DB_TCP.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "DB-TCP.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "44ci0dnuo_iq",
        "outputId": "dbf17b9e-5d1a-4db9-9c6f-ecef7b7b2575"
      },
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import os\n",
        "from google.colab import files \n",
        "print(tf.__version__)\n",
        "\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.6.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L6ahhvrJpEtE"
      },
      "source": [
        "!pip install -q pyyaml h5py \n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": "OK"
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "cbiX2rR9tlU0",
        "outputId": "1d161f86-c52b-4973-f784-1aa631eb854a"
      },
      "source": [
        "uoloaded=files.upload()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-42c0f3dd-e4de-4a0e-9c68-92d0e26a6e34\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-42c0f3dd-e4de-4a0e-9c68-92d0e26a6e34\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving eval_input.txt to eval_input.txt\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n0Ft-CaQwG14"
      },
      "source": [
        "training_set = pd.read_csv('input.txt', delim_whitespace=True)\n",
        "\n",
        "training_eval = pd.read_csv('eval_input.txt', delim_whitespace=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jwQtwLbWpJWQ",
        "outputId": "10aca898-dc27-4b45-e36d-3666990841bb"
      },
      "source": [
        "print(training_set.head())\n",
        "\n",
        "print(training_eval.head())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "   rtt  tmp  targetedvscwnd  diff  average  output\n",
            "1  0.0  0.0             0.0     0      0.0       0\n",
            "2  0.0  0.0             0.0     0      0.0       0\n",
            "3  0.0  0.0             0.0     0      0.0       0\n",
            "4  0.0  0.0             0.0     0      0.0       0\n",
            "5  0.0  0.0             0.0     0      0.0       0\n",
            "   rtt  tmp  targetedvscwnd  diff  average  output\n",
            "1  0.0  0.0             0.0     0      0.0       0\n",
            "2  0.0  0.0             0.0     0      0.0       0\n",
            "3  0.0  0.0             0.0     0      0.0       0\n",
            "4  0.0  0.0             0.0     0      0.0       0\n",
            "5  0.0  0.0             0.0     0      0.0       0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0PY2CUQOpLfb",
        "outputId": "d51a2e4f-cfb2-45f8-fb26-128099cf2b20"
      },
      "source": [
        "column=training_set[\"diff\"]\n",
        "print(column.max())\n",
        "print(column.min())\n",
        "column=column/1759\n",
        "print(column[100])\n",
        "\n",
        "column1=training_eval[\"diff\"]\n",
        "print(column1.max())\n",
        "print(column1.min())\n",
        "column1=column1/1759\n",
        "print(column1[100])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1759\n",
            "0\n",
            "0.014781125639567936\n",
            "1561\n",
            "0\n",
            "0.014781125639567936\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ilBrxLshpeqm",
        "outputId": "2541b8cb-3f61-425e-939e-0707a9f05458"
      },
      "source": [
        "training_set[\"diff\"]=column\n",
        "print(training_set)\n",
        "\n",
        "training_eval[\"diff\"]=column1\n",
        "print(training_eval)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "           rtt       tmp  targetedvscwnd      diff   average  output\n",
            "1     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "2     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "3     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "4     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "5     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "...        ...       ...             ...       ...       ...     ...\n",
            "6897  0.022585  0.974333        0.974053  0.043206  0.022585       0\n",
            "6898  0.022585  0.974333        0.974061  0.043206  0.022585       0\n",
            "6899  0.022585  0.974333        0.974061  0.043206  0.022585       0\n",
            "6900  0.022585  0.974333        0.974070  0.043206  0.022585       0\n",
            "6901  0.022585  0.974333        0.974070  0.043206  0.022585       0\n",
            "\n",
            "[6901 rows x 6 columns]\n",
            "           rtt       tmp  targetedvscwnd      diff   average  output\n",
            "1     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "2     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "3     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "4     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "5     0.000000  0.000000        0.000000  0.000000  0.000000       0\n",
            "...        ...       ...             ...       ...       ...     ...\n",
            "1487  0.022174  0.992387        0.992129  0.010802  0.022174       0\n",
            "1488  0.022174  0.992387        0.992133  0.010802  0.022174       0\n",
            "1489  0.022174  0.992387        0.992133  0.010802  0.022174       0\n",
            "1490  0.022174  0.992387        0.992133  0.010802  0.022174       0\n",
            "1491  0.022174  0.992387        0.992136  0.010802  0.022174       0\n",
            "\n",
            "[1491 rows x 6 columns]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vm5Hp9klpiBg"
      },
      "source": [
        "training_output=training_set.pop(\"output\")\n",
        "\n",
        "training_eval_output=training_eval.pop(\"output\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IiJut7u-pkgx"
      },
      "source": [
        "training_input=training_set\n",
        "\n",
        "training_eval_input=training_eval"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PhL586_QqaUC",
        "outputId": "bafe9f7d-b9db-4942-cc24-58208a81699e"
      },
      "source": [
        "print(training_output)\n",
        "print(training_input)\n",
        "\n",
        "print(training_eval_output)\n",
        "print(training_eval_input)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1       0\n",
            "2       0\n",
            "3       0\n",
            "4       0\n",
            "5       0\n",
            "       ..\n",
            "6897    0\n",
            "6898    0\n",
            "6899    0\n",
            "6900    0\n",
            "6901    0\n",
            "Name: output, Length: 6901, dtype: int64\n",
            "           rtt       tmp  targetedvscwnd      diff   average\n",
            "1     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "2     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "3     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "4     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "5     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "...        ...       ...             ...       ...       ...\n",
            "6897  0.022585  0.974333        0.974053  0.043206  0.022585\n",
            "6898  0.022585  0.974333        0.974061  0.043206  0.022585\n",
            "6899  0.022585  0.974333        0.974061  0.043206  0.022585\n",
            "6900  0.022585  0.974333        0.974070  0.043206  0.022585\n",
            "6901  0.022585  0.974333        0.974070  0.043206  0.022585\n",
            "\n",
            "[6901 rows x 5 columns]\n",
            "1       0\n",
            "2       0\n",
            "3       0\n",
            "4       0\n",
            "5       0\n",
            "       ..\n",
            "1487    0\n",
            "1488    0\n",
            "1489    0\n",
            "1490    0\n",
            "1491    0\n",
            "Name: output, Length: 1491, dtype: int64\n",
            "           rtt       tmp  targetedvscwnd      diff   average\n",
            "1     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "2     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "3     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "4     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "5     0.000000  0.000000        0.000000  0.000000  0.000000\n",
            "...        ...       ...             ...       ...       ...\n",
            "1487  0.022174  0.992387        0.992129  0.010802  0.022174\n",
            "1488  0.022174  0.992387        0.992133  0.010802  0.022174\n",
            "1489  0.022174  0.992387        0.992133  0.010802  0.022174\n",
            "1490  0.022174  0.992387        0.992133  0.010802  0.022174\n",
            "1491  0.022174  0.992387        0.992136  0.010802  0.022174\n",
            "\n",
            "[1491 rows x 5 columns]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "lQU1s8zcqcY8",
        "outputId": "286ec0f2-943f-4979-c70c-538ecece6f4c"
      },
      "source": [
        "training_input.rtt.hist(bins=20)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f0066ec5750>"
            ]
          },
          "metadata": {},
          "execution_count": 26
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "HK_C6q6Xqe8i",
        "outputId": "8c932e03-16d0-4996-f1d4-4e90a8e5bf61"
      },
      "source": [
        "training_input.tmp.hist(bins=20)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f0066b3a750>"
            ]
          },
          "metadata": {},
          "execution_count": 27
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "4ZeqfsumqghW",
        "outputId": "1105961d-533a-4e85-c2ea-13fd1b37a27d"
      },
      "source": [
        "training_input.targetedvscwnd.hist(bins=20)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f0066ad0810>"
            ]
          },
          "metadata": {},
          "execution_count": 28
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HxypuEQoqhty"
      },
      "source": [
        "training_x=np.array(training_input)\n",
        "training_y=np.array(training_output)\n",
        "\n",
        "\n",
        "training_eval_x=np.array(training_eval_input)\n",
        "training_eval_y=np.array(training_eval_output)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 135
        },
        "id": "W0KW4BCGqnJz",
        "outputId": "35cdd5b0-e54d-4a44-daa0-086f1e68d0b5"
      },
      "source": [
        "print(training_x.shape)\n",
        "print(training_y.shape)\n",
        "\n",
        "\n",
        "print(training_eval_x.shape)\n",
        "print(training_eval_y.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "SyntaxError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;36m  File \u001b[0;32m\"<ipython-input-30-54739fdf34e3>\"\u001b[0;36m, line \u001b[0;32m6\u001b[0m\n\u001b[0;31m    print(training_eval_y.shape\u001b[0m\n\u001b[0m                               ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m unexpected EOF while parsing\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1ZduC3bMqoqq"
      },
      "source": [
        "training_x=training_x.reshape(len(training_x),5,1)\n",
        "training_y=training_y.reshape(len(training_y))\n",
        "\n",
        "print(training_x.shape)\n",
        "print(training_y.shape)\n",
        "\n",
        "\n",
        "training_eval_x=training_eval_x.reshape(len(training_eval_x),5,1)\n",
        "training_eval_y=training_eval_y.reshape(len(training_eval_y))\n",
        "\n",
        "print(training_eval_x.shape)\n",
        "print(training_eval_y.shape"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RXvgQBnrqsBd"
      },
      "source": [
        " model=tf.keras.Sequential([\n",
        "    tf.keras.layers.Flatten(input_shape=(5, 1)),\n",
        "      tf.keras.layers.Dense(20, activation='relu', kernel_initializer=keras.initializers.GlorotNormal()),\n",
        "      keras.layers.Dropout(rate=0.2),\n",
        "      tf.keras.layers.Dense(25, activation='relu',  kernel_initializer=keras.initializers.GlorotNormal()),\n",
        "      keras.layers.Dropout(rate=0.2),\n",
        "      tf.keras.layers.Dense(20, activation='relu',  kernel_initializer=keras.initializers.GlorotNormal()),\n",
        "      keras.layers.Dropout(rate=0.2),\n",
        "      tf.keras.layers.Dense(3,activation='softmax' ), \n",
        "  ])\n",
        " \n",
        " model.compile(\n",
        "      loss='sparse_categorical_crossentropy',\n",
        "      optimizer='adam',\n",
        "      metrics='accuracy')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gsbH-LdJqjj5",
        "outputId": "927c26b3-b161-488f-cf5f-a62b50699d8c"
      },
      "source": [
        "model.fit(training_x,training_y,epochs=987)\n",
        "          #,   callbacks=[callbacks])\n",
        "model.save('DTCP')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.1470 - accuracy: 0.9493\n",
            "Epoch 2/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.1290 - accuracy: 0.9604\n",
            "Epoch 3/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.1155 - accuracy: 0.9609\n",
            "Epoch 4/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.1099 - accuracy: 0.9654\n",
            "Epoch 5/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.1031 - accuracy: 0.9683\n",
            "Epoch 6/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.1011 - accuracy: 0.9678\n",
            "Epoch 7/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0885 - accuracy: 0.9736\n",
            "Epoch 8/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0887 - accuracy: 0.9738\n",
            "Epoch 9/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0833 - accuracy: 0.9730\n",
            "Epoch 10/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0764 - accuracy: 0.9761\n",
            "Epoch 11/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0706 - accuracy: 0.9764\n",
            "Epoch 12/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0713 - accuracy: 0.9754\n",
            "Epoch 13/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0666 - accuracy: 0.9781\n",
            "Epoch 14/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0675 - accuracy: 0.9791\n",
            "Epoch 15/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0633 - accuracy: 0.9787\n",
            "Epoch 16/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0560 - accuracy: 0.9836\n",
            "Epoch 17/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0550 - accuracy: 0.9817\n",
            "Epoch 18/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0628 - accuracy: 0.9786\n",
            "Epoch 19/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0505 - accuracy: 0.9852\n",
            "Epoch 20/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0523 - accuracy: 0.9838\n",
            "Epoch 21/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0481 - accuracy: 0.9849\n",
            "Epoch 22/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0484 - accuracy: 0.9858\n",
            "Epoch 23/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0488 - accuracy: 0.9852\n",
            "Epoch 24/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0431 - accuracy: 0.9870\n",
            "Epoch 25/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0510 - accuracy: 0.9836\n",
            "Epoch 26/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0428 - accuracy: 0.9870\n",
            "Epoch 27/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0443 - accuracy: 0.9859\n",
            "Epoch 28/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0409 - accuracy: 0.9878\n",
            "Epoch 29/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0428 - accuracy: 0.9875\n",
            "Epoch 30/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0419 - accuracy: 0.9897\n",
            "Epoch 31/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0393 - accuracy: 0.9881\n",
            "Epoch 32/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0445 - accuracy: 0.9870\n",
            "Epoch 33/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0404 - accuracy: 0.9890\n",
            "Epoch 34/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0388 - accuracy: 0.9887\n",
            "Epoch 35/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0418 - accuracy: 0.9874\n",
            "Epoch 36/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0396 - accuracy: 0.9872\n",
            "Epoch 37/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0418 - accuracy: 0.9880\n",
            "Epoch 38/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0404 - accuracy: 0.9887\n",
            "Epoch 39/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0346 - accuracy: 0.9904\n",
            "Epoch 40/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0339 - accuracy: 0.9904\n",
            "Epoch 41/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0363 - accuracy: 0.9887\n",
            "Epoch 42/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0387 - accuracy: 0.9877\n",
            "Epoch 43/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0366 - accuracy: 0.9894\n",
            "Epoch 44/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0338 - accuracy: 0.9896\n",
            "Epoch 45/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0352 - accuracy: 0.9888\n",
            "Epoch 46/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0365 - accuracy: 0.9880\n",
            "Epoch 47/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0323 - accuracy: 0.9904\n",
            "Epoch 48/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0364 - accuracy: 0.9878\n",
            "Epoch 49/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0308 - accuracy: 0.9907\n",
            "Epoch 50/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0336 - accuracy: 0.9906\n",
            "Epoch 51/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0307 - accuracy: 0.9907\n",
            "Epoch 52/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0324 - accuracy: 0.9901\n",
            "Epoch 53/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0344 - accuracy: 0.9880\n",
            "Epoch 54/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0337 - accuracy: 0.9903\n",
            "Epoch 55/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0332 - accuracy: 0.9893\n",
            "Epoch 56/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0270 - accuracy: 0.9910\n",
