103 - val_loss: 0.0096\n",
            "Epoch 9/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0088 - val_loss: 0.0082\n",
            "Epoch 10/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0076 - val_loss: 0.0072\n",
            "Epoch 11/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0067 - val_loss: 0.0064\n",
            "Epoch 12/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0060 - val_loss: 0.0058\n",
            "Epoch 13/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0055 - val_loss: 0.0053\n",
            "Epoch 14/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0050 - val_loss: 0.0049\n",
            "Epoch 15/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0046 - val_loss: 0.0045\n",
            "Epoch 16/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0043 - val_loss: 0.0042\n",
            "Epoch 17/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0040 - val_loss: 0.0039\n",
            "Epoch 18/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0037 - val_loss: 0.0036\n",
            "Epoch 19/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0035 - val_loss: 0.0034\n",
            "Epoch 20/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0033 - val_loss: 0.0033\n",
            "Epoch 21/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0032 - val_loss: 0.0031\n",
            "Epoch 22/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0030 - val_loss: 0.0030\n",
            "Epoch 23/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0029 - val_loss: 0.0029\n",
            "Epoch 24/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0028 - val_loss: 0.0028\n",
            "Epoch 25/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0027 - val_loss: 0.0027\n",
            "Epoch 26/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0026 - val_loss: 0.0026\n",
            "Epoch 27/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0025 - val_loss: 0.0025\n",
            "Epoch 28/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0025 - val_loss: 0.0025\n",
            "Epoch 29/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0024 - val_loss: 0.0024\n",
            "Epoch 30/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0024 - val_loss: 0.0023\n",
            "Epoch 31/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0023 - val_loss: 0.0023\n",
            "Epoch 32/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0023 - val_loss: 0.0022\n",
            "Epoch 33/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0022 - val_loss: 0.0022\n",
            "Epoch 34/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0022 - val_loss: 0.0022\n",
            "Epoch 35/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0021 - val_loss: 0.0021\n",
            "Epoch 36/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0021 - val_loss: 0.0021\n",
            "Epoch 37/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0021 - val_loss: 0.0021\n",
            "Epoch 38/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0021 - val_loss: 0.0020\n",
            "Epoch 39/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0020 - val_loss: 0.0020\n",
            "Epoch 40/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0020 - val_loss: 0.0020\n",
            "Epoch 41/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0020 - val_loss: 0.0020\n",
            "Epoch 42/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0020 - val_loss: 0.0019\n",
            "Epoch 43/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0019 - val_loss: 0.0019\n",
            "Epoch 44/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0019 - val_loss: 0.0019\n",
            "Epoch 45/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0019 - val_loss: 0.0019\n",
            "Epoch 46/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0019 - val_loss: 0.0019\n",
            "Epoch 47/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0018 - val_loss: 0.0018\n",
            "Epoch 48/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 0.0018 - val_loss: 0.0018\n",
            "Epoch 49/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0018 - val_loss: 0.0018\n",
            "Epoch 50/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0018 - val_loss: 0.0018\n",
            "Epoch 51/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0017 - val_loss: 0.0017\n",
            "Epoch 52/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0017 - val_loss: 0.0017\n",
            "Epoch 53/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0017 - val_loss: 0.0017\n",
            "Epoch 54/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0017 - val_loss: 0.0017\n",
            "Epoch 55/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0017 - val_loss: 0.0016\n",
            "Epoch 56/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0016 - val_loss: 0.0016\n",
            "Epoch 57/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0016 - val_loss: 0.0016\n",
            "Epoch 58/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0016 - val_loss: 0.0016\n",
            "Epoch 59/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0016 - val_loss: 0.0016\n",
            "Epoch 60/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0016 - val_loss: 0.0016\n",
            "Epoch 61/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0015 - val_loss: 0.0015\n",
            "Epoch 62/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0015 - val_loss: 0.0015\n",
            "Epoch 63/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0015 - val_loss: 0.0015\n",
            "Epoch 64/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0015 - val_loss: 0.0015\n",
            "Epoch 65/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0015 - val_loss: 0.0015\n",
            "Epoch 66/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0015 - val_loss: 0.0015\n",
            "Epoch 67/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0014\n",
            "Epoch 68/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0014\n",
            "Epoch 69/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0014\n",
            "Epoch 70/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0014\n",
            "Epoch 71/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0014\n",
            "Epoch 72/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0014\n",
            "Epoch 73/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0014\n",
            "Epoch 74/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0014\n",
            "Epoch 75/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0014 - val_loss: 0.0013\n",
            "Epoch 76/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 77/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 78/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 79/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 80/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 81/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 82/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 83/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 84/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 85/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 86/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0013 - val_loss: 0.0013\n",
            "Epoch 87/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0013\n",
            "Epoch 88/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 89/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 90/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 91/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 92/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 93/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 94/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 95/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 96/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 97/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 98/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 99/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 100/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 101/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 102/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0012 - val_loss: 0.0012\n",
