rhiskey/classification_nn_seminarus.ipynb

## classification_nn_seminarus.ipynb
{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Classification_NN_Seminarus.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPFu4oGKana8Wrxwfjd8UX7",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/rhiskey/02c1b2818dd819ad514961ebfd7e4c3b/classification_nn_seminarus.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#‎Классификация с помощью нейронных сетей с использованием Python‎\n"
      ],
      "metadata": {
        "id": "Mhp5Fh66jaA1"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "IxxN8ZI4jHpq"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "fashion = keras.datasets.fashion_mnist\n",
        "(xtrain, ytrain), (xtest, ytest) = fashion.load_data()"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "imgIndex = 8\n",
        "image = xtrain[imgIndex]\n",
        "print(\"label: \", ytrain[imgIndex])\n",
        "plt.imshow(image)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "HD8RWEklj-8l",
        "outputId": "b578a433-441c-48b1-eee6-cd6dc2b93cb4"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "label:  5\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f57e72ff110>"
            ]
          },
          "metadata": {},
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "xtrain.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1hSalJp-kEaz",
        "outputId": "a486a716-f342-47d8-f354-a95d847a0d44"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(60000, 28, 28)"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "xtest.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lKsnnfo9kH6q",
        "outputId": "c467105e-06cf-43e3-9af6-83e65e8fda67"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(10000, 28, 28)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# ‎Построение архитектуры нейронной сети‎"
      ],
      "metadata": {
        "id": "3u0x9E4VkKHr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = keras.models.Sequential([\n",
        "    keras.layers.Flatten(input_shape=[28, 28]),\n",
        "    keras.layers.Dense(300, activation=\"relu\"),\n",
        "    keras.layers.Dense(100, activation=\"relu\"),\n",
        "    keras.layers.Dense(10, activation=\"softmax\"),\n",
        "])\n",
        "\n",
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rZNah7DSkPNb",
        "outputId": "67e47925-2bab-4e46-8bbc-be7637a3d5a2"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " flatten_1 (Flatten)         (None, 784)               0         \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 300)               235500    \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 100)               30100     \n",
            "                                                                 \n",
            " dense_5 (Dense)             (None, 10)                1010      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 266,610\n",
            "Trainable params: 266,610\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "xvalid, xtrain = xtrain[:5000]/255.0, xtrain[5000:]/255.0\n",
        "yvalid, ytrain = ytrain[:5000], ytrain[5000:]"
      ],
      "metadata": {
        "id": "sZ_NFRBRkVhr"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#‎Обучение модели классификации с помощью нейронных сетей‎"
      ],
      "metadata": {
        "id": "qbmWVgB9kcok"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(loss=\"sparse_categorical_crossentropy\",\n",
        "              optimizer=\"sgd\",\n",
        "              metrics=[\"accuracy\"])\n",
        "history = model.fit(xtrain, ytrain, epochs=30, validation_data=(xvalid, yvalid))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D3NcCFcWkgQc",
        "outputId": "3d5c6767-e702-447d-ec4a-2a19322170b5"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/30\n",
            "1719/1719 [==============================] - 6s 3ms/step - loss: 0.7261 - accuracy: 0.7662 - val_loss: 0.5003 - val_accuracy: 0.8334\n",