            "Epoch 57/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0350 - accuracy: 0.9890\n",
            "Epoch 58/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0332 - accuracy: 0.9900\n",
            "Epoch 59/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0363 - accuracy: 0.9894\n",
            "Epoch 60/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0336 - accuracy: 0.9891\n",
            "Epoch 61/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0287 - accuracy: 0.9916\n",
            "Epoch 62/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0261 - accuracy: 0.9925\n",
            "Epoch 63/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0308 - accuracy: 0.9903\n",
            "Epoch 64/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0279 - accuracy: 0.9920\n",
            "Epoch 65/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0340 - accuracy: 0.9887\n",
            "Epoch 66/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0248 - accuracy: 0.9930\n",
            "Epoch 67/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0280 - accuracy: 0.9919\n",
            "Epoch 68/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0309 - accuracy: 0.9903\n",
            "Epoch 69/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0284 - accuracy: 0.9929\n",
            "Epoch 70/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0275 - accuracy: 0.9913\n",
            "Epoch 71/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0250 - accuracy: 0.9922\n",
            "Epoch 72/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0248 - accuracy: 0.9923\n",
            "Epoch 73/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0285 - accuracy: 0.9916\n",
            "Epoch 74/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0284 - accuracy: 0.9906\n",
            "Epoch 75/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0270 - accuracy: 0.9920\n",
            "Epoch 76/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0266 - accuracy: 0.9916\n",
            "Epoch 77/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0320 - accuracy: 0.9909\n",
            "Epoch 78/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0304 - accuracy: 0.9919\n",
            "Epoch 79/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0272 - accuracy: 0.9915\n",
            "Epoch 80/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0290 - accuracy: 0.9909\n",
            "Epoch 81/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0292 - accuracy: 0.9912\n",
            "Epoch 82/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0272 - accuracy: 0.9917\n",
            "Epoch 83/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0297 - accuracy: 0.9919\n",
            "Epoch 84/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0228 - accuracy: 0.9942\n",
            "Epoch 85/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0270 - accuracy: 0.9919\n",
            "Epoch 86/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0261 - accuracy: 0.9920\n",
            "Epoch 87/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0240 - accuracy: 0.9930\n",
            "Epoch 88/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0212 - accuracy: 0.9943\n",
            "Epoch 89/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0288 - accuracy: 0.9916\n",
            "Epoch 90/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0261 - accuracy: 0.9926\n",
            "Epoch 91/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0271 - accuracy: 0.9910\n",
            "Epoch 92/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0253 - accuracy: 0.9930\n",
            "Epoch 93/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0250 - accuracy: 0.9920\n",
            "Epoch 94/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0253 - accuracy: 0.9919\n",
            "Epoch 95/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0302 - accuracy: 0.9904\n",
            "Epoch 96/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0239 - accuracy: 0.9923\n",
            "Epoch 97/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0216 - accuracy: 0.9939\n",
            "Epoch 98/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0261 - accuracy: 0.9916\n",
            "Epoch 99/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0231 - accuracy: 0.9929\n",
            "Epoch 100/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0228 - accuracy: 0.9928\n",
            "Epoch 101/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0229 - accuracy: 0.9922\n",
            "Epoch 102/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0243 - accuracy: 0.9941\n",
            "Epoch 103/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0250 - accuracy: 0.9917\n",
            "Epoch 104/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0297 - accuracy: 0.9903\n",
            "Epoch 105/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0200 - accuracy: 0.9951\n",
            "Epoch 106/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0268 - accuracy: 0.9916\n",
            "Epoch 107/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0221 - accuracy: 0.9926\n",
            "Epoch 108/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0228 - accuracy: 0.9929\n",
            "Epoch 109/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0233 - accuracy: 0.9938\n",
            "Epoch 110/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0246 - accuracy: 0.9926\n",
            "Epoch 111/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0244 - accuracy: 0.9930\n",
            "Epoch 112/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0221 - accuracy: 0.9935\n",
            "Epoch 113/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0235 - accuracy: 0.9930\n",
            "Epoch 114/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0235 - accuracy: 0.9928\n",
            "Epoch 115/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0277 - accuracy: 0.9916\n",
            "Epoch 116/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0257 - accuracy: 0.9928\n",
            "Epoch 117/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0237 - accuracy: 0.9930\n",
            "Epoch 118/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0257 - accuracy: 0.9932\n",
            "Epoch 119/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0175 - accuracy: 0.9942\n",
            "Epoch 120/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0291 - accuracy: 0.9909\n",
            "Epoch 121/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0165 - accuracy: 0.9954\n",
            "Epoch 122/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0254 - accuracy: 0.9917\n",
            "Epoch 123/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0240 - accuracy: 0.9930\n",
            "Epoch 124/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0174 - accuracy: 0.9945\n",
            "Epoch 125/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0220 - accuracy: 0.9942\n",
            "Epoch 126/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0199 - accuracy: 0.9938\n",
            "Epoch 127/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0188 - accuracy: 0.9948\n",
            "Epoch 128/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0303 - accuracy: 0.9897\n",
            "Epoch 129/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0186 - accuracy: 0.9933\n",
            "Epoch 130/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0228 - accuracy: 0.9929\n",
            "Epoch 131/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0242 - accuracy: 0.9919\n",
            "Epoch 132/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0214 - accuracy: 0.9930\n",
            "Epoch 133/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0231 - accuracy: 0.9926\n",
            "Epoch 134/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0163 - accuracy: 0.9946\n",
            "Epoch 135/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0207 - accuracy: 0.9933\n",
            "Epoch 136/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0234 - accuracy: 0.9935\n",
            "Epoch 137/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0254 - accuracy: 0.9920\n",
            "Epoch 138/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0206 - accuracy: 0.9930\n",
            "Epoch 139/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0235 - accuracy: 0.9933\n",
            "Epoch 140/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0210 - accuracy: 0.9925\n",
            "Epoch 141/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0158 - accuracy: 0.9945\n",
            "Epoch 142/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0202 - accuracy: 0.9938\n",
            "Epoch 143/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0236 - accuracy: 0.9928\n",
            "Epoch 144/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0197 - accuracy: 0.9948\n",
            "Epoch 145/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0242 - accuracy: 0.9925\n",
            "Epoch 146/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0215 - accuracy: 0.9933\n",
            "Epoch 147/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0195 - accuracy: 0.9948\n",
            "Epoch 148/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0218 - accuracy: 0.9935\n",
            "Epoch 149/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0197 - accuracy: 0.9930\n",
            "Epoch 150/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0199 - accuracy: 0.9941\n",
            "Epoch 151/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0153 - accuracy: 0.9955\n",
            "Epoch 152/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0195 - accuracy: 0.9935\n",
            "Epoch 153/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0236 - accuracy: 0.9915\n",
            "Epoch 154/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0207 - accuracy: 0.9923\n",
            "Epoch 155/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0173 - accuracy: 0.9955\n",
            "Epoch 156/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0185 - accuracy: 0.9933\n",
            "Epoch 157/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0214 - accuracy: 0.9928\n",
            "Epoch 158/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0166 - accuracy: 0.9946\n",
            "Epoch 159/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0224 - accuracy: 0.9922\n",
            "Epoch 160/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0169 - accuracy: 0.9946\n",
            "Epoch 161/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0193 - accuracy: 0.9943\n",
            "Epoch 162/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9951\n",
            "Epoch 163/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0168 - accuracy: 0.9955\n",
            "Epoch 164/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0205 - accuracy: 0.9938\n",
            "Epoch 165/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0146 - accuracy: 0.9946\n",
            "Epoch 166/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0186 - accuracy: 0.9942\n",
            "Epoch 167/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0164 - accuracy: 0.9954\n",
            "Epoch 168/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0191 - accuracy: 0.9936\n",
            "Epoch 169/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0168 - accuracy: 0.9942\n",
            "Epoch 170/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0190 - accuracy: 0.9932\n",
            "Epoch 171/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0191 - accuracy: 0.9928\n",
            "Epoch 172/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0185 - accuracy: 0.9928\n",
            "Epoch 173/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0174 - accuracy: 0.9941\n",
            "Epoch 174/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0147 - accuracy: 0.9946\n",
            "Epoch 175/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0195 - accuracy: 0.9936\n",
            "Epoch 176/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0209 - accuracy: 0.9941\n",
            "Epoch 177/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0197 - accuracy: 0.9936\n",
            "Epoch 178/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0200 - accuracy: 0.9932\n",
            "Epoch 179/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0168 - accuracy: 0.9946\n",
            "Epoch 180/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0182 - accuracy: 0.9930\n",
            "Epoch 181/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0163 - accuracy: 0.9948\n",
            "Epoch 182/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0191 - accuracy: 0.9945\n",
            "Epoch 183/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0188 - accuracy: 0.9942\n",
            "Epoch 184/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0190 - accuracy: 0.9939\n",
            "Epoch 185/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0164 - accuracy: 0.9955\n",
            "Epoch 186/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0158 - accuracy: 0.9952\n",
            "Epoch 187/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0188 - accuracy: 0.9942\n",
            "Epoch 188/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0153 - accuracy: 0.9948\n",
            "Epoch 189/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0155 - accuracy: 0.9939\n",
            "Epoch 190/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0206 - accuracy: 0.9936\n",
            "Epoch 191/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0182 - accuracy: 0.9943\n",
            "Epoch 192/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0189 - accuracy: 0.9938\n",
            "Epoch 193/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0163 - accuracy: 0.9957\n",
            "Epoch 194/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0193 - accuracy: 0.9939\n",
            "Epoch 195/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0149 - accuracy: 0.9958\n",
            "Epoch 196/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0212 - accuracy: 0.9930\n",
            "Epoch 197/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0169 - accuracy: 0.9941\n",
            "Epoch 198/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0159 - accuracy: 0.9952\n",
            "Epoch 199/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0153 - accuracy: 0.9957\n",
            "Epoch 200/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0150 - accuracy: 0.9941\n",
            "Epoch 201/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0125 - accuracy: 0.9961\n",
            "Epoch 202/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0169 - accuracy: 0.9951\n",
            "Epoch 203/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0150 - accuracy: 0.9951\n",
            "Epoch 204/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0195 - accuracy: 0.9941\n",
            "Epoch 205/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0158 - accuracy: 0.9958\n",
            "Epoch 206/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0182 - accuracy: 0.9932\n",
            "Epoch 207/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0163 - accuracy: 0.9957\n",
            "Epoch 208/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0137 - accuracy: 0.9951\n",
            "Epoch 209/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0150 - accuracy: 0.9951\n",
            "Epoch 210/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0183 - accuracy: 0.9939\n",
            "Epoch 211/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0181 - accuracy: 0.9936\n",
            "Epoch 212/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0201 - accuracy: 0.9930\n",
            "Epoch 213/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0208 - accuracy: 0.9929\n",
            "Epoch 214/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0133 - accuracy: 0.9959\n",
            "Epoch 215/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0180 - accuracy: 0.9942\n",
            "Epoch 216/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0230 - accuracy: 0.9930\n",
            "Epoch 217/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0156 - accuracy: 0.9955\n",
            "Epoch 218/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0162 - accuracy: 0.9945\n",
            "Epoch 219/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0236 - accuracy: 0.9923\n",
            "Epoch 220/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0157 - accuracy: 0.9946\n",
            "Epoch 221/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0151 - accuracy: 0.9958\n",
            "Epoch 222/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0135 - accuracy: 0.9957\n",
            "Epoch 223/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0166 - accuracy: 0.9951\n",
            "Epoch 224/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0157 - accuracy: 0.9949\n",
            "Epoch 225/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0207 - accuracy: 0.9933\n",
            "Epoch 226/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0219 - accuracy: 0.9929\n",
            "Epoch 227/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0155 - accuracy: 0.9952\n",
            "Epoch 228/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0135 - accuracy: 0.9958\n",
            "Epoch 229/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0180 - accuracy: 0.9938\n",
            "Epoch 230/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0160 - accuracy: 0.9946\n",
            "Epoch 231/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0158 - accuracy: 0.9951\n",
            "Epoch 232/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0166 - accuracy: 0.9952\n",
            "Epoch 233/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0187 - accuracy: 0.9946\n",
            "Epoch 234/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0164 - accuracy: 0.9942\n",