            "Epoch 103/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0012\n",
            "Epoch 104/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0012\n",
            "Epoch 105/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0012\n",
            "Epoch 106/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 107/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0012\n",
            "Epoch 108/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 109/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 110/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 111/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 112/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 113/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 114/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 115/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 116/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 117/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 118/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 119/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 120/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 121/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 122/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 123/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 124/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 125/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 0.0011 - val_loss: 0.0011\n",
            "Epoch 126/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0011\n",
            "Epoch 127/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0011\n",
            "Epoch 128/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0011\n",
            "Epoch 129/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0011\n",
            "Epoch 130/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0011\n",
            "Epoch 131/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0011\n",
            "Epoch 132/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0011\n",
            "Epoch 133/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0010\n",
            "Epoch 134/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 0.0010 - val_loss: 0.0011\n",
            "Epoch 135/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0010\n",
            "Epoch 136/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0010\n",
            "Epoch 137/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0010\n",
            "Epoch 138/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0010\n",
            "Epoch 139/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 0.0010 - val_loss: 0.0010\n",
            "Epoch 140/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.9831e-04 - val_loss: 0.0010\n",
            "Epoch 141/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.9430e-04 - val_loss: 0.0010\n",
            "Epoch 142/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.9518e-04 - val_loss: 0.0010\n",
            "Epoch 143/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.9130e-04 - val_loss: 0.0010\n",
            "Epoch 144/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.8928e-04 - val_loss: 0.0010\n",
            "Epoch 145/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.8471e-04 - val_loss: 0.0010\n",
            "Epoch 146/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.8457e-04 - val_loss: 0.0010\n",
            "Epoch 147/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.8138e-04 - val_loss: 9.9756e-04\n",
            "Epoch 148/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.7667e-04 - val_loss: 0.0010\n",
            "Epoch 149/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.7421e-04 - val_loss: 9.9473e-04\n",
            "Epoch 150/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.7124e-04 - val_loss: 9.9459e-04\n",
            "Epoch 151/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.7179e-04 - val_loss: 9.9566e-04\n",
            "Epoch 152/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.6636e-04 - val_loss: 9.9353e-04\n",
            "Epoch 153/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.6424e-04 - val_loss: 9.8897e-04\n",
            "Epoch 154/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.6209e-04 - val_loss: 9.8925e-04\n",
            "Epoch 155/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.6071e-04 - val_loss: 9.8283e-04\n",
            "Epoch 156/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.5540e-04 - val_loss: 9.7851e-04\n",
            "Epoch 157/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.5420e-04 - val_loss: 9.8373e-04\n",
            "Epoch 158/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.5257e-04 - val_loss: 9.9110e-04\n",
            "Epoch 159/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.5307e-04 - val_loss: 9.7336e-04\n",
            "Epoch 160/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.4556e-04 - val_loss: 9.7209e-04\n",
            "Epoch 161/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.4736e-04 - val_loss: 9.6968e-04\n",
            "Epoch 162/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.3958e-04 - val_loss: 9.7039e-04\n",
            "Epoch 163/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.4211e-04 - val_loss: 9.7341e-04\n",
            "Epoch 164/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.3782e-04 - val_loss: 9.6681e-04\n",
            "Epoch 165/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.3427e-04 - val_loss: 9.6525e-04\n",
            "Epoch 166/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.3408e-04 - val_loss: 9.6123e-04\n",
            "Epoch 167/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.3359e-04 - val_loss: 9.6178e-04\n",
            "Epoch 168/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.2999e-04 - val_loss: 9.7179e-04\n",
            "Epoch 169/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.2697e-04 - val_loss: 9.5397e-04\n",
            "Epoch 170/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.2411e-04 - val_loss: 9.5093e-04\n",
            "Epoch 171/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.2198e-04 - val_loss: 9.4822e-04\n",
            "Epoch 172/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.1875e-04 - val_loss: 9.5054e-04\n",
            "Epoch 173/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.1910e-04 - val_loss: 9.4663e-04\n",
            "Epoch 174/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.1477e-04 - val_loss: 9.5109e-04\n",
            "Epoch 175/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.1233e-04 - val_loss: 9.4850e-04\n",
            "Epoch 176/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.1011e-04 - val_loss: 9.4121e-04\n",
            "Epoch 177/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.0971e-04 - val_loss: 9.4490e-04\n",
            "Epoch 178/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.0922e-04 - val_loss: 9.3956e-04\n",
            "Epoch 179/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.0518e-04 - val_loss: 9.3629e-04\n",
            "Epoch 180/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.0360e-04 - val_loss: 9.3246e-04\n",
            "Epoch 181/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 9.0344e-04 - val_loss: 9.3569e-04\n",
            "Epoch 182/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.9772e-04 - val_loss: 9.2854e-04\n",
            "Epoch 183/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.9737e-04 - val_loss: 9.2956e-04\n",
            "Epoch 184/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.9448e-04 - val_loss: 9.4317e-04\n",
            "Epoch 185/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.9780e-04 - val_loss: 9.2341e-04\n",
            "Epoch 186/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.9229e-04 - val_loss: 9.2684e-04\n",
            "Epoch 187/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.9184e-04 - val_loss: 9.2035e-04\n",