            "Epoch 2/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.4881 - accuracy: 0.8313 - val_loss: 0.4471 - val_accuracy: 0.8508\n",
            "Epoch 3/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.4423 - accuracy: 0.8456 - val_loss: 0.4226 - val_accuracy: 0.8572\n",
            "Epoch 4/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.4142 - accuracy: 0.8544 - val_loss: 0.4043 - val_accuracy: 0.8608\n",
            "Epoch 5/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.3961 - accuracy: 0.8607 - val_loss: 0.3804 - val_accuracy: 0.8658\n",
            "Epoch 6/30\n",
            "1719/1719 [==============================] - 6s 3ms/step - loss: 0.3792 - accuracy: 0.8661 - val_loss: 0.3928 - val_accuracy: 0.8676\n",
            "Epoch 7/30\n",
            "1719/1719 [==============================] - 6s 3ms/step - loss: 0.3644 - accuracy: 0.8709 - val_loss: 0.3870 - val_accuracy: 0.8664\n",
            "Epoch 8/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.3536 - accuracy: 0.8742 - val_loss: 0.3474 - val_accuracy: 0.8786\n",
            "Epoch 9/30\n",
            "1719/1719 [==============================] - 6s 3ms/step - loss: 0.3440 - accuracy: 0.8780 - val_loss: 0.3449 - val_accuracy: 0.8784\n",
            "Epoch 10/30\n",
            "1719/1719 [==============================] - 6s 3ms/step - loss: 0.3341 - accuracy: 0.8802 - val_loss: 0.3493 - val_accuracy: 0.8792\n",
            "Epoch 11/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.3263 - accuracy: 0.8839 - val_loss: 0.3460 - val_accuracy: 0.8762\n",
            "Epoch 12/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.3177 - accuracy: 0.8862 - val_loss: 0.3436 - val_accuracy: 0.8764\n",
            "Epoch 13/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.3095 - accuracy: 0.8892 - val_loss: 0.3372 - val_accuracy: 0.8812\n",
            "Epoch 14/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.3032 - accuracy: 0.8913 - val_loss: 0.3389 - val_accuracy: 0.8810\n",
            "Epoch 15/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2958 - accuracy: 0.8927 - val_loss: 0.3370 - val_accuracy: 0.8786\n",
            "Epoch 16/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2912 - accuracy: 0.8959 - val_loss: 0.3258 - val_accuracy: 0.8856\n",
            "Epoch 17/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2856 - accuracy: 0.8967 - val_loss: 0.3243 - val_accuracy: 0.8860\n",
            "Epoch 18/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2802 - accuracy: 0.8994 - val_loss: 0.3257 - val_accuracy: 0.8820\n",
            "Epoch 19/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2757 - accuracy: 0.9000 - val_loss: 0.3181 - val_accuracy: 0.8838\n",
            "Epoch 20/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2702 - accuracy: 0.9024 - val_loss: 0.3218 - val_accuracy: 0.8886\n",
            "Epoch 21/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2651 - accuracy: 0.9036 - val_loss: 0.3174 - val_accuracy: 0.8862\n",
            "Epoch 22/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2604 - accuracy: 0.9067 - val_loss: 0.3142 - val_accuracy: 0.8852\n",
            "Epoch 23/30\n",
            "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2562 - accuracy: 0.9075 - val_loss: 0.3103 - val_accuracy: 0.8880\n",
            "Epoch 24/30\n",
            "1719/1719 [==============================] - 6s 4ms/step - loss: 0.2516 - accuracy: 0.9088 - val_loss: 0.3274 - val_accuracy: 0.8842\n",
            "Epoch 25/30\n",
            "1719/1719 [==============================] - 7s 4ms/step - loss: 0.2475 - accuracy: 0.9108 - val_loss: 0.2981 - val_accuracy: 0.8936\n",
            "Epoch 26/30\n",
            "1719/1719 [==============================] - 6s 4ms/step - loss: 0.2438 - accuracy: 0.9119 - val_loss: 0.3101 - val_accuracy: 0.8882\n",
            "Epoch 27/30\n",
            "1719/1719 [==============================] - 6s 4ms/step - loss: 0.2396 - accuracy: 0.9140 - val_loss: 0.3058 - val_accuracy: 0.8862\n",
            "Epoch 28/30\n",