            "Epoch 235/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0177 - accuracy: 0.9952\n",
            "Epoch 236/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0140 - accuracy: 0.9964\n",
            "Epoch 237/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0146 - accuracy: 0.9955\n",
            "Epoch 238/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0162 - accuracy: 0.9946\n",
            "Epoch 239/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0152 - accuracy: 0.9946\n",
            "Epoch 240/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0161 - accuracy: 0.9935\n",
            "Epoch 241/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0175 - accuracy: 0.9941\n",
            "Epoch 242/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0144 - accuracy: 0.9952\n",
            "Epoch 243/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0181 - accuracy: 0.9942\n",
            "Epoch 244/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0165 - accuracy: 0.9951\n",
            "Epoch 245/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0191 - accuracy: 0.9936\n",
            "Epoch 246/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0134 - accuracy: 0.9952\n",
            "Epoch 247/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0146 - accuracy: 0.9959\n",
            "Epoch 248/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0157 - accuracy: 0.9943\n",
            "Epoch 249/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0168 - accuracy: 0.9936\n",
            "Epoch 250/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0112 - accuracy: 0.9967\n",
            "Epoch 251/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0195 - accuracy: 0.9941\n",
            "Epoch 252/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0179 - accuracy: 0.9949\n",
            "Epoch 253/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0164 - accuracy: 0.9941\n",
            "Epoch 254/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0106 - accuracy: 0.9967\n",
            "Epoch 255/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0160 - accuracy: 0.9957\n",
            "Epoch 256/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0148 - accuracy: 0.9958\n",
            "Epoch 257/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0180 - accuracy: 0.9938\n",
            "Epoch 258/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0156 - accuracy: 0.9962\n",
            "Epoch 259/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0204 - accuracy: 0.9935\n",
            "Epoch 260/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0145 - accuracy: 0.9948\n",
            "Epoch 261/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0149 - accuracy: 0.9952\n",
            "Epoch 262/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0205 - accuracy: 0.9943\n",
            "Epoch 263/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0148 - accuracy: 0.9955\n",
            "Epoch 264/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0177 - accuracy: 0.9948\n",
            "Epoch 265/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0199 - accuracy: 0.9930\n",
            "Epoch 266/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0130 - accuracy: 0.9955\n",
            "Epoch 267/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0196 - accuracy: 0.9932\n",
            "Epoch 268/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0181 - accuracy: 0.9948\n",
            "Epoch 269/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0173 - accuracy: 0.9949\n",
            "Epoch 270/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0146 - accuracy: 0.9951\n",
            "Epoch 271/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0162 - accuracy: 0.9954\n",
            "Epoch 272/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0179 - accuracy: 0.9938\n",
            "Epoch 273/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0188 - accuracy: 0.9938\n",
            "Epoch 274/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0171 - accuracy: 0.9942\n",
            "Epoch 275/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0176 - accuracy: 0.9954\n",
            "Epoch 276/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0130 - accuracy: 0.9955\n",
            "Epoch 277/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0145 - accuracy: 0.9945\n",
            "Epoch 278/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0126 - accuracy: 0.9962\n",
            "Epoch 279/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0127 - accuracy: 0.9952\n",
            "Epoch 280/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0148 - accuracy: 0.9958\n",
            "Epoch 281/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0177 - accuracy: 0.9936\n",
            "Epoch 282/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0183 - accuracy: 0.9935\n",
            "Epoch 283/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0164 - accuracy: 0.9948\n",
            "Epoch 284/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0123 - accuracy: 0.9965\n",
            "Epoch 285/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0170 - accuracy: 0.9939\n",
            "Epoch 286/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0114 - accuracy: 0.9964\n",
            "Epoch 287/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0247 - accuracy: 0.9929\n",
            "Epoch 288/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0123 - accuracy: 0.9967\n",
            "Epoch 289/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0123 - accuracy: 0.9958\n",
            "Epoch 290/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0124 - accuracy: 0.9958\n",
            "Epoch 291/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0156 - accuracy: 0.9961\n",
            "Epoch 292/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0163 - accuracy: 0.9945\n",
            "Epoch 293/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9959\n",
            "Epoch 294/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0175 - accuracy: 0.9945\n",
            "Epoch 295/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0126 - accuracy: 0.9952\n",
            "Epoch 296/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0152 - accuracy: 0.9945\n",
            "Epoch 297/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0141 - accuracy: 0.9957\n",
            "Epoch 298/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0152 - accuracy: 0.9955\n",
            "Epoch 299/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0155 - accuracy: 0.9945\n",
            "Epoch 300/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0131 - accuracy: 0.9961\n",
            "Epoch 301/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0125 - accuracy: 0.9955\n",
            "Epoch 302/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9958\n",
            "Epoch 303/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0134 - accuracy: 0.9957\n",
            "Epoch 304/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0112 - accuracy: 0.9968\n",
            "Epoch 305/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0158 - accuracy: 0.9939\n",
            "Epoch 306/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0140 - accuracy: 0.9951\n",
            "Epoch 307/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0167 - accuracy: 0.9957\n",
            "Epoch 308/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0155 - accuracy: 0.9949\n",
            "Epoch 309/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0163 - accuracy: 0.9938\n",
            "Epoch 310/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0186 - accuracy: 0.9933\n",
            "Epoch 311/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0156 - accuracy: 0.9949\n",
            "Epoch 312/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0114 - accuracy: 0.9959\n",
            "Epoch 313/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0161 - accuracy: 0.9957\n",
            "Epoch 314/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0138 - accuracy: 0.9961\n",
            "Epoch 315/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0141 - accuracy: 0.9942\n",
            "Epoch 316/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0124 - accuracy: 0.9948\n",
            "Epoch 317/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0097 - accuracy: 0.9970\n",
            "Epoch 318/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0163 - accuracy: 0.9942\n",
            "Epoch 319/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0205 - accuracy: 0.9939\n",
            "Epoch 320/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0125 - accuracy: 0.9952\n",
            "Epoch 321/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0143 - accuracy: 0.9943\n",
            "Epoch 322/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0165 - accuracy: 0.9945\n",
            "Epoch 323/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0139 - accuracy: 0.9946\n",
            "Epoch 324/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0177 - accuracy: 0.9942\n",
            "Epoch 325/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0131 - accuracy: 0.9948\n",
            "Epoch 326/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0144 - accuracy: 0.9955\n",
            "Epoch 327/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0151 - accuracy: 0.9946\n",
            "Epoch 328/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0143 - accuracy: 0.9952\n",
            "Epoch 329/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0141 - accuracy: 0.9961\n",
            "Epoch 330/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0145 - accuracy: 0.9955\n",
            "Epoch 331/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0158 - accuracy: 0.9938\n",
            "Epoch 332/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0125 - accuracy: 0.9946\n",
            "Epoch 333/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0110 - accuracy: 0.9958\n",
            "Epoch 334/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0128 - accuracy: 0.9954\n",
            "Epoch 335/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0153 - accuracy: 0.9948\n",
            "Epoch 336/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0138 - accuracy: 0.9957\n",
            "Epoch 337/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9952\n",
            "Epoch 338/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9945\n",
            "Epoch 339/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0127 - accuracy: 0.9957\n",
            "Epoch 340/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0171 - accuracy: 0.9938\n",
            "Epoch 341/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0129 - accuracy: 0.9955\n",
            "Epoch 342/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0144 - accuracy: 0.9961\n",
            "Epoch 343/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0129 - accuracy: 0.9958\n",
            "Epoch 344/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0158 - accuracy: 0.9942\n",
            "Epoch 345/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0141 - accuracy: 0.9943\n",
            "Epoch 346/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0129 - accuracy: 0.9965\n",
            "Epoch 347/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0194 - accuracy: 0.9945\n",
            "Epoch 348/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0169 - accuracy: 0.9942\n",
            "Epoch 349/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0133 - accuracy: 0.9958\n",
            "Epoch 350/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0105 - accuracy: 0.9970\n",
            "Epoch 351/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0117 - accuracy: 0.9959\n",
            "Epoch 352/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0129 - accuracy: 0.9959\n",
            "Epoch 353/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0175 - accuracy: 0.9942\n",
            "Epoch 354/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0085 - accuracy: 0.9971\n",
            "Epoch 355/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0176 - accuracy: 0.9945\n",
            "Epoch 356/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0118 - accuracy: 0.9958\n",
            "Epoch 357/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0158 - accuracy: 0.9935\n",
            "Epoch 358/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0140 - accuracy: 0.9959\n",
            "Epoch 359/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0154 - accuracy: 0.9949\n",
            "Epoch 360/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0174 - accuracy: 0.9936\n",
            "Epoch 361/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0114 - accuracy: 0.9961\n",
            "Epoch 362/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9959\n",
            "Epoch 363/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0137 - accuracy: 0.9954\n",
            "Epoch 364/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0112 - accuracy: 0.9951\n",
            "Epoch 365/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0135 - accuracy: 0.9952\n",
            "Epoch 366/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0115 - accuracy: 0.9968\n",
            "Epoch 367/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0139 - accuracy: 0.9955\n",
            "Epoch 368/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0140 - accuracy: 0.9954\n",
            "Epoch 369/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0159 - accuracy: 0.9949\n",
            "Epoch 370/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0152 - accuracy: 0.9946\n",
            "Epoch 371/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0125 - accuracy: 0.9954\n",
            "Epoch 372/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0121 - accuracy: 0.9959\n",
            "Epoch 373/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0170 - accuracy: 0.9957\n",
            "Epoch 374/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0162 - accuracy: 0.9949\n",
            "Epoch 375/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0121 - accuracy: 0.9954\n",
            "Epoch 376/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0135 - accuracy: 0.9964\n",
            "Epoch 377/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0206 - accuracy: 0.9930\n",
            "Epoch 378/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0166 - accuracy: 0.9941\n",
            "Epoch 379/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0149 - accuracy: 0.9948\n",
            "Epoch 380/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0121 - accuracy: 0.9965\n",
            "Epoch 381/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9961\n",
            "Epoch 382/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0162 - accuracy: 0.9949\n",
            "Epoch 383/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0100 - accuracy: 0.9974\n",
            "Epoch 384/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0106 - accuracy: 0.9965\n",
            "Epoch 385/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0105 - accuracy: 0.9964\n",
            "Epoch 386/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0153 - accuracy: 0.9957\n",
            "Epoch 387/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0164 - accuracy: 0.9949\n",
            "Epoch 388/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0154 - accuracy: 0.9951\n",
            "Epoch 389/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0195 - accuracy: 0.9928\n",
            "Epoch 390/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0116 - accuracy: 0.9968\n",
            "Epoch 391/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0129 - accuracy: 0.9957\n",
            "Epoch 392/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0123 - accuracy: 0.9955\n",
            "Epoch 393/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0135 - accuracy: 0.9949\n",
            "Epoch 394/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0104 - accuracy: 0.9967\n",
            "Epoch 395/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0161 - accuracy: 0.9954\n",
            "Epoch 396/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0165 - accuracy: 0.9949\n",
            "Epoch 397/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0113 - accuracy: 0.9958\n",
            "Epoch 398/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0114 - accuracy: 0.9959\n",
            "Epoch 399/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0131 - accuracy: 0.9955\n",
            "Epoch 400/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0112 - accuracy: 0.9964\n",
            "Epoch 401/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0114 - accuracy: 0.9959\n",
            "Epoch 402/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0140 - accuracy: 0.9952\n",
            "Epoch 403/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0145 - accuracy: 0.9948\n",
            "Epoch 404/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0128 - accuracy: 0.9954\n",
            "Epoch 405/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0203 - accuracy: 0.9930\n",
            "Epoch 406/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0080 - accuracy: 0.9981\n",
            "Epoch 407/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9964\n",
            "Epoch 408/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0144 - accuracy: 0.9961\n",
            "Epoch 409/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0111 - accuracy: 0.9957\n",
            "Epoch 410/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0136 - accuracy: 0.9955\n",
            "Epoch 411/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0148 - accuracy: 0.9948\n",
            "Epoch 412/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0129 - accuracy: 0.9965\n",
            "Epoch 413/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0119 - accuracy: 0.9962\n",