            "Epoch 188/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.8586e-04 - val_loss: 9.3643e-04\n",
            "Epoch 189/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.8621e-04 - val_loss: 9.1481e-04\n",
            "Epoch 190/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.8475e-04 - val_loss: 9.1968e-04\n",
            "Epoch 191/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.8151e-04 - val_loss: 9.1510e-04\n",
            "Epoch 192/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.8091e-04 - val_loss: 9.1346e-04\n",
            "Epoch 193/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.8137e-04 - val_loss: 9.1661e-04\n",
            "Epoch 194/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.7652e-04 - val_loss: 9.0832e-04\n",
            "Epoch 195/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.7639e-04 - val_loss: 9.1294e-04\n",
            "Epoch 196/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.7148e-04 - val_loss: 9.1631e-04\n",
            "Epoch 197/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.7121e-04 - val_loss: 9.0490e-04\n",
            "Epoch 198/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.6902e-04 - val_loss: 9.1684e-04\n",
            "Epoch 199/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.7127e-04 - val_loss: 9.0295e-04\n",
            "Epoch 200/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.6730e-04 - val_loss: 9.1576e-04\n",
            "Epoch 201/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.6592e-04 - val_loss: 9.0165e-04\n",
            "Epoch 202/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.6561e-04 - val_loss: 9.0021e-04\n",
            "Epoch 203/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.6137e-04 - val_loss: 9.0075e-04\n",
            "Epoch 204/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.6030e-04 - val_loss: 8.9364e-04\n",
            "Epoch 205/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.6010e-04 - val_loss: 8.9348e-04\n",
            "Epoch 206/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.5731e-04 - val_loss: 8.9372e-04\n",
            "Epoch 207/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.5854e-04 - val_loss: 8.8994e-04\n",
            "Epoch 208/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.5468e-04 - val_loss: 8.8718e-04\n",
            "Epoch 209/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.5177e-04 - val_loss: 8.9020e-04\n",
            "Epoch 210/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.4924e-04 - val_loss: 8.8647e-04\n",
            "Epoch 211/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.4939e-04 - val_loss: 8.8684e-04\n",
            "Epoch 212/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.4988e-04 - val_loss: 8.8252e-04\n",
            "Epoch 213/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.4567e-04 - val_loss: 8.8011e-04\n",
            "Epoch 214/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.4586e-04 - val_loss: 8.9111e-04\n",
            "Epoch 215/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.4471e-04 - val_loss: 8.7684e-04\n",
            "Epoch 216/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.4491e-04 - val_loss: 8.9644e-04\n",
            "Epoch 217/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.4079e-04 - val_loss: 8.7753e-04\n",
            "Epoch 218/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.3960e-04 - val_loss: 8.7506e-04\n",
            "Epoch 219/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.3649e-04 - val_loss: 8.8520e-04\n",
            "Epoch 220/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.4165e-04 - val_loss: 8.7349e-04\n",
            "Epoch 221/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.3767e-04 - val_loss: 8.8467e-04\n",
            "Epoch 222/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.3562e-04 - val_loss: 8.9949e-04\n",
            "Epoch 223/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.3356e-04 - val_loss: 8.6846e-04\n",
            "Epoch 224/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.2935e-04 - val_loss: 8.7157e-04\n",
            "Epoch 225/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.2813e-04 - val_loss: 8.7286e-04\n",
            "Epoch 226/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.2849e-04 - val_loss: 8.7127e-04\n",
            "Epoch 227/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.2795e-04 - val_loss: 8.6426e-04\n",
            "Epoch 228/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.2420e-04 - val_loss: 8.6311e-04\n",
            "Epoch 229/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.2112e-04 - val_loss: 8.5769e-04\n",
            "Epoch 230/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.2193e-04 - val_loss: 8.6378e-04\n",
            "Epoch 231/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.2180e-04 - val_loss: 8.6505e-04\n",
            "Epoch 232/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.1834e-04 - val_loss: 8.5703e-04\n",
            "Epoch 233/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.1865e-04 - val_loss: 8.5840e-04\n",
            "Epoch 234/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.1626e-04 - val_loss: 8.6311e-04\n",
            "Epoch 235/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.1546e-04 - val_loss: 8.5277e-04\n",
            "Epoch 236/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.1258e-04 - val_loss: 8.6122e-04\n",
            "Epoch 237/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.1263e-04 - val_loss: 8.5087e-04\n",
            "Epoch 238/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.1274e-04 - val_loss: 8.5035e-04\n",
            "Epoch 239/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.1288e-04 - val_loss: 8.5194e-04\n",
            "Epoch 240/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.0680e-04 - val_loss: 8.5125e-04\n",
            "Epoch 241/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.0704e-04 - val_loss: 8.5019e-04\n",
            "Epoch 242/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.0747e-04 - val_loss: 8.5507e-04\n",
            "Epoch 243/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.0743e-04 - val_loss: 8.4875e-04\n",
            "Epoch 244/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.0432e-04 - val_loss: 8.4054e-04\n",
            "Epoch 245/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 8.0141e-04 - val_loss: 8.3865e-04\n",
            "Epoch 246/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 8.0314e-04 - val_loss: 8.4098e-04\n",
            "Epoch 247/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.9707e-04 - val_loss: 8.4171e-04\n",
            "Epoch 248/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.9727e-04 - val_loss: 8.4164e-04\n",
            "Epoch 249/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.9782e-04 - val_loss: 8.5405e-04\n",
            "Epoch 250/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.9508e-04 - val_loss: 8.3834e-04\n",
            "Epoch 251/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.9487e-04 - val_loss: 8.3278e-04\n",
            "Epoch 252/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.9354e-04 - val_loss: 8.4054e-04\n",
            "Epoch 253/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.9356e-04 - val_loss: 8.3540e-04\n",
            "Epoch 254/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8979e-04 - val_loss: 8.3267e-04\n",
            "Epoch 255/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8957e-04 - val_loss: 8.3266e-04\n",
            "Epoch 256/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8699e-04 - val_loss: 8.2833e-04\n",
            "Epoch 257/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8733e-04 - val_loss: 8.4543e-04\n",
            "Epoch 258/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8594e-04 - val_loss: 8.3683e-04\n",
            "Epoch 259/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8484e-04 - val_loss: 8.3557e-04\n",
            "Epoch 260/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8235e-04 - val_loss: 8.2442e-04\n",