            "1719/1719 [==============================] - 6s 4ms/step - loss: 0.2357 - accuracy: 0.9153 - val_loss: 0.3105 - val_accuracy: 0.8880\n",
            "Epoch 29/30\n",
            "1719/1719 [==============================] - 6s 4ms/step - loss: 0.2322 - accuracy: 0.9159 - val_loss: 0.3112 - val_accuracy: 0.8874\n",
            "Epoch 30/30\n",
            "1719/1719 [==============================] - 6s 4ms/step - loss: 0.2276 - accuracy: 0.9185 - val_loss: 0.2958 - val_accuracy: 0.8934\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new = xtest[:5]\n",
        "predictions = model.predict(new)\n",
        "predictions"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6hgiga9Mkkir",
        "outputId": "4c191145-0504-4e4d-c100-b52c8e581fda"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 1.],\n",
              "       [0., 0., 1., 0., 0., 0., 0., 0., 0., 0.],\n",
              "       [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "       [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "       [0., 0., 0., 0., 0., 0., 1., 0., 0., 0.]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "classes = np.argmax(predictions, axis=1)\n",
        "classes"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BL8H1xeBkohU",
        "outputId": "824f6b0b-04aa-41b6-d479-212d7d63647b"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9, 2, 1, 1, 6])"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        ""
      ],
      "metadata": {
        "id": "Rll3I921ksGU"
      }
    }
  ]
}
	{
	"nbformat": 4,
	"nbformat_minor": 0,
	"metadata": {
	"colab": {
	"name": "Classification_NN_Seminarus.ipynb",
	"provenance": [],
	"collapsed_sections": [],
	"authorship_tag": "ABX9TyPFu4oGKana8Wrxwfjd8UX7",
	"include_colab_link": true
	},
	"kernelspec": {
	"name": "python3",
	"display_name": "Python 3"
	},
	"language_info": {
	"name": "python"
	}
	},
	"cells": [
	{
	"cell_type": "markdown",
	"metadata": {
	"id": "view-in-github",
	"colab_type": "text"
	},
	"source": [
	"<a href=\"https://colab.research.google.com/gist/rhiskey/02c1b2818dd819ad514961ebfd7e4c3b/classification_nn_seminarus.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
	]
	},
	{
	"cell_type": "markdown",
	"source": [
	"#‎Классификация с помощью нейронных сетей с использованием Python‎\n"
	],
	"metadata": {
	"id": "Mhp5Fh66jaA1"
	}
	},
	{
	"cell_type": "code",
	"execution_count": 10,
	"metadata": {
	"id": "IxxN8ZI4jHpq"
	},
	"outputs": [],
	"source": [
	"import tensorflow as tf\n",
	"from tensorflow import keras\n",
	"import numpy as np\n",
	"import matplotlib.pyplot as plt\n",
	"\n",
	"fashion = keras.datasets.fashion_mnist\n",
	"(xtrain, ytrain), (xtest, ytest) = fashion.load_data()"
	]
	},
	{
	"cell_type": "code",
	"source": [
	"imgIndex = 8\n",
	"image = xtrain[imgIndex]\n",
	"print(\"label: \", ytrain[imgIndex])\n",
	"plt.imshow(image)"
	],
	"metadata": {
	"colab": {
	"base_uri": "https://localhost:8080/",
	"height": 300
	},
	"id": "HD8RWEklj-8l",
	"outputId": "b578a433-441c-48b1-eee6-cd6dc2b93cb4"
	},
	"execution_count": 11,
	"outputs": [
	{
	"output_type": "stream",
	"name": "stdout",
	"text": [
	"label: 5\n"
	]
	},
	{
	"output_type": "execute_result",
	"data": {
	"text/plain": [
	"<matplotlib.image.AxesImage at 0x7f57e72ff110>"
	]
	},
	"metadata": {},
	"execution_count": 11
	},
	{
	"output_type": "display_data",
	"data": {
	"text/plain": [
	"<Figure size 432x288 with 1 Axes>"
	],
	"image/png": "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\n"
	},
	"metadata": {
	"needs_background": "light"
	}
	}
	]
	},
	{
	"cell_type": "code",
	"source": [
	"xtrain.shape"
	],
	"metadata": {
	"colab": {
	"base_uri": "https://localhost:8080/"
	},
	"id": "1hSalJp-kEaz",
	"outputId": "a486a716-f342-47d8-f354-a95d847a0d44"
	},
	"execution_count": 12,
	"outputs": [
	{
	"output_type": "execute_result",
	"data": {
	"text/plain": [
	"(60000, 28, 28)"
	]
	},
	"metadata": {},
	"execution_count": 12
	}
	]
	},
	{
	"cell_type": "code",
	"source": [
	"xtest.shape"
	],
	"metadata": {
	"colab": {
	"base_uri": "https://localhost:8080/"