            "Epoch 414/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0133 - accuracy: 0.9951\n",
            "Epoch 415/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0197 - accuracy: 0.9948\n",
            "Epoch 416/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0080 - accuracy: 0.9974\n",
            "Epoch 417/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0092 - accuracy: 0.9971\n",
            "Epoch 418/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9957\n",
            "Epoch 419/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0106 - accuracy: 0.9962\n",
            "Epoch 420/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0109 - accuracy: 0.9961\n",
            "Epoch 421/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9961\n",
            "Epoch 422/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0127 - accuracy: 0.9957\n",
            "Epoch 423/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0131 - accuracy: 0.9957\n",
            "Epoch 424/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0123 - accuracy: 0.9961\n",
            "Epoch 425/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0163 - accuracy: 0.9948\n",
            "Epoch 426/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0110 - accuracy: 0.9971\n",
            "Epoch 427/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0171 - accuracy: 0.9941\n",
            "Epoch 428/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0116 - accuracy: 0.9961\n",
            "Epoch 429/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0132 - accuracy: 0.9948\n",
            "Epoch 430/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0163 - accuracy: 0.9941\n",
            "Epoch 431/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0097 - accuracy: 0.9964\n",
            "Epoch 432/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0149 - accuracy: 0.9954\n",
            "Epoch 433/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0184 - accuracy: 0.9939\n",
            "Epoch 434/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0077 - accuracy: 0.9977\n",
            "Epoch 435/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0174 - accuracy: 0.9945\n",
            "Epoch 436/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0141 - accuracy: 0.9962\n",
            "Epoch 437/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0138 - accuracy: 0.9949\n",
            "Epoch 438/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0096 - accuracy: 0.9972\n",
            "Epoch 439/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0094 - accuracy: 0.9967\n",
            "Epoch 440/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0110 - accuracy: 0.9971\n",
            "Epoch 441/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0155 - accuracy: 0.9951\n",
            "Epoch 442/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0120 - accuracy: 0.9961\n",
            "Epoch 443/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0165 - accuracy: 0.9955\n",
            "Epoch 444/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0103 - accuracy: 0.9970\n",
            "Epoch 445/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0107 - accuracy: 0.9961\n",
            "Epoch 446/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0138 - accuracy: 0.9952\n",
            "Epoch 447/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0117 - accuracy: 0.9965\n",
            "Epoch 448/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0132 - accuracy: 0.9946\n",
            "Epoch 449/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0098 - accuracy: 0.9968\n",
            "Epoch 450/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0096 - accuracy: 0.9965\n",
            "Epoch 451/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0115 - accuracy: 0.9964\n",
            "Epoch 452/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0126 - accuracy: 0.9957\n",
            "Epoch 453/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0162 - accuracy: 0.9941\n",
            "Epoch 454/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0067 - accuracy: 0.9980\n",
            "Epoch 455/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0194 - accuracy: 0.9938\n",
            "Epoch 456/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9954\n",
            "Epoch 457/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0114 - accuracy: 0.9964\n",
            "Epoch 458/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0115 - accuracy: 0.9970\n",
            "Epoch 459/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0140 - accuracy: 0.9951\n",
            "Epoch 460/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0121 - accuracy: 0.9957\n",
            "Epoch 461/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0124 - accuracy: 0.9962\n",
            "Epoch 462/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0142 - accuracy: 0.9961\n",
            "Epoch 463/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0114 - accuracy: 0.9951\n",
            "Epoch 464/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0117 - accuracy: 0.9962\n",
            "Epoch 465/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0115 - accuracy: 0.9959\n",
            "Epoch 466/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0112 - accuracy: 0.9967\n",
            "Epoch 467/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0117 - accuracy: 0.9965\n",
            "Epoch 468/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0113 - accuracy: 0.9965\n",
            "Epoch 469/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0115 - accuracy: 0.9964\n",
            "Epoch 470/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0074 - accuracy: 0.9974\n",
            "Epoch 471/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0078 - accuracy: 0.9970\n",
            "Epoch 472/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0100 - accuracy: 0.9970\n",
            "Epoch 473/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0123 - accuracy: 0.9955\n",
            "Epoch 474/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0088 - accuracy: 0.9967\n",
            "Epoch 475/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9967\n",
            "Epoch 476/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0128 - accuracy: 0.9962\n",
            "Epoch 477/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0200 - accuracy: 0.9928\n",
            "Epoch 478/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0122 - accuracy: 0.9961\n",
            "Epoch 479/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0122 - accuracy: 0.9961\n",
            "Epoch 480/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0194 - accuracy: 0.9943\n",
            "Epoch 481/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9965\n",
            "Epoch 482/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0133 - accuracy: 0.9957\n",
            "Epoch 483/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0154 - accuracy: 0.9949\n",
            "Epoch 484/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0150 - accuracy: 0.9957\n",
            "Epoch 485/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0145 - accuracy: 0.9951\n",
            "Epoch 486/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0106 - accuracy: 0.9958\n",
            "Epoch 487/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0079 - accuracy: 0.9977\n",
            "Epoch 488/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0177 - accuracy: 0.9948\n",
            "Epoch 489/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0102 - accuracy: 0.9970\n",
            "Epoch 490/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0135 - accuracy: 0.9952\n",
            "Epoch 491/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0113 - accuracy: 0.9961\n",
            "Epoch 492/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0102 - accuracy: 0.9975\n",
            "Epoch 493/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0133 - accuracy: 0.9952\n",
            "Epoch 494/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0114 - accuracy: 0.9957\n",
            "Epoch 495/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0095 - accuracy: 0.9961\n",
            "Epoch 496/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0079 - accuracy: 0.9968\n",
            "Epoch 497/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0093 - accuracy: 0.9964\n",
            "Epoch 498/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0090 - accuracy: 0.9971\n",
            "Epoch 499/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0105 - accuracy: 0.9964\n",
            "Epoch 500/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0149 - accuracy: 0.9948\n",
            "Epoch 501/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0123 - accuracy: 0.9967\n",
            "Epoch 502/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0192 - accuracy: 0.9929\n",
            "Epoch 503/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0101 - accuracy: 0.9961\n",
            "Epoch 504/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0094 - accuracy: 0.9974\n",
            "Epoch 505/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0088 - accuracy: 0.9968\n",
            "Epoch 506/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0144 - accuracy: 0.9954\n",
            "Epoch 507/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0089 - accuracy: 0.9972\n",
            "Epoch 508/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0116 - accuracy: 0.9962\n",
            "Epoch 509/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0123 - accuracy: 0.9959\n",
            "Epoch 510/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0076 - accuracy: 0.9972\n",
            "Epoch 511/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0089 - accuracy: 0.9971\n",
            "Epoch 512/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0107 - accuracy: 0.9970\n",
            "Epoch 513/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0133 - accuracy: 0.9959\n",
            "Epoch 514/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0122 - accuracy: 0.9961\n",
            "Epoch 515/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9965\n",
            "Epoch 516/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0129 - accuracy: 0.9955\n",
            "Epoch 517/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0134 - accuracy: 0.9964\n",
            "Epoch 518/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0113 - accuracy: 0.9967\n",
            "Epoch 519/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0130 - accuracy: 0.9965\n",
            "Epoch 520/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0096 - accuracy: 0.9968\n",
            "Epoch 521/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0105 - accuracy: 0.9971\n",
            "Epoch 522/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0109 - accuracy: 0.9968\n",
            "Epoch 523/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0109 - accuracy: 0.9972\n",
            "Epoch 524/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0120 - accuracy: 0.9955\n",
            "Epoch 525/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0101 - accuracy: 0.9961\n",
            "Epoch 526/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0109 - accuracy: 0.9961\n",
            "Epoch 527/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0094 - accuracy: 0.9967\n",
            "Epoch 528/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0069 - accuracy: 0.9972\n",
            "Epoch 529/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0095 - accuracy: 0.9964\n",
            "Epoch 530/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0126 - accuracy: 0.9954\n",
            "Epoch 531/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0153 - accuracy: 0.9952\n",
            "Epoch 532/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0098 - accuracy: 0.9961\n",
            "Epoch 533/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0102 - accuracy: 0.9965\n",
            "Epoch 534/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0074 - accuracy: 0.9971\n",
            "Epoch 535/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0129 - accuracy: 0.9961\n",
            "Epoch 536/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0089 - accuracy: 0.9970\n",
            "Epoch 537/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0101 - accuracy: 0.9967\n",
            "Epoch 538/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0089 - accuracy: 0.9967\n",
            "Epoch 539/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0108 - accuracy: 0.9970\n",
            "Epoch 540/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0140 - accuracy: 0.9957\n",
            "Epoch 541/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0067 - accuracy: 0.9974\n",
            "Epoch 542/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0104 - accuracy: 0.9967\n",
            "Epoch 543/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0088 - accuracy: 0.9964\n",
            "Epoch 544/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0077 - accuracy: 0.9967\n",
            "Epoch 545/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0215 - accuracy: 0.9930\n",
            "Epoch 546/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0117 - accuracy: 0.9951\n",
            "Epoch 547/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0118 - accuracy: 0.9954\n",
            "Epoch 548/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0113 - accuracy: 0.9957\n",
            "Epoch 549/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0109 - accuracy: 0.9967\n",
            "Epoch 550/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0131 - accuracy: 0.9946\n",
            "Epoch 551/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0135 - accuracy: 0.9946\n",
            "Epoch 552/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0077 - accuracy: 0.9974\n",
            "Epoch 553/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0114 - accuracy: 0.9959\n",
            "Epoch 554/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0083 - accuracy: 0.9967\n",
            "Epoch 555/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9964\n",
            "Epoch 556/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0105 - accuracy: 0.9962\n",
            "Epoch 557/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0073 - accuracy: 0.9972\n",
            "Epoch 558/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0170 - accuracy: 0.9939\n",
            "Epoch 559/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0109 - accuracy: 0.9962\n",
            "Epoch 560/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0115 - accuracy: 0.9970\n",
            "Epoch 561/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0148 - accuracy: 0.9954\n",
            "Epoch 562/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9965\n",
            "Epoch 563/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0067 - accuracy: 0.9972\n",
            "Epoch 564/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0142 - accuracy: 0.9961\n",
            "Epoch 565/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0128 - accuracy: 0.9955\n",
            "Epoch 566/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0104 - accuracy: 0.9965\n",
            "Epoch 567/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0117 - accuracy: 0.9957\n",
            "Epoch 568/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0097 - accuracy: 0.9961\n",
            "Epoch 569/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0123 - accuracy: 0.9964\n",
            "Epoch 570/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0095 - accuracy: 0.9977\n",
            "Epoch 571/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0132 - accuracy: 0.9949\n",
            "Epoch 572/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0141 - accuracy: 0.9949\n",
            "Epoch 573/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0086 - accuracy: 0.9968\n",
            "Epoch 574/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9965\n",
            "Epoch 575/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0072 - accuracy: 0.9972\n",
            "Epoch 576/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0085 - accuracy: 0.9964\n",
            "Epoch 577/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0079 - accuracy: 0.9974\n",
            "Epoch 578/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0100 - accuracy: 0.9964\n",
            "Epoch 579/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0143 - accuracy: 0.9952\n",
            "Epoch 580/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0092 - accuracy: 0.9968\n",
            "Epoch 581/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0103 - accuracy: 0.9962\n",
            "Epoch 582/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0111 - accuracy: 0.9964\n",
            "Epoch 583/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0084 - accuracy: 0.9970\n",
            "Epoch 584/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0080 - accuracy: 0.9970\n",
            "Epoch 585/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0098 - accuracy: 0.9961\n",
            "Epoch 586/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0068 - accuracy: 0.9978\n",
            "Epoch 587/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0087 - accuracy: 0.9968\n",
            "Epoch 588/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0073 - accuracy: 0.9975\n",
            "Epoch 589/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0115 - accuracy: 0.9957\n",
            "Epoch 590/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0095 - accuracy: 0.9962\n",
            "Epoch 591/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0100 - accuracy: 0.9957\n",
            "Epoch 592/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0120 - accuracy: 0.9964\n",