            "Epoch 261/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8121e-04 - val_loss: 8.4000e-04\n",
            "Epoch 262/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.8060e-04 - val_loss: 8.2379e-04\n",
            "Epoch 263/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.7865e-04 - val_loss: 8.2160e-04\n",
            "Epoch 264/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.7896e-04 - val_loss: 8.2365e-04\n",
            "Epoch 265/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.7673e-04 - val_loss: 8.3157e-04\n",
            "Epoch 266/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.7682e-04 - val_loss: 8.1570e-04\n",
            "Epoch 267/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.7388e-04 - val_loss: 8.1624e-04\n",
            "Epoch 268/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.7484e-04 - val_loss: 8.2490e-04\n",
            "Epoch 269/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.7379e-04 - val_loss: 8.2116e-04\n",
            "Epoch 270/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.7151e-04 - val_loss: 8.1604e-04\n",
            "Epoch 271/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.6945e-04 - val_loss: 8.2179e-04\n",
            "Epoch 272/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.6872e-04 - val_loss: 8.1114e-04\n",
            "Epoch 273/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.6719e-04 - val_loss: 8.1217e-04\n",
            "Epoch 274/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.6484e-04 - val_loss: 8.1886e-04\n",
            "Epoch 275/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.6625e-04 - val_loss: 8.1044e-04\n",
            "Epoch 276/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.6268e-04 - val_loss: 8.1226e-04\n",
            "Epoch 277/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.6280e-04 - val_loss: 8.0647e-04\n",
            "Epoch 278/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.6043e-04 - val_loss: 8.0532e-04\n",
            "Epoch 279/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.6130e-04 - val_loss: 8.0341e-04\n",
            "Epoch 280/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.5899e-04 - val_loss: 8.0606e-04\n",
            "Epoch 281/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.6428e-04 - val_loss: 8.0381e-04\n",
            "Epoch 282/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.5806e-04 - val_loss: 8.0872e-04\n",
            "Epoch 283/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.5648e-04 - val_loss: 8.0495e-04\n",
            "Epoch 284/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.5568e-04 - val_loss: 8.0007e-04\n",
            "Epoch 285/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.5376e-04 - val_loss: 8.0530e-04\n",
            "Epoch 286/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.5319e-04 - val_loss: 8.0218e-04\n",
            "Epoch 287/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.5149e-04 - val_loss: 8.0538e-04\n",
            "Epoch 288/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.5364e-04 - val_loss: 7.9866e-04\n",
            "Epoch 289/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.4928e-04 - val_loss: 8.0379e-04\n",
            "Epoch 290/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.4903e-04 - val_loss: 8.0263e-04\n",
            "Epoch 291/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.4923e-04 - val_loss: 7.9829e-04\n",
            "Epoch 292/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.4910e-04 - val_loss: 7.9439e-04\n",
            "Epoch 293/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.4670e-04 - val_loss: 7.9095e-04\n",
            "Epoch 294/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.4498e-04 - val_loss: 7.9687e-04\n",
            "Epoch 295/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.4367e-04 - val_loss: 7.9423e-04\n",
            "Epoch 296/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.4400e-04 - val_loss: 7.9354e-04\n",
            "Epoch 297/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.4398e-04 - val_loss: 7.8881e-04\n",
            "Epoch 298/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3925e-04 - val_loss: 7.8900e-04\n",
            "Epoch 299/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3826e-04 - val_loss: 7.8963e-04\n",
            "Epoch 300/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.4047e-04 - val_loss: 7.8116e-04\n",
            "Epoch 301/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3830e-04 - val_loss: 7.9005e-04\n",
            "Epoch 302/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3628e-04 - val_loss: 7.8564e-04\n",
            "Epoch 303/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.3844e-04 - val_loss: 7.8608e-04\n",
            "Epoch 304/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3325e-04 - val_loss: 7.8512e-04\n",
            "Epoch 305/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3442e-04 - val_loss: 7.8783e-04\n",
            "Epoch 306/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3202e-04 - val_loss: 7.9027e-04\n",
            "Epoch 307/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3627e-04 - val_loss: 7.8010e-04\n",
            "Epoch 308/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3257e-04 - val_loss: 7.8481e-04\n",
            "Epoch 309/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.3006e-04 - val_loss: 7.8430e-04\n",
            "Epoch 310/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.2999e-04 - val_loss: 7.7642e-04\n",
            "Epoch 311/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.2897e-04 - val_loss: 7.7891e-04\n",
            "Epoch 312/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.2641e-04 - val_loss: 7.7786e-04\n",
            "Epoch 313/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.3212e-04 - val_loss: 7.7444e-04\n",
            "Epoch 314/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.2579e-04 - val_loss: 7.7745e-04\n",
            "Epoch 315/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.2425e-04 - val_loss: 7.7227e-04\n",
            "Epoch 316/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.2516e-04 - val_loss: 7.7062e-04\n",
            "Epoch 317/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.2317e-04 - val_loss: 7.7159e-04\n",
            "Epoch 318/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.2075e-04 - val_loss: 7.7345e-04\n",
            "Epoch 319/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.2077e-04 - val_loss: 7.7225e-04\n",
            "Epoch 320/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.2059e-04 - val_loss: 7.7050e-04\n",
            "Epoch 321/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1760e-04 - val_loss: 7.6942e-04\n",
            "Epoch 322/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1881e-04 - val_loss: 7.7477e-04\n",
            "Epoch 323/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1899e-04 - val_loss: 7.7388e-04\n",
            "Epoch 324/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.1732e-04 - val_loss: 7.7409e-04\n",
            "Epoch 325/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.1877e-04 - val_loss: 7.7321e-04\n",
            "Epoch 326/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1613e-04 - val_loss: 7.6586e-04\n",
            "Epoch 327/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1230e-04 - val_loss: 7.6282e-04\n",
            "Epoch 328/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.1413e-04 - val_loss: 7.7030e-04\n",
            "Epoch 329/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1227e-04 - val_loss: 7.6623e-04\n",
            "Epoch 330/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1108e-04 - val_loss: 7.5931e-04\n",
            "Epoch 331/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.0987e-04 - val_loss: 7.5674e-04\n",
            "Epoch 332/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1086e-04 - val_loss: 7.6247e-04\n",
            "Epoch 333/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.1069e-04 - val_loss: 7.5869e-04\n",