	},
	"id": "lKsnnfo9kH6q",
	"outputId": "c467105e-06cf-43e3-9af6-83e65e8fda67"
	},
	"execution_count": 13,
	"outputs": [
	{
	"output_type": "execute_result",
	"data": {
	"text/plain": [
	"(10000, 28, 28)"
	]
	},
	"metadata": {},
	"execution_count": 13
	}
	]
	},
	{
	"cell_type": "markdown",
	"source": [
	"# ‎Построение архитектуры нейронной сети‎"
	],
	"metadata": {
	"id": "3u0x9E4VkKHr"
	}
	},
	{
	"cell_type": "code",
	"source": [
	"model = keras.models.Sequential([\n",
	" keras.layers.Flatten(input_shape=[28, 28]),\n",
	" keras.layers.Dense(300, activation=\"relu\"),\n",
	" keras.layers.Dense(100, activation=\"relu\"),\n",
	" keras.layers.Dense(10, activation=\"softmax\"),\n",
	"])\n",
	"\n",
	"model.summary()"
	],
	"metadata": {
	"colab": {
	"base_uri": "https://localhost:8080/"
	},
	"id": "rZNah7DSkPNb",
	"outputId": "67e47925-2bab-4e46-8bbc-be7637a3d5a2"
	},
	"execution_count": 14,
	"outputs": [
	{
	"output_type": "stream",
	"name": "stdout",
	"text": [
	"Model: \"sequential_1\"\n",
	"_________________________________________________________________\n",
	" Layer (type) Output Shape Param # \n",
	"=================================================================\n",
	" flatten_1 (Flatten) (None, 784) 0 \n",
	" \n",
	" dense_3 (Dense) (None, 300) 235500 \n",
	" \n",
	" dense_4 (Dense) (None, 100) 30100 \n",
	" \n",
	" dense_5 (Dense) (None, 10) 1010 \n",
	" \n",
	"=================================================================\n",
	"Total params: 266,610\n",
	"Trainable params: 266,610\n",
	"Non-trainable params: 0\n",
	"_________________________________________________________________\n"
	]
	}
	]
	},
	{
	"cell_type": "code",
	"source": [
	"xvalid, xtrain = xtrain[:5000]/255.0, xtrain[5000:]/255.0\n",
	"yvalid, ytrain = ytrain[:5000], ytrain[5000:]"
	],
	"metadata": {
	"id": "sZ_NFRBRkVhr"
	},
	"execution_count": 15,
	"outputs": []
	},
	{
	"cell_type": "markdown",
	"source": [
	"#‎Обучение модели классификации с помощью нейронных сетей‎"
	],
	"metadata": {
	"id": "qbmWVgB9kcok"
	}
	},
	{
	"cell_type": "code",
	"source": [
	"model.compile(loss=\"sparse_categorical_crossentropy\",\n",
	" optimizer=\"sgd\",\n",
	" metrics=[\"accuracy\"])\n",
	"history = model.fit(xtrain, ytrain, epochs=30, validation_data=(xvalid, yvalid))"
	],
	"metadata": {
	"colab": {
	"base_uri": "https://localhost:8080/"
	},
	"id": "D3NcCFcWkgQc",
	"outputId": "3d5c6767-e702-447d-ec4a-2a19322170b5"
	},
	"execution_count": 16,
	"outputs": [
	{
	"output_type": "stream",
	"name": "stdout",
	"text": [
	"Epoch 1/30\n",
	"1719/1719 [==============================] - 6s 3ms/step - loss: 0.7261 - accuracy: 0.7662 - val_loss: 0.5003 - val_accuracy: 0.8334\n",
	"Epoch 2/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.4881 - accuracy: 0.8313 - val_loss: 0.4471 - val_accuracy: 0.8508\n",
	"Epoch 3/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.4423 - accuracy: 0.8456 - val_loss: 0.4226 - val_accuracy: 0.8572\n",
	"Epoch 4/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.4142 - accuracy: 0.8544 - val_loss: 0.4043 - val_accuracy: 0.8608\n",
	"Epoch 5/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.3961 - accuracy: 0.8607 - val_loss: 0.3804 - val_accuracy: 0.8658\n",
	"Epoch 6/30\n",
	"1719/1719 [==============================] - 6s 3ms/step - loss: 0.3792 - accuracy: 0.8661 - val_loss: 0.3928 - val_accuracy: 0.8676\n",
	"Epoch 7/30\n",
	"1719/1719 [==============================] - 6s 3ms/step - loss: 0.3644 - accuracy: 0.8709 - val_loss: 0.3870 - val_accuracy: 0.8664\n",
	"Epoch 8/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.3536 - accuracy: 0.8742 - val_loss: 0.3474 - val_accuracy: 0.8786\n",
	"Epoch 9/30\n",
	"1719/1719 [==============================] - 6s 3ms/step - loss: 0.3440 - accuracy: 0.8780 - val_loss: 0.3449 - val_accuracy: 0.8784\n",
	"Epoch 10/30\n",
	"1719/1719 [==============================] - 6s 3ms/step - loss: 0.3341 - accuracy: 0.8802 - val_loss: 0.3493 - val_accuracy: 0.8792\n",
	"Epoch 11/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.3263 - accuracy: 0.8839 - val_loss: 0.3460 - val_accuracy: 0.8762\n",