            "Epoch 593/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0093 - accuracy: 0.9970\n",
            "Epoch 594/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0068 - accuracy: 0.9974\n",
            "Epoch 595/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0068 - accuracy: 0.9972\n",
            "Epoch 596/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0140 - accuracy: 0.9957\n",
            "Epoch 597/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0100 - accuracy: 0.9968\n",
            "Epoch 598/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0096 - accuracy: 0.9970\n",
            "Epoch 599/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0109 - accuracy: 0.9967\n",
            "Epoch 600/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0126 - accuracy: 0.9952\n",
            "Epoch 601/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0111 - accuracy: 0.9959\n",
            "Epoch 602/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0091 - accuracy: 0.9964\n",
            "Epoch 603/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0088 - accuracy: 0.9975\n",
            "Epoch 604/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0137 - accuracy: 0.9955\n",
            "Epoch 605/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0115 - accuracy: 0.9968\n",
            "Epoch 606/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0065 - accuracy: 0.9983\n",
            "Epoch 607/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0056 - accuracy: 0.9983\n",
            "Epoch 608/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0072 - accuracy: 0.9965\n",
            "Epoch 609/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9970\n",
            "Epoch 610/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0176 - accuracy: 0.9946\n",
            "Epoch 611/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0128 - accuracy: 0.9954\n",
            "Epoch 612/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0095 - accuracy: 0.9962\n",
            "Epoch 613/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0136 - accuracy: 0.9957\n",
            "Epoch 614/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0066 - accuracy: 0.9978\n",
            "Epoch 615/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0130 - accuracy: 0.9957\n",
            "Epoch 616/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0144 - accuracy: 0.9946\n",
            "Epoch 617/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0079 - accuracy: 0.9970\n",
            "Epoch 618/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0080 - accuracy: 0.9971\n",
            "Epoch 619/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0114 - accuracy: 0.9958\n",
            "Epoch 620/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0062 - accuracy: 0.9981\n",
            "Epoch 621/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0073 - accuracy: 0.9970\n",
            "Epoch 622/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0070 - accuracy: 0.9980\n",
            "Epoch 623/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0089 - accuracy: 0.9970\n",
            "Epoch 624/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0134 - accuracy: 0.9957\n",
            "Epoch 625/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0126 - accuracy: 0.9951\n",
            "Epoch 626/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0063 - accuracy: 0.9974\n",
            "Epoch 627/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0107 - accuracy: 0.9967\n",
            "Epoch 628/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0088 - accuracy: 0.9965\n",
            "Epoch 629/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0089 - accuracy: 0.9971\n",
            "Epoch 630/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0130 - accuracy: 0.9954\n",
            "Epoch 631/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0098 - accuracy: 0.9957\n",
            "Epoch 632/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0104 - accuracy: 0.9959\n",
            "Epoch 633/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0113 - accuracy: 0.9962\n",
            "Epoch 634/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0158 - accuracy: 0.9942\n",
            "Epoch 635/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0103 - accuracy: 0.9967\n",
            "Epoch 636/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0062 - accuracy: 0.9984\n",
            "Epoch 637/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0096 - accuracy: 0.9971\n",
            "Epoch 638/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0124 - accuracy: 0.9952\n",
            "Epoch 639/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0101 - accuracy: 0.9968\n",
            "Epoch 640/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0068 - accuracy: 0.9978\n",
            "Epoch 641/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0067 - accuracy: 0.9975\n",
            "Epoch 642/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0120 - accuracy: 0.9964\n",
            "Epoch 643/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0122 - accuracy: 0.9964\n",
            "Epoch 644/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0100 - accuracy: 0.9964\n",
            "Epoch 645/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0089 - accuracy: 0.9967\n",
            "Epoch 646/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0117 - accuracy: 0.9972\n",
            "Epoch 647/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0097 - accuracy: 0.9965\n",
            "Epoch 648/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0078 - accuracy: 0.9974\n",
            "Epoch 649/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0093 - accuracy: 0.9965\n",
            "Epoch 650/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0095 - accuracy: 0.9968\n",
            "Epoch 651/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0120 - accuracy: 0.9957\n",
            "Epoch 652/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0133 - accuracy: 0.9958\n",
            "Epoch 653/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0165 - accuracy: 0.9951\n",
            "Epoch 654/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0108 - accuracy: 0.9961\n",
            "Epoch 655/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0077 - accuracy: 0.9965\n",
            "Epoch 656/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0090 - accuracy: 0.9964\n",
            "Epoch 657/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0230 - accuracy: 0.9941\n",
            "Epoch 658/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0147 - accuracy: 0.9957\n",
            "Epoch 659/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0111 - accuracy: 0.9961\n",
            "Epoch 660/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0124 - accuracy: 0.9962\n",
            "Epoch 661/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0074 - accuracy: 0.9970\n",
            "Epoch 662/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0100 - accuracy: 0.9968\n",
            "Epoch 663/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0101 - accuracy: 0.9968\n",
            "Epoch 664/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0129 - accuracy: 0.9954\n",
            "Epoch 665/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0110 - accuracy: 0.9961\n",
            "Epoch 666/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0082 - accuracy: 0.9971\n",
            "Epoch 667/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0045 - accuracy: 0.9983\n",
            "Epoch 668/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0128 - accuracy: 0.9951\n",
            "Epoch 669/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0086 - accuracy: 0.9970\n",
            "Epoch 670/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0088 - accuracy: 0.9967\n",
            "Epoch 671/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0122 - accuracy: 0.9952\n",
            "Epoch 672/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0083 - accuracy: 0.9972\n",
            "Epoch 673/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9975\n",
            "Epoch 674/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0053 - accuracy: 0.9981\n",
            "Epoch 675/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0078 - accuracy: 0.9972\n",
            "Epoch 676/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0139 - accuracy: 0.9957\n",
            "Epoch 677/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0053 - accuracy: 0.9983\n",
            "Epoch 678/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0085 - accuracy: 0.9978\n",
            "Epoch 679/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0079 - accuracy: 0.9972\n",
            "Epoch 680/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0092 - accuracy: 0.9967\n",
            "Epoch 681/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0043 - accuracy: 0.9986\n",
            "Epoch 682/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0097 - accuracy: 0.9977\n",
            "Epoch 683/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9965\n",
            "Epoch 684/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0150 - accuracy: 0.9959\n",
            "Epoch 685/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0092 - accuracy: 0.9965\n",
            "Epoch 686/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0091 - accuracy: 0.9971\n",
            "Epoch 687/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0115 - accuracy: 0.9967\n",
            "Epoch 688/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0096 - accuracy: 0.9967\n",
            "Epoch 689/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0063 - accuracy: 0.9971\n",
            "Epoch 690/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0103 - accuracy: 0.9965\n",
            "Epoch 691/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0108 - accuracy: 0.9958\n",
            "Epoch 692/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0126 - accuracy: 0.9962\n",
            "Epoch 693/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0071 - accuracy: 0.9980\n",
            "Epoch 694/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0187 - accuracy: 0.9935\n",
            "Epoch 695/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0084 - accuracy: 0.9971\n",
            "Epoch 696/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0087 - accuracy: 0.9970\n",
            "Epoch 697/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0082 - accuracy: 0.9968\n",
            "Epoch 698/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0088 - accuracy: 0.9965\n",
            "Epoch 699/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0105 - accuracy: 0.9970\n",
            "Epoch 700/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0045 - accuracy: 0.9984\n",
            "Epoch 701/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0091 - accuracy: 0.9971\n",
            "Epoch 702/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0218 - accuracy: 0.9929\n",
            "Epoch 703/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0135 - accuracy: 0.9952\n",
            "Epoch 704/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0139 - accuracy: 0.9955\n",
            "Epoch 705/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0090 - accuracy: 0.9965\n",
            "Epoch 706/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0083 - accuracy: 0.9977\n",
            "Epoch 707/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0094 - accuracy: 0.9958\n",
            "Epoch 708/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9959\n",
            "Epoch 709/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0082 - accuracy: 0.9975\n",
            "Epoch 710/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0089 - accuracy: 0.9972\n",
            "Epoch 711/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0107 - accuracy: 0.9961\n",
            "Epoch 712/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0105 - accuracy: 0.9959\n",
            "Epoch 713/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0098 - accuracy: 0.9968\n",
            "Epoch 714/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0109 - accuracy: 0.9958\n",
            "Epoch 715/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0107 - accuracy: 0.9957\n",
            "Epoch 716/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0160 - accuracy: 0.9942\n",
            "Epoch 717/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0053 - accuracy: 0.9990\n",
            "Epoch 718/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0145 - accuracy: 0.9952\n",
            "Epoch 719/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0085 - accuracy: 0.9970\n",
            "Epoch 720/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0060 - accuracy: 0.9980\n",
            "Epoch 721/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0059 - accuracy: 0.9980\n",
            "Epoch 722/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0090 - accuracy: 0.9962\n",
            "Epoch 723/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0092 - accuracy: 0.9974\n",
            "Epoch 724/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0169 - accuracy: 0.9943\n",
            "Epoch 725/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0093 - accuracy: 0.9968\n",
            "Epoch 726/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0125 - accuracy: 0.9959\n",
            "Epoch 727/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0060 - accuracy: 0.9975\n",
            "Epoch 728/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0103 - accuracy: 0.9961\n",
            "Epoch 729/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0109 - accuracy: 0.9958\n",
            "Epoch 730/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0096 - accuracy: 0.9964\n",
            "Epoch 731/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0053 - accuracy: 0.9981\n",
            "Epoch 732/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0154 - accuracy: 0.9955\n",
            "Epoch 733/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0120 - accuracy: 0.9955\n",
            "Epoch 734/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0151 - accuracy: 0.9948\n",
            "Epoch 735/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0096 - accuracy: 0.9965\n",
            "Epoch 736/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0090 - accuracy: 0.9964\n",
            "Epoch 737/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0142 - accuracy: 0.9946\n",
            "Epoch 738/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0100 - accuracy: 0.9961\n",
            "Epoch 739/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0090 - accuracy: 0.9959\n",
            "Epoch 740/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0099 - accuracy: 0.9972\n",
            "Epoch 741/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0113 - accuracy: 0.9962\n",
            "Epoch 742/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0115 - accuracy: 0.9951\n",
            "Epoch 743/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0126 - accuracy: 0.9954\n",
            "Epoch 744/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0105 - accuracy: 0.9965\n",
            "Epoch 745/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0077 - accuracy: 0.9975\n",
            "Epoch 746/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0101 - accuracy: 0.9959\n",
            "Epoch 747/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0152 - accuracy: 0.9954\n",
            "Epoch 748/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9964\n",
            "Epoch 749/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0107 - accuracy: 0.9962\n",
            "Epoch 750/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0085 - accuracy: 0.9971\n",
            "Epoch 751/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0081 - accuracy: 0.9978\n",
            "Epoch 752/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0078 - accuracy: 0.9975\n",
            "Epoch 753/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0109 - accuracy: 0.9958\n",
            "Epoch 754/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0170 - accuracy: 0.9939\n",
            "Epoch 755/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0057 - accuracy: 0.9980\n",
            "Epoch 756/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0073 - accuracy: 0.9978\n",
            "Epoch 757/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0104 - accuracy: 0.9961\n",
            "Epoch 758/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0088 - accuracy: 0.9970\n",
            "Epoch 759/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0093 - accuracy: 0.9965\n",
            "Epoch 760/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0118 - accuracy: 0.9952\n",
            "Epoch 761/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0107 - accuracy: 0.9965\n",
            "Epoch 762/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0102 - accuracy: 0.9962\n",
            "Epoch 763/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0102 - accuracy: 0.9964\n",
            "Epoch 764/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0112 - accuracy: 0.9961\n",
            "Epoch 765/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0092 - accuracy: 0.9972\n",
            "Epoch 766/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0099 - accuracy: 0.9967\n",
            "Epoch 767/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0090 - accuracy: 0.9972\n",
            "Epoch 768/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0103 - accuracy: 0.9965\n",
            "Epoch 769/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0132 - accuracy: 0.9952\n",
            "Epoch 770/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0091 - accuracy: 0.9964\n",