            "Epoch 334/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.0591e-04 - val_loss: 7.6027e-04\n",
            "Epoch 335/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.0525e-04 - val_loss: 7.5879e-04\n",
            "Epoch 336/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.0432e-04 - val_loss: 7.5805e-04\n",
            "Epoch 337/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.0850e-04 - val_loss: 7.6846e-04\n",
            "Epoch 338/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.0587e-04 - val_loss: 7.5445e-04\n",
            "Epoch 339/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.0376e-04 - val_loss: 7.6959e-04\n",
            "Epoch 340/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 7.0300e-04 - val_loss: 7.5625e-04\n",
            "Epoch 341/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.0005e-04 - val_loss: 7.6606e-04\n",
            "Epoch 342/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9971e-04 - val_loss: 7.5955e-04\n",
            "Epoch 343/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.0281e-04 - val_loss: 7.6147e-04\n",
            "Epoch 344/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.9917e-04 - val_loss: 7.5713e-04\n",
            "Epoch 345/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9769e-04 - val_loss: 7.4866e-04\n",
            "Epoch 346/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9783e-04 - val_loss: 7.6758e-04\n",
            "Epoch 347/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.9970e-04 - val_loss: 7.5875e-04\n",
            "Epoch 348/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 7.0320e-04 - val_loss: 7.6247e-04\n",
            "Epoch 349/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9991e-04 - val_loss: 7.5163e-04\n",
            "Epoch 350/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.9552e-04 - val_loss: 7.5391e-04\n",
            "Epoch 351/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9643e-04 - val_loss: 7.5660e-04\n",
            "Epoch 352/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9223e-04 - val_loss: 7.4888e-04\n",
            "Epoch 353/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9265e-04 - val_loss: 7.5140e-04\n",
            "Epoch 354/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9175e-04 - val_loss: 7.4562e-04\n",
            "Epoch 355/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9025e-04 - val_loss: 7.5752e-04\n",
            "Epoch 356/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8978e-04 - val_loss: 7.4849e-04\n",
            "Epoch 357/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8910e-04 - val_loss: 7.4570e-04\n",
            "Epoch 358/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8905e-04 - val_loss: 7.4584e-04\n",
            "Epoch 359/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8796e-04 - val_loss: 7.5806e-04\n",
            "Epoch 360/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.9063e-04 - val_loss: 7.4481e-04\n",
            "Epoch 361/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.9229e-04 - val_loss: 7.4503e-04\n",
            "Epoch 362/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8379e-04 - val_loss: 7.4042e-04\n",
            "Epoch 363/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.8788e-04 - val_loss: 7.4083e-04\n",
            "Epoch 364/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8545e-04 - val_loss: 7.4677e-04\n",
            "Epoch 365/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8694e-04 - val_loss: 7.3909e-04\n",
            "Epoch 366/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.8233e-04 - val_loss: 7.3787e-04\n",
            "Epoch 367/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8211e-04 - val_loss: 7.4066e-04\n",
            "Epoch 368/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8065e-04 - val_loss: 7.4359e-04\n",
            "Epoch 369/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8185e-04 - val_loss: 7.5401e-04\n",
            "Epoch 370/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8560e-04 - val_loss: 7.3753e-04\n",
            "Epoch 371/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.8405e-04 - val_loss: 7.3437e-04\n",
            "Epoch 372/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7879e-04 - val_loss: 7.3479e-04\n",
            "Epoch 373/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7815e-04 - val_loss: 7.3474e-04\n",
            "Epoch 374/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.7934e-04 - val_loss: 7.4256e-04\n",
            "Epoch 375/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.7761e-04 - val_loss: 7.3069e-04\n",
            "Epoch 376/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7696e-04 - val_loss: 7.4045e-04\n",
            "Epoch 377/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7680e-04 - val_loss: 7.3379e-04\n",
            "Epoch 378/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7498e-04 - val_loss: 7.3517e-04\n",
            "Epoch 379/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7424e-04 - val_loss: 7.3116e-04\n",
            "Epoch 380/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7651e-04 - val_loss: 7.3495e-04\n",
            "Epoch 381/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7319e-04 - val_loss: 7.2883e-04\n",
            "Epoch 382/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7246e-04 - val_loss: 7.3342e-04\n",
            "Epoch 383/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7393e-04 - val_loss: 7.3282e-04\n",
            "Epoch 384/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7049e-04 - val_loss: 7.2809e-04\n",
            "Epoch 385/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6942e-04 - val_loss: 7.3055e-04\n",
            "Epoch 386/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6916e-04 - val_loss: 7.2700e-04\n",
            "Epoch 387/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.7034e-04 - val_loss: 7.2948e-04\n",
            "Epoch 388/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6839e-04 - val_loss: 7.2834e-04\n",
            "Epoch 389/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.7033e-04 - val_loss: 7.3512e-04\n",
            "Epoch 390/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6766e-04 - val_loss: 7.3095e-04\n",
            "Epoch 391/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6582e-04 - val_loss: 7.2337e-04\n",
            "Epoch 392/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6560e-04 - val_loss: 7.2522e-04\n",
            "Epoch 393/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6590e-04 - val_loss: 7.4559e-04\n",
            "Epoch 394/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6561e-04 - val_loss: 7.2497e-04\n",
            "Epoch 395/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6371e-04 - val_loss: 7.2301e-04\n",
            "Epoch 396/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6396e-04 - val_loss: 7.3621e-04\n",
            "Epoch 397/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6346e-04 - val_loss: 7.3197e-04\n",
            "Epoch 398/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6215e-04 - val_loss: 7.2286e-04\n",
            "Epoch 399/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6574e-04 - val_loss: 7.2603e-04\n",
            "Epoch 400/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6050e-04 - val_loss: 7.2225e-04\n",
            "Epoch 401/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6179e-04 - val_loss: 7.3407e-04\n",
            "Epoch 402/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6025e-04 - val_loss: 7.2001e-04\n",
            "Epoch 403/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.6229e-04 - val_loss: 7.2030e-04\n",
            "Epoch 404/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5905e-04 - val_loss: 7.2285e-04\n",
            "Epoch 405/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5896e-04 - val_loss: 7.1668e-04\n",
            "Epoch 406/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5743e-04 - val_loss: 7.1356e-04\n",