	"Epoch 12/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.3177 - accuracy: 0.8862 - val_loss: 0.3436 - val_accuracy: 0.8764\n",
	"Epoch 13/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.3095 - accuracy: 0.8892 - val_loss: 0.3372 - val_accuracy: 0.8812\n",
	"Epoch 14/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.3032 - accuracy: 0.8913 - val_loss: 0.3389 - val_accuracy: 0.8810\n",
	"Epoch 15/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2958 - accuracy: 0.8927 - val_loss: 0.3370 - val_accuracy: 0.8786\n",
	"Epoch 16/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2912 - accuracy: 0.8959 - val_loss: 0.3258 - val_accuracy: 0.8856\n",
	"Epoch 17/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2856 - accuracy: 0.8967 - val_loss: 0.3243 - val_accuracy: 0.8860\n",
	"Epoch 18/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2802 - accuracy: 0.8994 - val_loss: 0.3257 - val_accuracy: 0.8820\n",
	"Epoch 19/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2757 - accuracy: 0.9000 - val_loss: 0.3181 - val_accuracy: 0.8838\n",
	"Epoch 20/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2702 - accuracy: 0.9024 - val_loss: 0.3218 - val_accuracy: 0.8886\n",
	"Epoch 21/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2651 - accuracy: 0.9036 - val_loss: 0.3174 - val_accuracy: 0.8862\n",
	"Epoch 22/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2604 - accuracy: 0.9067 - val_loss: 0.3142 - val_accuracy: 0.8852\n",
	"Epoch 23/30\n",
	"1719/1719 [==============================] - 5s 3ms/step - loss: 0.2562 - accuracy: 0.9075 - val_loss: 0.3103 - val_accuracy: 0.8880\n",
	"Epoch 24/30\n",
	"1719/1719 [==============================] - 6s 4ms/step - loss: 0.2516 - accuracy: 0.9088 - val_loss: 0.3274 - val_accuracy: 0.8842\n",
	"Epoch 25/30\n",
	"1719/1719 [==============================] - 7s 4ms/step - loss: 0.2475 - accuracy: 0.9108 - val_loss: 0.2981 - val_accuracy: 0.8936\n",
	"Epoch 26/30\n",
	"1719/1719 [==============================] - 6s 4ms/step - loss: 0.2438 - accuracy: 0.9119 - val_loss: 0.3101 - val_accuracy: 0.8882\n",
	"Epoch 27/30\n",
	"1719/1719 [==============================] - 6s 4ms/step - loss: 0.2396 - accuracy: 0.9140 - val_loss: 0.3058 - val_accuracy: 0.8862\n",
	"Epoch 28/30\n",
	"1719/1719 [==============================] - 6s 4ms/step - loss: 0.2357 - accuracy: 0.9153 - val_loss: 0.3105 - val_accuracy: 0.8880\n",
	"Epoch 29/30\n",
	"1719/1719 [==============================] - 6s 4ms/step - loss: 0.2322 - accuracy: 0.9159 - val_loss: 0.3112 - val_accuracy: 0.8874\n",
	"Epoch 30/30\n",
	"1719/1719 [==============================] - 6s 4ms/step - loss: 0.2276 - accuracy: 0.9185 - val_loss: 0.2958 - val_accuracy: 0.8934\n"
	]
	}
	]
	},
	{
	"cell_type": "code",
	"source": [
	"new = xtest[:5]\n",
	"predictions = model.predict(new)\n",
	"predictions"
	],
	"metadata": {
	"colab": {
	"base_uri": "https://localhost:8080/"
	},
	"id": "6hgiga9Mkkir",
	"outputId": "4c191145-0504-4e4d-c100-b52c8e581fda"
	},
	"execution_count": 17,
	"outputs": [
	{
	"output_type": "execute_result",
	"data": {
	"text/plain": [
	"array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 1.],\n",
	" [0., 0., 1., 0., 0., 0., 0., 0., 0., 0.],\n",
	" [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
	" [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
	" [0., 0., 0., 0., 0., 0., 1., 0., 0., 0.]], dtype=float32)"
	]
	},
	"metadata": {},
	"execution_count": 17
	}
	]
	},
	{
	"cell_type": "code",
	"source": [
	"classes = np.argmax(predictions, axis=1)\n",
	"classes"
	],
	"metadata": {
	"colab": {
	"base_uri": "https://localhost:8080/"
	},
	"id": "BL8H1xeBkohU",
	"outputId": "824f6b0b-04aa-41b6-d479-212d7d63647b"
	},
	"execution_count": 18,
	"outputs": [
	{
	"output_type": "execute_result",
	"data": {
	"text/plain": [
	"array([9, 2, 1, 1, 6])"
	]
	},
	"metadata": {},
	"execution_count": 18
	}
	]
	},
	{
	"cell_type": "markdown",
	"source": [
	""
	],
	"metadata": {
	"id": "Rll3I921ksGU"
	}
	}
	]
	}