            "Epoch 771/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0084 - accuracy: 0.9972\n",
            "Epoch 772/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0045 - accuracy: 0.9991\n",
            "Epoch 773/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0129 - accuracy: 0.9948\n",
            "Epoch 774/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0049 - accuracy: 0.9986\n",
            "Epoch 775/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0054 - accuracy: 0.9981\n",
            "Epoch 776/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0092 - accuracy: 0.9980\n",
            "Epoch 777/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0116 - accuracy: 0.9952\n",
            "Epoch 778/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0098 - accuracy: 0.9964\n",
            "Epoch 779/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0110 - accuracy: 0.9965\n",
            "Epoch 780/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0094 - accuracy: 0.9957\n",
            "Epoch 781/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0096 - accuracy: 0.9972\n",
            "Epoch 782/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0152 - accuracy: 0.9942\n",
            "Epoch 783/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0087 - accuracy: 0.9970\n",
            "Epoch 784/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0075 - accuracy: 0.9974\n",
            "Epoch 785/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0137 - accuracy: 0.9968\n",
            "Epoch 786/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0098 - accuracy: 0.9959\n",
            "Epoch 787/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0075 - accuracy: 0.9970\n",
            "Epoch 788/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0099 - accuracy: 0.9962\n",
            "Epoch 789/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0115 - accuracy: 0.9955\n",
            "Epoch 790/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0095 - accuracy: 0.9970\n",
            "Epoch 791/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0050 - accuracy: 0.9984\n",
            "Epoch 792/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0097 - accuracy: 0.9959\n",
            "Epoch 793/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0078 - accuracy: 0.9972\n",
            "Epoch 794/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9961\n",
            "Epoch 795/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0057 - accuracy: 0.9975\n",
            "Epoch 796/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0068 - accuracy: 0.9978\n",
            "Epoch 797/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0091 - accuracy: 0.9977\n",
            "Epoch 798/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0087 - accuracy: 0.9970\n",
            "Epoch 799/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0106 - accuracy: 0.9967\n",
            "Epoch 800/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0102 - accuracy: 0.9958\n",
            "Epoch 801/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0105 - accuracy: 0.9955\n",
            "Epoch 802/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0105 - accuracy: 0.9968\n",
            "Epoch 803/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0101 - accuracy: 0.9957\n",
            "Epoch 804/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0096 - accuracy: 0.9967\n",
            "Epoch 805/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0140 - accuracy: 0.9954\n",
            "Epoch 806/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0122 - accuracy: 0.9959\n",
            "Epoch 807/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0134 - accuracy: 0.9954\n",
            "Epoch 808/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0095 - accuracy: 0.9968\n",
            "Epoch 809/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0114 - accuracy: 0.9964\n",
            "Epoch 810/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0118 - accuracy: 0.9952\n",
            "Epoch 811/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0061 - accuracy: 0.9986\n",
            "Epoch 812/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0101 - accuracy: 0.9972\n",
            "Epoch 813/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9957\n",
            "Epoch 814/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0161 - accuracy: 0.9954\n",
            "Epoch 815/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0124 - accuracy: 0.9955\n",
            "Epoch 816/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0053 - accuracy: 0.9986\n",
            "Epoch 817/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0078 - accuracy: 0.9977\n",
            "Epoch 818/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0069 - accuracy: 0.9981\n",
            "Epoch 819/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0075 - accuracy: 0.9972\n",
            "Epoch 820/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0093 - accuracy: 0.9970\n",
            "Epoch 821/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0085 - accuracy: 0.9974\n",
            "Epoch 822/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0108 - accuracy: 0.9959\n",
            "Epoch 823/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0088 - accuracy: 0.9968\n",
            "Epoch 824/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0106 - accuracy: 0.9965\n",
            "Epoch 825/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0076 - accuracy: 0.9968\n",
            "Epoch 826/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0081 - accuracy: 0.9964\n",
            "Epoch 827/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0142 - accuracy: 0.9945\n",
            "Epoch 828/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0122 - accuracy: 0.9952\n",
            "Epoch 829/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0071 - accuracy: 0.9975\n",
            "Epoch 830/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0135 - accuracy: 0.9958\n",
            "Epoch 831/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0067 - accuracy: 0.9978\n",
            "Epoch 832/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0079 - accuracy: 0.9971\n",
            "Epoch 833/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0077 - accuracy: 0.9977\n",
            "Epoch 834/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0125 - accuracy: 0.9945\n",
            "Epoch 835/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0086 - accuracy: 0.9959\n",
            "Epoch 836/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0094 - accuracy: 0.9957\n",
            "Epoch 837/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0080 - accuracy: 0.9968\n",
            "Epoch 838/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0063 - accuracy: 0.9977\n",
            "Epoch 839/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0059 - accuracy: 0.9988\n",
            "Epoch 840/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0076 - accuracy: 0.9968\n",
            "Epoch 841/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0048 - accuracy: 0.9991\n",
            "Epoch 842/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0104 - accuracy: 0.9962\n",
            "Epoch 843/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0123 - accuracy: 0.9957\n",
            "Epoch 844/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0085 - accuracy: 0.9972\n",
            "Epoch 845/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9965\n",
            "Epoch 846/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0082 - accuracy: 0.9977\n",
            "Epoch 847/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0044 - accuracy: 0.9987\n",
            "Epoch 848/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9943\n",
            "Epoch 849/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0095 - accuracy: 0.9959\n",
            "Epoch 850/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0078 - accuracy: 0.9971\n",
            "Epoch 851/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0098 - accuracy: 0.9975\n",
            "Epoch 852/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0124 - accuracy: 0.9958\n",
            "Epoch 853/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0106 - accuracy: 0.9974\n",
            "Epoch 854/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9962\n",
            "Epoch 855/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0087 - accuracy: 0.9967\n",
            "Epoch 856/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0104 - accuracy: 0.9959\n",
            "Epoch 857/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0059 - accuracy: 0.9980\n",
            "Epoch 858/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0058 - accuracy: 0.9978\n",
            "Epoch 859/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0062 - accuracy: 0.9974\n",
            "Epoch 860/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0141 - accuracy: 0.9959\n",
            "Epoch 861/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0081 - accuracy: 0.9972\n",
            "Epoch 862/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0145 - accuracy: 0.9948\n",
            "Epoch 863/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0124 - accuracy: 0.9949\n",
            "Epoch 864/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0035 - accuracy: 0.9988\n",
            "Epoch 865/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0073 - accuracy: 0.9974\n",
            "Epoch 866/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0123 - accuracy: 0.9954\n",
            "Epoch 867/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0082 - accuracy: 0.9975\n",
            "Epoch 868/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0118 - accuracy: 0.9954\n",
            "Epoch 869/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0092 - accuracy: 0.9958\n",
            "Epoch 870/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0126 - accuracy: 0.9961\n",
            "Epoch 871/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0098 - accuracy: 0.9968\n",
            "Epoch 872/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0078 - accuracy: 0.9974\n",
            "Epoch 873/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0064 - accuracy: 0.9978\n",
            "Epoch 874/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0086 - accuracy: 0.9964\n",
            "Epoch 875/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0145 - accuracy: 0.9954\n",
            "Epoch 876/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0068 - accuracy: 0.9972\n",
            "Epoch 877/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0065 - accuracy: 0.9983\n",
            "Epoch 878/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0084 - accuracy: 0.9970\n",
            "Epoch 879/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0086 - accuracy: 0.9965\n",
            "Epoch 880/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0067 - accuracy: 0.9971\n",
            "Epoch 881/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0053 - accuracy: 0.9981\n",
            "Epoch 882/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0093 - accuracy: 0.9974\n",
            "Epoch 883/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0132 - accuracy: 0.9959\n",
            "Epoch 884/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0125 - accuracy: 0.9957\n",
            "Epoch 885/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0106 - accuracy: 0.9957\n",
            "Epoch 886/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0061 - accuracy: 0.9983\n",
            "Epoch 887/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0138 - accuracy: 0.9952\n",
            "Epoch 888/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9955\n",
            "Epoch 889/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0084 - accuracy: 0.9978\n",
            "Epoch 890/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0088 - accuracy: 0.9967\n",
            "Epoch 891/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0125 - accuracy: 0.9961\n",
            "Epoch 892/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0086 - accuracy: 0.9970\n",
            "Epoch 893/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0101 - accuracy: 0.9970\n",
            "Epoch 894/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0063 - accuracy: 0.9978\n",
            "Epoch 895/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0066 - accuracy: 0.9975\n",
            "Epoch 896/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0062 - accuracy: 0.9974\n",
            "Epoch 897/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0076 - accuracy: 0.9972\n",
            "Epoch 898/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0107 - accuracy: 0.9968\n",
            "Epoch 899/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0143 - accuracy: 0.9949\n",
            "Epoch 900/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0072 - accuracy: 0.9975\n",
            "Epoch 901/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0064 - accuracy: 0.9977\n",
            "Epoch 902/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0116 - accuracy: 0.9955\n",
            "Epoch 903/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0076 - accuracy: 0.9972\n",
            "Epoch 904/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0085 - accuracy: 0.9972\n",
            "Epoch 905/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0073 - accuracy: 0.9980\n",
            "Epoch 906/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0061 - accuracy: 0.9977\n",
            "Epoch 907/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0086 - accuracy: 0.9965\n",
            "Epoch 908/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0083 - accuracy: 0.9971\n",
            "Epoch 909/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0091 - accuracy: 0.9958\n",
            "Epoch 910/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0073 - accuracy: 0.9974\n",
            "Epoch 911/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0123 - accuracy: 0.9962\n",
            "Epoch 912/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0070 - accuracy: 0.9972\n",
            "Epoch 913/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0156 - accuracy: 0.9942\n",
            "Epoch 914/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0089 - accuracy: 0.9964\n",
            "Epoch 915/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0089 - accuracy: 0.9972\n",
            "Epoch 916/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0107 - accuracy: 0.9971\n",
            "Epoch 917/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0084 - accuracy: 0.9971\n",
            "Epoch 918/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0131 - accuracy: 0.9964\n",
            "Epoch 919/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9954\n",
            "Epoch 920/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9968\n",
            "Epoch 921/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0052 - accuracy: 0.9974\n",
            "Epoch 922/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0115 - accuracy: 0.9962\n",
            "Epoch 923/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0063 - accuracy: 0.9983\n",
            "Epoch 924/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0122 - accuracy: 0.9967\n",
            "Epoch 925/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0097 - accuracy: 0.9961\n",
            "Epoch 926/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0087 - accuracy: 0.9972\n",
            "Epoch 927/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0088 - accuracy: 0.9972\n",
            "Epoch 928/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0073 - accuracy: 0.9977\n",
            "Epoch 929/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0080 - accuracy: 0.9975\n",
            "Epoch 930/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0130 - accuracy: 0.9951\n",
            "Epoch 931/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0076 - accuracy: 0.9972\n",
            "Epoch 932/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0103 - accuracy: 0.9965\n",
            "Epoch 933/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0106 - accuracy: 0.9958\n",
            "Epoch 934/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0125 - accuracy: 0.9964\n",
            "Epoch 935/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0079 - accuracy: 0.9977\n",
            "Epoch 936/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0114 - accuracy: 0.9962\n",
            "Epoch 937/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0120 - accuracy: 0.9958\n",
            "Epoch 938/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0132 - accuracy: 0.9951\n",
            "Epoch 939/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0084 - accuracy: 0.9971\n",
            "Epoch 940/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0080 - accuracy: 0.9968\n",
            "Epoch 941/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0083 - accuracy: 0.9975\n",
            "Epoch 942/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0081 - accuracy: 0.9967\n",
            "Epoch 943/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0072 - accuracy: 0.9974\n",
            "Epoch 944/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0083 - accuracy: 0.9975\n",
            "Epoch 945/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0096 - accuracy: 0.9971\n",
            "Epoch 946/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0069 - accuracy: 0.9971\n",
            "Epoch 947/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0049 - accuracy: 0.9980\n",
            "Epoch 948/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0105 - accuracy: 0.9961\n",
            "Epoch 949/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0126 - accuracy: 0.9945\n",