            "Epoch 407/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5704e-04 - val_loss: 7.1900e-04\n",
            "Epoch 408/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.5802e-04 - val_loss: 7.1624e-04\n",
            "Epoch 409/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5625e-04 - val_loss: 7.1674e-04\n",
            "Epoch 410/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5667e-04 - val_loss: 7.1372e-04\n",
            "Epoch 411/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5577e-04 - val_loss: 7.1488e-04\n",
            "Epoch 412/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5446e-04 - val_loss: 7.2202e-04\n",
            "Epoch 413/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5618e-04 - val_loss: 7.1342e-04\n",
            "Epoch 414/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5366e-04 - val_loss: 7.2250e-04\n",
            "Epoch 415/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5423e-04 - val_loss: 7.1208e-04\n",
            "Epoch 416/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5158e-04 - val_loss: 7.1651e-04\n",
            "Epoch 417/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5221e-04 - val_loss: 7.1287e-04\n",
            "Epoch 418/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5158e-04 - val_loss: 7.1879e-04\n",
            "Epoch 419/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5244e-04 - val_loss: 7.1242e-04\n",
            "Epoch 420/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4814e-04 - val_loss: 7.1814e-04\n",
            "Epoch 421/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5061e-04 - val_loss: 7.1132e-04\n",
            "Epoch 422/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5071e-04 - val_loss: 7.1343e-04\n",
            "Epoch 423/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4993e-04 - val_loss: 7.1101e-04\n",
            "Epoch 424/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4636e-04 - val_loss: 7.1473e-04\n",
            "Epoch 425/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4803e-04 - val_loss: 7.1676e-04\n",
            "Epoch 426/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4993e-04 - val_loss: 7.2353e-04\n",
            "Epoch 427/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.5022e-04 - val_loss: 7.1306e-04\n",
            "Epoch 428/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4897e-04 - val_loss: 7.0954e-04\n",
            "Epoch 429/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.4495e-04 - val_loss: 7.1199e-04\n",
            "Epoch 430/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.4989e-04 - val_loss: 7.1693e-04\n",
            "Epoch 431/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.4512e-04 - val_loss: 7.1156e-04\n",
            "Epoch 432/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.4489e-04 - val_loss: 7.0865e-04\n",
            "Epoch 433/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.4274e-04 - val_loss: 7.0894e-04\n",
            "Epoch 434/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.4394e-04 - val_loss: 7.0720e-04\n",
            "Epoch 435/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4451e-04 - val_loss: 7.0538e-04\n",
            "Epoch 436/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.4205e-04 - val_loss: 7.0388e-04\n",
            "Epoch 437/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3996e-04 - val_loss: 7.0335e-04\n",
            "Epoch 438/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.3936e-04 - val_loss: 7.0519e-04\n",
            "Epoch 439/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4150e-04 - val_loss: 7.0403e-04\n",
            "Epoch 440/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.3892e-04 - val_loss: 7.1124e-04\n",
            "Epoch 441/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.3892e-04 - val_loss: 7.0427e-04\n",
            "Epoch 442/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3879e-04 - val_loss: 7.0135e-04\n",
            "Epoch 443/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3573e-04 - val_loss: 7.0691e-04\n",
            "Epoch 444/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3916e-04 - val_loss: 7.0414e-04\n",
            "Epoch 445/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3752e-04 - val_loss: 6.9910e-04\n",
            "Epoch 446/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.3627e-04 - val_loss: 7.0791e-04\n",
            "Epoch 447/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.4112e-04 - val_loss: 7.0501e-04\n",
            "Epoch 448/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3820e-04 - val_loss: 7.0001e-04\n",
            "Epoch 449/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3419e-04 - val_loss: 7.0332e-04\n",
            "Epoch 450/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3428e-04 - val_loss: 7.0226e-04\n",
            "Epoch 451/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3349e-04 - val_loss: 7.0150e-04\n",
            "Epoch 452/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3855e-04 - val_loss: 7.1093e-04\n",
            "Epoch 453/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.3492e-04 - val_loss: 6.9986e-04\n",
            "Epoch 454/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3365e-04 - val_loss: 6.9520e-04\n",
            "Epoch 455/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3268e-04 - val_loss: 7.0060e-04\n",
            "Epoch 456/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.3210e-04 - val_loss: 6.9512e-04\n",
            "Epoch 457/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2998e-04 - val_loss: 6.9764e-04\n",
            "Epoch 458/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2941e-04 - val_loss: 6.9321e-04\n",
            "Epoch 459/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2998e-04 - val_loss: 7.0221e-04\n",
            "Epoch 460/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3152e-04 - val_loss: 7.0327e-04\n",
            "Epoch 461/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2882e-04 - val_loss: 6.9434e-04\n",
            "Epoch 462/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3171e-04 - val_loss: 6.9570e-04\n",
            "Epoch 463/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.3302e-04 - val_loss: 6.9758e-04\n",
            "Epoch 464/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.2737e-04 - val_loss: 6.8980e-04\n",
            "Epoch 465/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.2591e-04 - val_loss: 6.9168e-04\n",
            "Epoch 466/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2598e-04 - val_loss: 6.9071e-04\n",
            "Epoch 467/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2673e-04 - val_loss: 6.9268e-04\n",
            "Epoch 468/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2753e-04 - val_loss: 6.9515e-04\n",
            "Epoch 469/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2557e-04 - val_loss: 7.0429e-04\n",
            "Epoch 470/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2792e-04 - val_loss: 6.9539e-04\n",
            "Epoch 471/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2413e-04 - val_loss: 6.8995e-04\n",
            "Epoch 472/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.2541e-04 - val_loss: 7.0091e-04\n",
            "Epoch 473/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.2408e-04 - val_loss: 6.8579e-04\n",
            "Epoch 474/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2373e-04 - val_loss: 6.8753e-04\n",
            "Epoch 475/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.2438e-04 - val_loss: 6.8794e-04\n",
            "Epoch 476/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2193e-04 - val_loss: 6.8928e-04\n",
            "Epoch 477/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2264e-04 - val_loss: 6.8684e-04\n",
            "Epoch 478/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.2090e-04 - val_loss: 6.8803e-04\n",
            "Epoch 479/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.1974e-04 - val_loss: 6.8754e-04\n",
            "Epoch 480/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1992e-04 - val_loss: 6.8965e-04\n",