            "Epoch 950/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0106 - accuracy: 0.9970\n",
            "Epoch 951/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0093 - accuracy: 0.9967\n",
            "Epoch 952/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0117 - accuracy: 0.9962\n",
            "Epoch 953/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0106 - accuracy: 0.9961\n",
            "Epoch 954/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0121 - accuracy: 0.9968\n",
            "Epoch 955/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0086 - accuracy: 0.9967\n",
            "Epoch 956/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0114 - accuracy: 0.9957\n",
            "Epoch 957/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0083 - accuracy: 0.9970\n",
            "Epoch 958/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0090 - accuracy: 0.9965\n",
            "Epoch 959/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0114 - accuracy: 0.9948\n",
            "Epoch 960/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0106 - accuracy: 0.9958\n",
            "Epoch 961/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0086 - accuracy: 0.9970\n",
            "Epoch 962/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0103 - accuracy: 0.9971\n",
            "Epoch 963/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0101 - accuracy: 0.9964\n",
            "Epoch 964/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0065 - accuracy: 0.9972\n",
            "Epoch 965/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0107 - accuracy: 0.9970\n",
            "Epoch 966/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0124 - accuracy: 0.9949\n",
            "Epoch 967/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0102 - accuracy: 0.9957\n",
            "Epoch 968/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0088 - accuracy: 0.9972\n",
            "Epoch 969/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0102 - accuracy: 0.9964\n",
            "Epoch 970/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0130 - accuracy: 0.9957\n",
            "Epoch 971/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0088 - accuracy: 0.9971\n",
            "Epoch 972/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0056 - accuracy: 0.9981\n",
            "Epoch 973/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0079 - accuracy: 0.9964\n",
            "Epoch 974/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0113 - accuracy: 0.9967\n",
            "Epoch 975/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0087 - accuracy: 0.9970\n",
            "Epoch 976/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0071 - accuracy: 0.9974\n",
            "Epoch 977/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0076 - accuracy: 0.9971\n",
            "Epoch 978/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0077 - accuracy: 0.9971\n",
            "Epoch 979/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0049 - accuracy: 0.9983\n",
            "Epoch 980/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0099 - accuracy: 0.9961\n",
            "Epoch 981/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0084 - accuracy: 0.9970\n",
            "Epoch 982/987\n",
            "216/216 [==============================] - 1s 3ms/step - loss: 0.0076 - accuracy: 0.9967\n",
            "Epoch 983/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0119 - accuracy: 0.9967\n",
            "Epoch 984/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0129 - accuracy: 0.9958\n",
            "Epoch 985/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0064 - accuracy: 0.9975\n",
            "Epoch 986/987\n",
            "216/216 [==============================] - 0s 2ms/step - loss: 0.0076 - accuracy: 0.9968\n",
            "Epoch 987/987\n",
            "216/216 [==============================] - 1s 2ms/step - loss: 0.0069 - accuracy: 0.9974\n",
            "INFO:tensorflow:Assets written to: DTCP/assets\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dkOYzKinq5sh",
        "outputId": "6f787197-b858-49d8-836a-c1d548f92a29"
      },
      "source": [
        "model.evaluate(training_eval_x,training_eval_y)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "47/47 [==============================] - 0s 1ms/step - loss: 0.0195 - accuracy: 0.9946\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.019548757001757622, 0.9946344494819641]"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RrIwMF2vq7a8",
        "outputId": "7d296dbf-6776-4431-cbd9-9af6e891b3a0"
      },
      "source": [
        "model.layers[4].weights"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[]"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "odmVYI2Eq_nw",
        "outputId": "141c878f-39c2-46ad-8bce-7adcfae56309"
      },
      "source": [
        "print(model.trainable_variables) "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[<tf.Variable 'dense/kernel:0' shape=(5, 20) dtype=float32, numpy=\n",
            "array([[ 1.2568754e+00,  3.2954230e+00, -2.1101822e-01, -1.8991193e+00,\n",
            "         2.0779605e+00, -1.4748269e+00, -1.5639724e+00,  2.8569624e-01,\n",
            "         3.0372572e+00, -1.2590669e+00,  1.8296706e-02,  2.4473085e-01,\n",
            "        -3.1412533e-01,  2.4580088e+00, -3.6297286e-01, -1.3986223e+00,\n",
            "        -1.2096174e+00, -1.4558980e+00, -1.6393703e+00, -5.1419002e-01],\n",
            "       [-1.4407817e+00, -7.2510630e-01,  3.3513077e-02,  1.0197828e+00,\n",
            "        -5.7455856e-01,  1.1235126e+00,  1.1690556e+00, -4.7079492e-01,\n",
            "        -3.2550189e-01,  9.9263579e-01, -6.1484253e-01, -8.1711572e-01,\n",
            "        -2.9480696e-01, -6.3636196e-01, -1.6822517e-01,  8.2638514e-01,\n",
            "         1.1353626e+00,  1.6640463e-01,  7.3603541e-01, -2.3118262e-01],\n",
            "       [-2.4223689e-02,  9.9394415e-03, -3.3568147e-01,  1.2735659e-02,\n",
            "        -1.8980642e-01, -5.9371614e-03,  1.2968641e-02,  4.4121605e-01,\n",
            "         1.2882789e-02, -6.8002697e-03, -3.9469221e-01, -1.6335398e-01,\n",
            "         2.5585577e-01, -1.2368269e-01, -2.7644351e-01, -1.3032127e-02,\n",
            "        -3.2593762e-03,  6.6950339e-01,  9.9933028e-02,  7.6842293e-02],\n",
            "       [ 3.4318713e-03,  4.6488428e-03, -3.2811898e-01, -1.5370537e-02,\n",
            "         7.8937024e-02, -1.5077378e-02, -2.7313305e-02, -5.1652163e-01,\n",
            "        -1.6633573e-01, -2.1841029e-02, -9.9157527e-02, -7.8395581e-01,\n",
            "         2.2329704e-01,  2.4545342e-02, -5.8121610e-02, -8.8360667e-02,\n",
            "        -2.1606108e-02, -3.5217637e-01, -1.3444623e-01, -2.0891137e-01],\n",
            "       [-3.7851308e-02,  4.8147902e-01,  6.0623562e-01, -1.4450383e-01,\n",
            "         1.2927182e+00, -1.3309167e-01,  1.0760899e-01,  2.7477533e-01,\n",
            "         9.6830148e-01, -8.9811742e-02,  1.6236387e-01, -1.5093888e+00,\n",
            "        -6.8527389e-01,  1.6338751e+00, -6.9734678e-02,  4.2846161e-01,\n",
            "         3.1469505e-02,  1.8208628e+00,  8.4229678e-01, -1.6775341e-01]],\n",
            "      dtype=float32)>, <tf.Variable 'dense/bias:0' shape=(20,) dtype=float32, numpy=\n",
            "array([ 0.4015911 , -0.00447368,  0.        , -0.21385682, -0.02189895,\n",
            "       -0.25424594, -0.3049441 , -0.08542651, -0.01856359, -0.2525568 ,\n",
            "        0.        ,  0.66114116, -0.05117083, -0.01219233,  0.        ,\n",
            "       -0.68075335, -0.3146357 , -0.72337633, -0.6974354 ,  0.        ],\n",
            "      dtype=float32)>, <tf.Variable 'dense_1/kernel:0' shape=(20, 25) dtype=float32, numpy=\n",
            "array([[ 1.18797386e+00, -2.87597299e+00, -6.17139864e+00,\n",
            "         2.79745966e-01, -2.01194811e+00,  3.20219541e+00,\n",
            "        -4.20407414e-01, -3.59228897e+00,  1.68589294e+00,\n",
            "        -5.51922098e-02,  1.27775714e-01,  2.34151363e-01,\n",
            "         2.74781793e-01, -5.07098794e-01, -2.86591101e+00,\n",
            "         9.17308256e-02, -3.39802474e-01, -1.92235112e+00,\n",
            "         2.62766171e+00, -3.10012698e-01,  4.18226607e-03,\n",
            "         2.67304748e-01,  2.23150659e+00, -2.95496082e+00,\n",
            "        -2.17769623e+00],\n",
            "       [ 4.39212382e-01, -1.18571532e+00, -2.33495522e+00,\n",
            "        -2.35851121e+00, -1.14180720e+00,  9.17277277e-01,\n",
            "        -2.89875698e+00, -1.25774324e+00,  2.24436283e+00,\n",
            "        -2.23721313e+00,  4.39598942e+00, -1.58186567e+00,\n",
            "        -2.48679113e+00, -2.45085239e+00, -1.19591069e+00,\n",
            "         4.13064432e+00, -1.35063970e+00, -1.26779723e+00,\n",
            "        -1.91980861e-02, -2.20336628e+00,  4.44846249e+00,\n",
            "         4.06886768e+00,  1.03478909e+00, -9.21642959e-01,\n",
            "        -3.02985406e+00],\n",
            "       [-4.17104095e-01, -2.56565273e-01,  1.16979241e-01,\n",
            "        -3.41001242e-01,  2.91854680e-01,  6.20750012e-03,\n",
            "        -3.14088240e-02,  2.12960765e-01, -6.53994530e-02,\n",
            "        -2.46698499e-01,  1.91282362e-01,  4.65388983e-01,\n",
            "        -2.22433299e-01, -1.98698625e-01,  2.66260445e-01,\n",
            "         1.81171253e-01, -1.81933239e-01, -8.53323936e-03,\n",
            "        -2.26716146e-01, -3.59722286e-01,  1.07106321e-01,\n",
            "         4.08617526e-01, -6.38640895e-02,  8.03029537e-03,\n",
            "        -4.55954336e-02],\n",
            "       [-3.31473804e+00,  2.16833740e-01, -6.90198690e-02,\n",
            "        -7.65083253e-01,  2.02395424e-01, -3.23059344e+00,\n",
            "         5.81331611e-01,  2.57255495e-01, -1.95257425e+00,\n",
            "         1.02040100e+00, -1.80387294e+00, -5.55928424e-02,\n",
            "        -2.99410131e-02, -1.86289083e-02,  3.88513833e-01,\n",
            "        -3.73904891e-02,  6.20951271e-03,  2.25783542e-01,\n",
            "         1.71408826e-03,  1.83393583e-02, -1.33410895e+00,\n",
            "         1.81241129e-02, -3.60793781e+00,  4.36982006e-01,\n",
            "        -1.69888869e-01],\n",
            "       [-1.60681486e-01, -1.31957009e-01, -1.46161699e+00,\n",
            "        -3.11045289e+00, -5.02946019e-01,  2.30240136e-01,\n",
            "         2.83613831e-01, -9.61295441e-02,  7.95324087e-01,\n",
            "        -1.41568804e+00,  1.23133862e+00, -1.42662084e+00,\n",
            "        -1.84650078e-01,  7.23245144e-02,  2.15862021e-02,\n",
            "         8.85130405e-01, -7.31060624e-01, -8.82298410e-01,\n",
            "        -3.17557693e-01,  2.30092317e-01,  1.18438089e+00,\n",
            "         1.14174342e+00,  8.61971006e-02, -5.30435033e-02,\n",
            "        -2.67590809e+00],\n",
            "       [-3.53504992e+00,  4.74672943e-01, -3.87961455e-02,\n",
            "        -5.61640978e-01,  5.31032205e-01, -3.18435550e+00,\n",
            "        -9.07034948e-02,  4.21582729e-01, -1.96719182e+00,\n",
            "        -1.41410278e-02, -1.47197521e+00,  9.43615019e-01,\n",
            "        -1.02200210e-01, -1.79640204e-02,  3.55818212e-01,\n",
            "        -2.23372392e-02, -4.05110139e-03,  5.41334599e-02,\n",
            "        -4.98001650e-02,  3.34820300e-02, -1.24222481e+00,\n",
            "         2.02981420e-02, -2.92465067e+00,  1.64382517e-01,\n",
            "        -8.57979245e-03],\n",
            "       [-4.20597124e+00,  5.65884292e-01, -1.73727069e-02,\n",
            "        -5.46285677e+00,  2.30022654e-01, -3.82838655e+00,\n",
            "        -6.22417331e-02,  5.46257317e-01, -2.19329500e+00,\n",
            "         1.48028526e-02, -2.04239154e+00, -1.28071839e-02,\n",
            "        -6.56346008e-02,  7.77286172e-01,  3.28658223e-01,\n",
            "         1.32157858e-02,  9.87668216e-01,  4.29412544e-01,\n",
            "        -2.38056085e-03, -1.80538800e-02, -1.48078871e+00,\n",
            "        -5.64193632e-03, -3.89703012e+00,  7.34591186e-01,\n",
            "        -2.81734824e-01],\n",
            "       [-3.96483958e-01,  8.80362749e-01,  3.39590478e+00,\n",
            "        -1.48165488e+00,  1.49046767e+00, -2.53214419e-01,\n",
            "        -5.73068500e-01,  1.45299673e+00,  1.88100904e-01,\n",
            "        -4.27297860e-01,  6.64187014e-01, -3.76002669e-01,\n",
            "        -8.43906999e-01, -2.57789105e-01,  7.73255765e-01,\n",
            "         2.28672838e+00, -3.75695854e-01,  1.94810212e+00,\n",
            "        -5.92110932e-01,  1.05252586e-01,  9.81273174e-01,\n",
            "         2.63612795e+00, -6.28433585e-01,  2.21432972e+00,\n",
            "        -1.63036859e+00],\n",
            "       [-3.66809726e-01, -4.37472314e-02, -2.14173123e-01,\n",
            "        -2.03254986e+00, -2.23313663e-02, -1.23064056e-01,\n",
            "        -1.66291046e+00,  1.35391608e-01,  3.41515213e-01,\n",
            "        -7.04943955e-01,  9.08158720e-01, -5.88754356e-01,\n",
            "        -1.98209512e+00, -8.77699852e-01, -3.53751853e-02,\n",
            "         5.52475929e-01, -6.18922114e-01, -1.38860643e-01,\n",
            "         2.50646800e-01, -1.00195694e+00,  5.26072741e-01,\n",
            "         5.31982899e-01,  1.47453994e-01,  5.70510561e-03,\n",
            "        -2.94822454e+00],\n",
            "       [-3.16278195e+00,  3.63158137e-01, -7.57431462e-02,\n",
            "        -6.63409352e-01,  2.08675504e-01, -2.18237686e+00,\n",
            "         4.03052159e-02,  2.11031735e-01, -1.08568120e+00,\n",
            "         5.26637537e-03, -1.24360430e+00, -5.04205078e-02,\n",
            "         6.82417631e-01, -4.71283756e-02,  2.33043969e-01,\n",
            "         4.24906872e-02, -1.81124471e-02,  3.77760222e-03,\n",
            "        -6.73418120e-02,  1.50276432e-02, -9.42162991e-01,\n",
            "        -1.02581404e-01, -2.84145451e+00,  5.10029085e-02,\n",
            "         3.40192695e-04],\n",
            "       [ 2.62658179e-01,  1.41392142e-01, -1.94501266e-01,\n",
            "        -3.81340571e-02, -1.00173026e-01,  1.96163625e-01,\n",
            "         1.29138097e-01,  2.99501307e-02, -1.70260459e-01,\n",
            "        -1.94704160e-02, -4.24346365e-02,  1.81279644e-01,\n",
            "         4.47349370e-01,  2.44209617e-01,  5.89431487e-02,\n",
            "         2.93926708e-02, -2.64277101e-01, -3.08796316e-02,\n",
            "         3.50203305e-01,  7.48913512e-02, -2.33043239e-01,\n",
            "         1.98288545e-01,  3.52792263e-01, -4.57919203e-02,\n",
            "         7.14992508e-02],\n",
            "       [ 5.49717128e-01,  6.34528041e-01,  5.91397226e-01,\n",
            "         5.76294899e-01, -1.03582823e+00,  6.15795851e-01,\n",
            "        -2.95905066e+00,  5.45917153e-01, -4.13279593e-01,\n",
            "        -1.93600738e+00, -2.84983009e-01, -2.18417907e+00,\n",
            "        -2.82606745e+00, -2.53702283e+00,  7.69667029e-01,\n",
            "        -1.56496689e-01, -1.38824260e+00, -1.32627177e+00,\n",
            "         8.30164015e-01, -1.38496304e+00, -3.50354254e-01,\n",
            "        -3.01813036e-01,  5.85001469e-01,  5.75274706e-01,\n",
            "        -3.21359491e+00],\n",
            "       [-1.23807475e-01, -3.73103470e-02, -7.07434490e-02,\n",
            "        -2.54398537e+00,  7.91931152e-02, -8.87218788e-02,\n",
            "        -6.71827793e-01,  1.43094258e-02, -6.92976126e-03,\n",
            "         3.62728573e-02,  2.01446041e-01, -8.98312479e-02,\n",
            "        -4.98593450e-01,  5.04847988e-02, -2.75943689e-02,\n",
            "         3.07825327e-01,  2.68787116e-01,  5.11204451e-02,\n",
            "        -2.00371638e-01,  9.96439755e-02,  1.36011958e-01,\n",
            "         3.65377277e-01, -1.57596350e-01, -1.24124885e-02,\n",
            "        -1.28857660e+00],\n",
            "       [-5.97193511e-03, -2.82800317e-01, -2.14075732e+00,\n",
            "        -3.86194420e+00, -7.99989760e-01,  2.19183773e-01,\n",
            "         1.34331495e-01, -2.26609558e-01,  5.83257675e-01,\n",
            "         1.00718364e-01,  1.43910837e+00, -8.03119123e-01,\n",
            "        -2.58454472e-01,  4.03257877e-01, -4.00815196e-02,\n",
            "         1.30775237e+00, -1.21508741e+00, -1.29169762e+00,\n",
            "        -4.61712703e-02,  7.25161806e-02,  1.51767218e+00,\n",
            "         1.73992252e+00,  4.92864937e-01, -1.35515479e-03,\n",
            "        -2.36322975e+00],\n",
            "       [ 1.41885653e-01,  2.56750733e-01,  2.43790343e-01,\n",
            "        -1.12570964e-01,  1.55612618e-01,  1.57167241e-01,\n",
            "        -3.26569192e-02,  2.56003588e-01,  2.07620040e-01,\n",
            "        -1.05060115e-01, -4.04294342e-01, -3.65878642e-01,\n",
            "        -3.54863293e-02,  2.10746720e-01,  2.02738434e-01,\n",
            "         1.35712296e-01, -3.93587835e-02, -1.95556313e-01,\n",
            "         1.73507541e-01,  3.01556706e-01, -1.76385924e-01,\n",
            "         2.44364977e-01,  5.51154353e-02, -2.77206630e-01,\n",
            "         3.37717652e-01],\n",
            "       [ 8.28411043e-01,  1.24837017e+00, -3.38165712e+00,\n",
            "         3.81870413e+00,  1.86498618e+00,  4.16873693e-01,\n",
            "         4.16058683e+00,  2.42281213e-01,  2.97244608e-01,\n",
            "         1.44389355e+00, -1.53919673e+00,  2.02293181e+00,\n",
            "         3.35310173e+00,  3.04836774e+00,  5.67694783e-01,\n",
            "        -2.08629084e+00,  1.01457095e+00, -9.34198573e-02,\n",
            "         1.77614284e+00,  1.18147874e+00, -1.81878877e+00,\n",
            "        -2.37124133e+00,  2.86731094e-01, -1.90514970e+00,\n",
            "         4.00555325e+00],\n",
            "       [-3.90805745e+00,  1.15105018e-01,  3.01490501e-02,\n",
            "        -3.21949577e+00,  7.66370744e-02, -3.49621248e+00,\n",
            "        -1.30720392e-01,  4.02955234e-01, -1.91663456e+00,\n",
            "         2.18385011e-02, -1.53474343e+00, -5.48867062e-02,\n",
            "        -1.03427321e-01, -8.02480131e-02,  6.00317240e-01,\n",
            "        -9.79394279e-03, -6.38945103e-02,  1.41956449e-01,\n",
            "         1.08678782e+00,  1.07564962e+00, -1.09972870e+00,\n",
            "         5.41156940e-02, -3.58504677e+00,  5.46932638e-01,\n",
            "        -1.62164614e-01],\n",
            "       [-3.26866508e+00,  2.95719695e+00, -1.56670260e+00,\n",
            "         5.00024438e-01,  7.98059702e-01, -2.30257916e+00,\n",
            "         2.01756001e+00,  2.26585698e+00, -1.08992851e+00,\n",
            "         4.68266994e-01, -1.31356215e+00,  7.12188721e-01,\n",
            "         1.70030391e+00,  1.21994865e+00,  2.58395314e+00,\n",
            "        -7.20793903e-01,  4.26700503e-01, -8.42710674e-01,\n",
            "         7.56326020e-01,  6.35789394e-01, -1.67603004e+00,\n",
            "        -8.29373240e-01, -2.03172684e+00,  8.08169484e-01,\n",
            "         1.78165781e+00],\n",
            "       [ 9.40895855e-01,  5.45848131e-01, -3.05596447e+00,\n",
            "         4.19028330e+00,  5.56188583e-01,  4.10801709e-01,\n",
            "         2.45228767e+00, -7.86048114e-01,  7.64133260e-02,\n",
            "         6.45146370e-01, -1.63668501e+00,  1.11722815e+00,\n",
            "         3.08394289e+00,  1.56604433e+00, -1.70103416e-01,\n",
            "        -2.40462136e+00,  1.04157281e+00, -2.12190413e+00,\n",
            "         1.49796832e+00,  9.82381344e-01, -1.75272536e+00,\n",
            "        -2.28997874e+00,  2.03865901e-01, -2.42374730e+00,\n",
            "         3.66883302e+00],\n",
            "       [ 3.99354696e-01,  1.46472469e-01,  6.08907863e-02,\n",
            "         3.64278585e-01,  1.30180940e-01, -1.53074250e-01,\n",
            "        -4.58347574e-02, -1.07719162e-02,  1.65144816e-01,\n",
            "         1.40273124e-01,  2.80909359e-01,  4.36556280e-01,\n",
            "         1.70541853e-02, -3.38557422e-01,  4.22089368e-01,\n",
            "         1.57506149e-02, -3.33664417e-02,  3.61670613e-01,\n",
            "         4.03964043e-01,  2.05313489e-02, -2.70029068e-01,\n",
            "        -8.81234109e-02,  9.04880613e-02,  2.36495718e-01,\n",
            "         1.01340306e-03]], dtype=float32)>, <tf.Variable 'dense_1/bias:0' shape=(25,) dtype=float32, numpy=\n",
            "array([ 0.21278425,  0.02485727,  0.13240507,  0.0885734 ,  0.02698254,\n",
            "        0.12168364, -0.35765147,  0.02479931, -0.02286869, -0.8120361 ,\n",
            "       -0.07405329, -0.743961  , -0.3483459 , -0.5728679 , -0.02093392,\n",
            "       -0.07405394, -0.8639088 ,  0.04731729, -0.8286478 , -0.9186759 ,\n",
            "       -0.01823124, -0.04481741,  0.03070896,  0.03484697,  0.04338821],\n",
            "      dtype=float32)>, <tf.Variable 'dense_2/kernel:0' shape=(25, 20) dtype=float32, numpy=\n",
            "array([[-1.37106895e+00, -1.95757866e-01,  2.81130433e-01,\n",
            "         1.23604035e+00, -2.11556435e+00, -1.18754894e-01,\n",
            "        -1.41087741e-01,  7.02633262e-01, -3.61723602e-01,\n",
            "        -3.17726463e-01,  1.29645419e+00, -4.90082800e-01,\n",
            "         7.42943048e-01, -1.75630593e+00,  4.48431492e-01,\n",
            "        -6.95673376e-02, -3.74399394e-01, -3.38245571e-01,\n",
            "        -1.47014225e+00,  9.76732746e-03],\n",
            "       [ 7.42407084e-01,  9.32163358e-01, -3.19642138e+00,\n",
            "        -1.17090322e-01,  2.17057560e-02,  1.14489579e+00,\n",
            "         7.63291419e-01,  4.32192832e-02,  9.08479840e-03,\n",
            "         2.85997093e-01,  7.13760778e-02, -4.09807153e-02,\n",
            "        -7.45644867e-02, -3.59951593e-02, -1.76647529e-01,\n",
            "         8.30624104e-01,  7.28754401e-02,  9.95742436e-03,\n",
            "         1.37599677e-01, -3.77349949e+00],\n",
            "       [-2.17950508e-01, -1.25010133e-01, -9.50772583e-01,\n",
            "         6.74474478e-01,  1.07491720e+00,  9.83283281e-01,\n",
            "        -2.37818345e-01,  1.14920938e+00,  2.04039171e-01,\n",
            "        -1.36667863e-01, -1.00239731e-01,  3.30819845e-01,\n",
            "         1.52853858e+00,  1.04031086e+00, -2.10773289e-01,\n",
            "         2.23066702e-01, -1.02506816e-01,  1.16050109e-01,\n",
            "         1.16676219e-01, -8.71037006e-01],\n",
            "       [ 7.19105005e-01,  1.52168834e+00,  1.96234703e+00,\n",
            "         1.44606006e+00, -7.36804545e-01,  3.59879911e-01,\n",
            "         1.66680205e+00,  6.42388821e-01, -1.39575410e+00,\n",
            "         2.43253446e+00,  1.68335259e+00, -1.47454429e+00,\n",
            "         5.72902799e-01, -9.83620405e-01,  1.79440260e+00,\n",
            "         1.06584990e+00,  2.28164768e+00, -1.25600183e+00,\n",
            "         2.51362491e+00,  1.24426961e+00],\n",
            "       [ 1.13190031e+00,  9.80264843e-01, -2.68627954e+00,\n",
            "         4.39595096e-02, -3.83564867e-02,  9.62141514e-01,\n",
            "         9.90410507e-01, -2.33299937e-03,  2.69160829e-02,\n",
            "         4.84338701e-02,  5.21025918e-02,  6.93460852e-02,\n",
            "        -1.70717105e-01, -7.33650923e-02, -5.31271435e-02,\n",
            "         1.17962730e+00,  1.08752243e-01,  1.55805200e-02,\n",
            "        -1.75558865e-01, -3.10717082e+00],\n",
            "       [-2.94360542e+00, -3.58704567e-01,  9.86781716e-01,\n",
            "         1.38475001e-01, -4.68882036e+00, -6.68421388e-01,\n",
            "        -5.52905262e-01,  8.26073810e-02, -1.76343486e-01,\n",
            "         6.70795798e-01,  1.97089696e+00, -3.24964136e-01,\n",
            "        -1.53394505e-01, -5.09696531e+00, -5.21226227e-01,\n",
            "        -1.71701282e-01,  4.90260720e-01, -1.91549882e-01,\n",
            "         1.70136288e-01,  1.02787578e+00],\n",
            "       [-3.52968550e+00, -3.83283043e+00,  2.42120832e-01,\n",
            "         5.93138576e-01, -1.97433770e+00, -7.63494492e-01,\n",
            "        -3.51530719e+00,  1.92518508e+00, -1.71901762e+00,\n",
            "         1.12325644e+00,  1.71963620e+00, -1.48616803e+00,\n",
            "         1.91332746e+00, -1.76424456e+00,  1.08613741e+00,\n",
            "        -3.88892817e+00,  1.40701342e+00, -1.76295209e+00,\n",
            "         1.25920510e+00,  6.34904504e-01],\n",
            "       [ 1.05767298e+00,  1.27101982e+00, -1.64045036e+00,\n",
            "        -4.84923236e-02,  6.91182688e-02,  1.41834092e+00,\n",
            "         1.29680133e+00, -1.02555841e-01, -4.03936533e-03,\n",
            "         3.37301083e-02, -2.38589626e-02,  7.28680473e-03,\n",
            "        -1.26076201e-02,  5.57620591e-03, -1.61057562e-01,\n",
            "         1.07217419e+00,  1.30557507e-01, -2.50251610e-02,\n",
            "         1.29631579e-01, -1.34510529e+00],\n",
            "       [-5.83204031e-01, -1.09199369e+00,  7.92334080e-01,\n",
            "         1.48326203e-01, -2.71656418e+00, -1.42415726e+00,\n",
            "        -1.14530420e+00, -5.40448070e-01,  3.09976906e-01,\n",
            "         2.63804615e-01,  1.83088756e+00,  7.17422545e-01,\n",
            "        -1.89158201e-01, -2.00095868e+00,  3.68595034e-01,\n",
            "        -1.38329887e+00,  4.14130628e-01,  2.81455249e-01,\n",
            "         1.35552645e-01,  1.02115977e+00],\n",
            "       [-4.66074467e+00, -4.24292040e+00,  3.98810923e-01,\n",
            "         2.96530485e+00, -2.70769739e+00, -8.06524694e-01,\n",
            "        -3.97322536e+00,  2.89852691e+00, -2.54905009e+00,\n",
            "         2.96225286e+00,  2.71431518e+00, -2.66583467e+00,\n",
            "         3.35374427e+00, -2.25717735e+00,  3.20019078e+00,\n",
            "        -3.40829206e+00,  2.80654430e+00, -2.38197351e+00,\n",
            "         3.48954916e+00,  1.35140747e-01],\n",
            "       [ 5.50786965e-02, -3.12323391e-01,  4.11943525e-01,\n",
            "        -2.15596747e+00, -4.75841731e-01, -6.85836434e-01,\n",
            "        -4.28490937e-01, -1.74863875e+00,  3.39030623e+00,\n",
            "        -2.74069697e-01,  5.19854188e-01,  3.30088091e+00,\n",
            "        -1.72345161e+00, -4.79159653e-01, -4.00175750e-01,\n",
            "        -5.42319655e-01, -4.32632416e-01,  3.37642598e+00,\n",
            "        -5.62909603e-01,  4.16660041e-01],\n",
            "       [-4.01227379e+00, -3.96446872e+00,  2.27928773e-01,\n",
            "         2.85389352e+00, -2.58234477e+00, -5.74067295e-01,\n",
            "        -3.68997741e+00,  2.69011164e+00, -2.34326267e+00,\n",
            "         2.61366272e+00,  1.72753131e+00, -2.39522982e+00,\n",
            "         2.58267045e+00, -2.40097046e+00,  2.23841429e+00,\n",
            "        -3.37589717e+00,  2.47177696e+00, -2.46080995e+00,\n",
            "         2.63702750e+00, -3.01059335e-01],\n",
            "       [-3.13098955e+00, -3.17710018e+00,  6.05774105e-01,\n",
            "         1.29586637e+00, -1.70934141e+00, -4.64469880e-01,\n",
            "        -2.78803015e+00,  1.32680964e+00, -1.05347812e+00,\n",
            "         1.55018461e+00,  1.18325901e+00, -1.03180671e+00,\n",
            "         1.66466069e+00, -1.04251862e+00,  1.29948962e+00,\n",
            "        -2.78618503e+00,  1.18298054e+00, -1.08574283e+00,\n",
            "         1.86405671e+00,  6.35698557e-01],\n",
            "       [-4.25055695e+00, -4.23292398e+00,  4.67422485e-01,\n",
            "         2.13445830e+00, -2.55955815e+00, -9.18377340e-01,\n",
            "        -4.65385103e+00,  2.26395226e+00, -2.34848452e+00,\n",
            "         2.04916835e+00,  1.82859206e+00, -1.73714018e+00,\n",
            "         1.68478847e+00, -2.12124014e+00,  1.15505612e+00,\n",
            "        -2.67286706e+00,  1.70242143e+00, -2.13080883e+00,\n",
            "         1.65183508e+00,  2.55584419e-01],\n",
            "       [ 1.33231926e+00,  1.17378962e+00, -2.79626751e+00,\n",
            "         6.24807253e-02, -2.99933855e-03,  9.40094888e-01,\n",
            "         1.28863955e+00, -3.97671983e-02,  1.72024891e-02,\n",
            "        -6.99940994e-02,  7.56654665e-02, -5.95220849e-02,\n",
            "        -2.06309520e-02, -8.48022997e-02,  2.19842479e-01,\n",
            "         1.04863489e+00,  2.06629232e-01, -4.28705104e-02,\n",
            "         1.32341562e-02, -3.39417195e+00],\n",
            "       [ 1.03744835e-01,  9.39796120e-02,  1.41048655e-01,\n",
            "        -5.33524454e-01, -3.36879998e-01,  4.01720516e-02,\n",
            "        -4.13954929e-02, -8.78586352e-01,  7.11868048e-01,\n",
            "        -1.58975124e-01,  1.28006518e-01,  1.00311565e+00,\n",
            "        -2.03721851e-01, -7.25417063e-02,  1.40034240e-02,\n",
            "        -6.00227565e-02,  7.35842958e-02,  1.03033769e+00,\n",
            "         1.69022977e-02,  2.58614987e-01],\n",
            "       [-3.29727268e+00, -3.90099216e+00,  1.24911293e-01,\n",
            "         2.77738857e+00, -2.81592298e+00, -8.97019923e-01,\n",
            "        -3.56814861e+00,  3.86614919e+00, -2.72708321e+00,\n",
            "         3.34277463e+00,  3.10080910e+00, -2.62477064e+00,\n",
            "         3.21223831e+00, -2.60217690e+00,  2.97934198e+00,\n",
            "        -3.72033858e+00,  3.25430250e+00, -2.36061049e+00,\n",
            "         3.34155655e+00, -2.48346299e-01],\n",
            "       [ 1.25011361e+00,  9.69940960e-01, -2.59489918e+00,\n",
            "        -5.28838951e-03,  8.69165286e-02,  7.36517847e-01,\n",
            "         1.24572182e+00,  9.01879296e-02,  1.29037932e-01,\n",
            "         3.99002880e-02,  4.29807007e-02,  1.31246731e-01,\n",
            "         3.63215543e-02, -1.00822840e-02, -5.57630211e-02,\n",
            "         9.24349844e-01,  5.65398447e-02,  3.28815132e-02,\n",
            "         5.49349748e-02, -3.20103931e+00],\n",
            "       [-2.25362492e+00, -1.64978397e+00,  2.80971098e+00,\n",
            "         1.99449635e+00, -2.13697314e+00, -9.38699767e-02,\n",
            "        -1.87677252e+00,  4.53923166e-01, -1.63592041e+00,\n",
            "         1.94780242e+00,  2.35228777e+00, -1.85535705e+00,\n",
            "         1.17909396e+00, -1.90854788e+00,  2.09852481e+00,\n",
            "        -1.85109091e+00,  2.20319843e+00, -1.53960586e+00,\n",
            "         1.58165991e+00,  2.48356414e+00],\n",
            "       [-3.39139128e+00, -3.61735487e+00,  2.00170621e-01,\n",
            "         3.70931578e+00, -2.87332153e+00, -7.18156517e-01,\n",
            "        -3.90857887e+00,  3.61433458e+00, -2.67337871e+00,\n",
            "         3.36759448e+00,  3.06046462e+00, -2.26608539e+00,\n",
            "         3.98008275e+00, -2.50300407e+00,  3.44350290e+00,\n",
            "        -2.72808146e+00,  3.10024476e+00, -2.58766460e+00,\n",
            "         3.20682645e+00, -1.31528557e-03],\n",
            "       [-1.03443019e-01, -2.65923023e-01,  3.16848695e-01,\n",
            "        -1.53331482e+00, -7.05794096e-01, -7.01019287e-01,\n",
            "        -4.79343027e-01, -1.71064401e+00,  2.18469262e+00,\n",
            "        -2.26060137e-01,  2.28982896e-01,  2.03866601e+00,\n",
            "        -1.62004411e+00, -5.54349840e-01, -1.85360715e-01,\n",
            "        -7.36982286e-01, -3.23236197e-01,  2.35978532e+00,\n",
            "        -5.09303033e-01,  3.40170056e-01],\n",
            "       [ 6.07365593e-02,  9.20755938e-02,  2.48744681e-01,\n",
            "        -6.44599617e-01, -2.28939787e-01,  6.16500503e-04,\n",
            "         5.16574085e-02, -1.52493083e+00,  8.24309170e-01,\n",
            "        -1.42468318e-01,  1.69995710e-01,  7.62157023e-01,\n",
            "        -6.57210350e-01, -4.14824113e-02, -1.81093305e-01,\n",
            "        -1.25935748e-01, -9.29722041e-02,  1.07380700e+00,\n",
            "        -4.12606113e-02,  2.44331077e-01],\n",
            "       [-2.88310385e+00, -1.05549467e+00,  2.24274707e+00,\n",
            "         4.64269876e-01, -5.04123211e+00, -4.73697364e-01,\n",
            "        -1.11064148e+00,  3.15436721e-01, -1.11591339e-01,\n",
            "        -3.24717045e-01,  2.54524970e+00, -2.45407566e-01,\n",
            "         2.14928672e-01, -5.77885866e+00,  4.43278044e-01,\n",
            "        -4.64324594e-01, -2.23123953e-01, -1.29376426e-01,\n",
            "         7.20118225e-01,  1.98199010e+00],\n",
            "       [ 1.04501963e+00,  1.25292158e+00, -2.53648400e+00,\n",
            "         9.31477323e-02,  7.04570487e-02,  9.13295150e-01,\n",
            "         1.24075818e+00, -8.12623948e-02,  1.12265252e-01,\n",
            "         9.71206129e-02,  5.35775051e-02,  1.45740835e-02,\n",
            "        -9.94490087e-03,  6.57281652e-03, -3.69089171e-02,\n",
            "         1.32676053e+00,  2.92462651e-02,  6.59406409e-02,\n",
            "        -8.18695053e-02, -2.58535123e+00],\n",
            "       [ 8.50898549e-02,  1.14296722e+00,  8.81053627e-01,\n",
            "         1.80097449e+00, -3.79653126e-01,  2.33158207e+00,\n",
            "         4.96619880e-01,  2.66920495e+00, -1.43702614e+00,\n",
            "         2.20093489e+00,  1.67548132e+00, -1.24845970e+00,\n",
            "         2.24425983e+00, -5.89089572e-01,  1.88781607e+00,\n",
            "         1.29604614e+00,  1.46546221e+00, -1.40917408e+00,\n",
            "         1.79773116e+00,  7.39986956e-01]], dtype=float32)>, <tf.Variable 'dense_2/bias:0' shape=(20,) dtype=float32, numpy=\n",
            "array([-0.00737699,  0.02757048, -0.00705412, -0.12698066,  0.15174192,\n",
            "        0.10275084,  0.04888716, -0.23903844,  0.19757234, -0.2808261 ,\n",
            "       -0.30747756,  0.1466565 , -0.03508269,  0.30211303, -0.4564334 ,\n",
            "       -0.02007052, -0.386415  ,  0.2437417 , -0.24377114,  0.01950282],\n",
            "      dtype=float32)>, <tf.Variable 'dense_3/kernel:0' shape=(20, 3) dtype=float32, numpy=\n",
            "array([[ 0.25123736,  0.22631282, -3.6605737 ],\n",
            "       [ 0.24387698,  0.26910096, -2.8904524 ],\n",
            "       [ 0.64999896, -1.9616166 ,  0.5547164 ],\n",
            "       [ 1.0764358 , -0.84158224, -0.5900102 ],\n",
            "       [-2.9159482 ,  2.3456259 , -1.8084701 ],\n",
            "       [-0.08717763, -0.09687141, -2.1798325 ],\n",
            "       [ 0.4445945 ,  0.47585112, -2.6848705 ],\n",
            "       [ 0.45977628, -1.0291592 , -2.163774  ],\n",
            "       [-4.1604505 ,  0.44729415,  0.86152035],\n",
            "       [ 0.796446  , -0.43716472, -1.5019774 ],\n",
            "       [ 0.22014801, -1.4964887 ,  0.17273706],\n",
            "       [-4.3966236 ,  0.37638408,  0.82037205],\n",
            "       [ 0.37851334, -0.9891751 , -2.9208982 ],\n",
            "       [-4.0458894 ,  1.9196274 , -0.3913224 ],\n",
            "       [ 0.9178704 , -0.49897528, -1.5463753 ],\n",
            "       [ 0.43982682,  0.43365166, -2.6349797 ],\n",
            "       [ 0.8992024 , -0.8323096 , -1.7191231 ],\n",
            "       [-3.8832114 ,  0.3988492 ,  0.7904539 ],\n",
            "       [ 0.64550483, -0.7561582 , -2.3891754 ],\n",
            "       [ 1.0002311 , -1.8148804 ,  1.1932877 ]], dtype=float32)>, <tf.Variable 'dense_3/bias:0' shape=(3,) dtype=float32, numpy=array([-3.0285997 ,  0.8732097 ,  0.85611695], dtype=float32)>]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uSgDNaZprCVd",
        "outputId": "670fd319-98b5-4292-dd88-af5bd3c16cf6"
      },
      "source": [
        "x=np.array([0.0540525,\t0.407113,\t0.2552,\t931\t,0.0754249\t])\n",
        "x=x.reshape(1,5,1)\n",
        "print(model.predict(x))\n",
        "print(np.argmax(model.predict(x)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[0. 0. 1.]]\n",
            "2\n"
          ]
        }
      ]
    }
  ]
}