            "Epoch 481/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2357e-04 - val_loss: 6.9156e-04\n",
            "Epoch 482/500\n",
            "69/69 [==============================] - 1s 14ms/step - loss: 6.2080e-04 - val_loss: 6.8319e-04\n",
            "Epoch 483/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1843e-04 - val_loss: 6.8574e-04\n",
            "Epoch 484/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1902e-04 - val_loss: 6.8161e-04\n",
            "Epoch 485/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1620e-04 - val_loss: 6.8550e-04\n",
            "Epoch 486/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1896e-04 - val_loss: 6.8610e-04\n",
            "Epoch 487/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1859e-04 - val_loss: 6.8558e-04\n",
            "Epoch 488/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1684e-04 - val_loss: 6.8726e-04\n",
            "Epoch 489/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.2240e-04 - val_loss: 6.8696e-04\n",
            "Epoch 490/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1876e-04 - val_loss: 6.8490e-04\n",
            "Epoch 491/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1698e-04 - val_loss: 6.8167e-04\n",
            "Epoch 492/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1485e-04 - val_loss: 6.8426e-04\n",
            "Epoch 493/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1668e-04 - val_loss: 6.8423e-04\n",
            "Epoch 494/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1553e-04 - val_loss: 6.8719e-04\n",
            "Epoch 495/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1410e-04 - val_loss: 6.7941e-04\n",
            "Epoch 496/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1127e-04 - val_loss: 6.8609e-04\n",
            "Epoch 497/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1492e-04 - val_loss: 6.8757e-04\n",
            "Epoch 498/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1369e-04 - val_loss: 6.9113e-04\n",
            "Epoch 499/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1255e-04 - val_loss: 6.8430e-04\n",
            "Epoch 500/500\n",
            "69/69 [==============================] - 1s 15ms/step - loss: 6.1076e-04 - val_loss: 6.8011e-04\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q0BG10VNU6wA",
        "outputId": "8d3dc99f-1865-4b8f-8f36-f19657f65495",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "predicted_mean = regression.predict(np.atleast_2d(sim_data_noisy[:]+0.5))\n",
        "training_var_unknown_mean = (sim_data[:]+0.5-predicted_mean)**2.\n",
        "training_var_unknown_mean.shape"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(80000, 128)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "w-3qy3I8WYWZ",
        "outputId": "0cae0544-a8fc-44c7-a9b6-866a57d69f96",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "model_instance = demo.simpler_leaky(128,128, learning_rate=1e-3)\n",
        "regression_var_unknown_mean = model_instance.model()\n",
        "history = regression_var_unknown_mean.fit(sim_data_noisy[:validation_index]+0.5,\n",
        "                                          training_var_unknown_mean[:validation_index]*1e3,\n",
        "                                          epochs =100,batch_size =512, shuffle=True,\n",
        "                                          validation_data = (sim_data_noisy[validation_index:]+0.5,\n",
        "                                                             training_var_unknown_mean[validation_index:]*1e3))"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "128\n",
            "Model: \"functional_21\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_11 (InputLayer)        [(None, 128)]             0         \n",
            "_________________________________________________________________\n",
            "dense_32 (Dense)             (None, 128)               16512     \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_22 (LeakyReLU)   (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "dense_33 (Dense)             (None, 256)               33024     \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_23 (LeakyReLU)   (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "dense_34 (Dense)             (None, 128)               32896     \n",
            "=================================================================\n",
            "Total params: 82,432\n",
            "Trainable params: 82,432\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/100\n",
            "69/69 [==============================] - 1s 13ms/step - loss: 1.7517 - val_loss: 2.1388\n",
            "Epoch 2/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.7147 - val_loss: 2.1364\n",
            "Epoch 3/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.7087 - val_loss: 2.1264\n",
            "Epoch 4/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.6910 - val_loss: 2.0897\n",
            "Epoch 5/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.6697 - val_loss: 2.0688\n",
            "Epoch 6/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.6389 - val_loss: 2.0595\n",
            "Epoch 7/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.5826 - val_loss: 2.0261\n",
            "Epoch 8/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.4924 - val_loss: 2.0076\n",
            "Epoch 9/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.4401 - val_loss: 1.9851\n",
            "Epoch 10/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.4006 - val_loss: 1.9727\n",
            "Epoch 11/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.3791 - val_loss: 1.9594\n",
            "Epoch 12/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.3567 - val_loss: 1.9219\n",
            "Epoch 13/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.3510 - val_loss: 1.9136\n",
            "Epoch 14/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.3331 - val_loss: 1.9066\n",
            "Epoch 15/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.3173 - val_loss: 1.8670\n",
            "Epoch 16/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.3059 - val_loss: 1.8457\n",
            "Epoch 17/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.2904 - val_loss: 1.8390\n",
            "Epoch 18/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.2748 - val_loss: 1.8102\n",
            "Epoch 19/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.2677 - val_loss: 1.7832\n",
            "Epoch 20/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.2505 - val_loss: 1.7667\n",
            "Epoch 21/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.2370 - val_loss: 1.7522\n",
            "Epoch 22/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.2233 - val_loss: 1.7278\n",
            "Epoch 23/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.2182 - val_loss: 1.7149\n",
            "Epoch 24/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.2035 - val_loss: 1.7344\n",
            "Epoch 25/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1995 - val_loss: 1.7113\n",
            "Epoch 26/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1939 - val_loss: 1.7540\n",
            "Epoch 27/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.1894 - val_loss: 1.7046\n",
            "Epoch 28/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1813 - val_loss: 1.6743\n",
            "Epoch 29/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1732 - val_loss: 1.6796\n",
            "Epoch 30/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.1633 - val_loss: 1.6737\n",
            "Epoch 31/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1598 - val_loss: 1.6631\n",
            "Epoch 32/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1560 - val_loss: 1.6627\n",
            "Epoch 33/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1470 - val_loss: 1.6549\n",
            "Epoch 34/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1414 - val_loss: 1.6509\n",
            "Epoch 35/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.1369 - val_loss: 1.6506\n",
            "Epoch 36/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1326 - val_loss: 1.6412\n",
            "Epoch 37/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1221 - val_loss: 1.6400\n",
            "Epoch 38/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.1212 - val_loss: 1.6338\n",
            "Epoch 39/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1180 - val_loss: 1.6412\n",
            "Epoch 40/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.1074 - val_loss: 1.6268\n",
            "Epoch 41/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1090 - val_loss: 1.6378\n",
            "Epoch 42/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1032 - val_loss: 1.6320\n",
            "Epoch 43/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.1003 - val_loss: 1.6164\n",
            "Epoch 44/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0918 - val_loss: 1.6146\n",
            "Epoch 45/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0905 - val_loss: 1.6185\n",
            "Epoch 46/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0852 - val_loss: 1.6449\n",
            "Epoch 47/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0823 - val_loss: 1.6312\n",
            "Epoch 48/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0788 - val_loss: 1.6145\n",
            "Epoch 49/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0742 - val_loss: 1.5971\n",
            "Epoch 50/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0751 - val_loss: 1.6063\n",
            "Epoch 51/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0709 - val_loss: 1.5978\n",
            "Epoch 52/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0660 - val_loss: 1.6202\n",
            "Epoch 53/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0630 - val_loss: 1.6095\n",
            "Epoch 54/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0577 - val_loss: 1.6001\n",
            "Epoch 55/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0556 - val_loss: 1.5949\n",
            "Epoch 56/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0529 - val_loss: 1.6198\n",
            "Epoch 57/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0526 - val_loss: 1.5882\n",
            "Epoch 58/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0446 - val_loss: 1.5800\n",
            "Epoch 59/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0440 - val_loss: 1.5765\n",
            "Epoch 60/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0414 - val_loss: 1.5762\n",
            "Epoch 61/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0366 - val_loss: 1.5986\n",
            "Epoch 62/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0329 - val_loss: 1.5826\n",
            "Epoch 63/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0353 - val_loss: 1.5939\n",
            "Epoch 64/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0286 - val_loss: 1.5944\n",
            "Epoch 65/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0316 - val_loss: 1.5697\n",
            "Epoch 66/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0233 - val_loss: 1.5626\n",
            "Epoch 67/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0205 - val_loss: 1.5738\n",
            "Epoch 68/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0176 - val_loss: 1.5774\n",
            "Epoch 69/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0186 - val_loss: 1.5931\n",
            "Epoch 70/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0129 - val_loss: 1.5605\n",
            "Epoch 71/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0110 - val_loss: 1.5634\n",
            "Epoch 72/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0054 - val_loss: 1.6205\n",
            "Epoch 73/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0065 - val_loss: 1.5644\n",
            "Epoch 74/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 1.0059 - val_loss: 1.5672\n",
            "Epoch 75/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 1.0007 - val_loss: 1.5714\n",
            "Epoch 76/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9970 - val_loss: 1.5538\n",
            "Epoch 77/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 0.9935 - val_loss: 1.5574\n",
            "Epoch 78/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9948 - val_loss: 1.5513\n",
            "Epoch 79/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9902 - val_loss: 1.5468\n",
            "Epoch 80/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9854 - val_loss: 1.5466\n",
            "Epoch 81/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9858 - val_loss: 1.5503\n",
            "Epoch 82/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 0.9882 - val_loss: 1.5537\n",
            "Epoch 83/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9804 - val_loss: 1.5704\n",
            "Epoch 84/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9787 - val_loss: 1.5504\n",
            "Epoch 85/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9755 - val_loss: 1.5538\n",
            "Epoch 86/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9785 - val_loss: 1.5555\n",
            "Epoch 87/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9708 - val_loss: 1.5522\n",
            "Epoch 88/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9686 - val_loss: 1.5519\n",
            "Epoch 89/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9692 - val_loss: 1.5814\n",
            "Epoch 90/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9639 - val_loss: 1.5897\n",
            "Epoch 91/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9596 - val_loss: 1.5386\n",
            "Epoch 92/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9579 - val_loss: 1.5782\n",
            "Epoch 93/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9580 - val_loss: 1.5457\n",
            "Epoch 94/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 0.9561 - val_loss: 1.5390\n",
            "Epoch 95/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9617 - val_loss: 1.5490\n",
            "Epoch 96/100\n",
            "69/69 [==============================] - 1s 11ms/step - loss: 0.9509 - val_loss: 1.5409\n",
            "Epoch 97/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9556 - val_loss: 1.5430\n",
            "Epoch 98/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9507 - val_loss: 1.5768\n",
            "Epoch 99/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9439 - val_loss: 1.5482\n",
            "Epoch 100/100\n",
            "69/69 [==============================] - 1s 12ms/step - loss: 0.9419 - val_loss: 1.5450\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fZ0KDMs8Wh--",
        "outputId": "204287b3-2c76-4b1b-d6fb-4a910983c4a2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "for i_index in np.arange(57+int(n_training*8),63+int(n_training*8)):\n",
        "  print(i_index)\n",
        "  predicted_var_obs = (regression_var_unknown_mean.predict(np.atleast_2d(sim_data_noisy[i_index]+0.5))[0])*1e-3\n",
        "  predicted_mean_obs = (regression.predict(np.atleast_2d(sim_data_noisy[i_index,]+0.5))[0])\n",
        "  plt.figure(figsize=(12,4))\n",
        "  _ = plt.subplot(1,2,1)\n",
        "  _ = plt.ylabel(r'$h_+$ strain',fontsize=14)\n",
        "  plt.fill_between(np.linspace(0,1./16,128),\n",
        "                  predicted_mean_obs-0.5- np.sqrt(predicted_var_obs),\n",
        "                  predicted_mean_obs + np.sqrt(predicted_var_obs)-0.5, alpha=0.5)\n",
        "  plt.scatter(np.linspace(0,1./16,128),sim_data_noisy[i_index,], alpha=0.8)\n",
        "  plt.plot(np.linspace(0,1./16,128),sim_data[i_index,], c='C2')\n",
        "  plt.xlabel(r'$t\\ /$ seconds',fontsize=14), plt.show()"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "40057\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "40058\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "40059\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "40060\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "40061\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "40062\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}