3. Machine Learning & Deep Learning/191001_(2) _ MNIST.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"191001_(2) ; TF_dataset_mnist.ipynb","provenance":[],"collapsed_sections":[]},"kernelspec":{"name":"python3","display_name":"Python 3"}},"cells":[{"cell_type":"markdown","metadata":{"id":"QIef5lFZyt-A","colab_type":"text"},"source":["## 1"]},{"cell_type":"code","metadata":{"id":"jYsyrMUpfoQn","colab_type":"code","colab":{}},"source":["import tensorflow as tf"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"tMgeYDOEgbCB","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":55},"outputId":"79556aab-f354-4389-82cd-32a621c74351","executionInfo":{"status":"ok","timestamp":1569902963121,"user_tz":-540,"elapsed":1009,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["mnist = tf.keras.datasets.mnist\n","(X_train, y_train), (X_test, y_test) = mnist.load_data()"],"execution_count":2,"outputs":[{"output_type":"stream","text":["Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n","11493376/11490434 [==============================] - 0s 0us/step\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"y91qlVAOgiU6","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":36},"outputId":"d91035c1-bfe3-4e10-c100-4a09696e59f7","executionInfo":{"status":"ok","timestamp":1569903045056,"user_tz":-540,"elapsed":421,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["print(type(X_train), type(y_train), type(X_test), type(y_test))"],"execution_count":3,"outputs":[{"output_type":"stream","text":["<class 'numpy.ndarray'> <class 'numpy.ndarray'> <class 'numpy.ndarray'> <class 'numpy.ndarray'>\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"v3n6CHAkg2d8","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":55},"outputId":"f51d9a7f-8761-406e-a0fd-63a5f8afc288","executionInfo":{"status":"ok","timestamp":1569903082864,"user_tz":-540,"elapsed":415,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["print(X_train.shape, y_train.shape)\n","print(X_test.shape, y_test.shape)"],"execution_count":4,"outputs":[{"output_type":"stream","text":["(60000, 28, 28) (60000,)\n","(10000, 28, 28) (10000,)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"lZsg4Nqzg_tr","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":92},"outputId":"5ef70aef-8fdd-46ee-fecb-fedb49cc60ed","executionInfo":{"status":"ok","timestamp":1569903456341,"user_tz":-540,"elapsed":526,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["print(X_train.shape[0])\n","print(X_test.shape[0])\n","print(y_train.shape[0])\n","print(y_test.shape[0])"],"execution_count":5,"outputs":[{"output_type":"stream","text":["60000\n","10000\n","60000\n","10000\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"7IcbYFfgia2q","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":36},"outputId":"65d15cf2-9430-422f-d964-c6534cdd69ba","executionInfo":{"status":"ok","timestamp":1569903477155,"user_tz":-540,"elapsed":441,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["print(X_train.shape[1])"],"execution_count":6,"outputs":[{"output_type":"stream","text":["28\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"Ujz03VPszCFI","colab_type":"text"},"source":["## 2"]},{"cell_type":"code","metadata":{"id":"iI_0u7Wvif-K","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":92},"outputId":"8b80d365-8bf9-4c65-b008-5476ad9d9627","executionInfo":{"status":"ok","timestamp":1569904656244,"user_tz":-540,"elapsed":945,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["from tensorflow.examples.tutorials.mnist import input_data\n","mnist2 = input_data.read_data_sets(\"./mnist/data/\", one_hot=True)"],"execution_count":27,"outputs":[{"output_type":"stream","text":["Extracting ./mnist/data/train-images-idx3-ubyte.gz\n","Extracting ./mnist/data/train-labels-idx1-ubyte.gz\n","Extracting ./mnist/data/t10k-images-idx3-ubyte.gz\n","Extracting ./mnist/data/t10k-labels-idx1-ubyte.gz\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"NrCPDK4xkPih","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":73},"outputId":"b29d1b65-5765-4363-80af-4fa412d3215a","executionInfo":{"status":"ok","timestamp":1569904658967,"user_tz":-540,"elapsed":527,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["print(type(mnist2.train))\n","print(type(mnist2.train.images), type(mnist2.test.images))\n","print(type(mnist2.train.labels), type(mnist2.test.labels))"],"execution_count":28,"outputs":[{"output_type":"stream","text":["<class 'tensorflow.contrib.learn.python.learn.datasets.mnist.DataSet'>\n","<class 'numpy.ndarray'> <class 'numpy.ndarray'>\n","<class 'numpy.ndarray'> <class 'numpy.ndarray'>\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"P3VvffG_kSCq","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":73},"outputId":"c4e391e0-ffd1-485a-c6fb-68ffa5b186d7","executionInfo":{"status":"ok","timestamp":1569904707699,"user_tz":-540,"elapsed":472,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["print(mnist2.train.images.shape, mnist2.train.labels.shape)\n","print(mnist2.test.images.shape, mnist2.test.labels.shape)\n","print(mnist2.validation.images.shape, mnist2.validation.labels.shape)"],"execution_count":32,"outputs":[{"output_type":"stream","text":["(55000, 784) (55000, 10)\n","(10000, 784) (10000, 10)\n","(5000, 784) (5000, 10)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"NhwKLrignppB","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":36},"outputId":"b919b759-1b4d-4eb6-d036-19fba13f003d","executionInfo":{"status":"ok","timestamp":1569905000430,"user_tz":-540,"elapsed":481,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["mnist2.train.images[5].shape"],"execution_count":34,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(784,)"]},"metadata":{"tags":[]},"execution_count":34}]},{"cell_type":"code","metadata":{"id":"ta5q80gHk9o6","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":36},"outputId":"fd58a516-2131-4add-941d-c15f686bb198","executionInfo":{"status":"ok","timestamp":1569904226451,"user_tz":-540,"elapsed":483,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["mnist2.train.labels[5]"],"execution_count":17,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([0., 0., 0., 0., 0., 0., 0., 0., 1., 0.])"]},"metadata":{"tags":[]},"execution_count":17}]},{"cell_type":"code","metadata":{"id":"tmWQ7Rx8lHXa","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":288},"outputId":"ff18c4dd-3ef2-4288-ba2a-f8bbb676c58f","executionInfo":{"status":"ok","timestamp":1569904676932,"user_tz":-540,"elapsed":674,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["%matplotlib inline\n","import matplotlib.pyplot as plt\n","plt.imshow(mnist2.train.images[5].reshape(28,28))"],"execution_count":30,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.image.AxesImage at 0x7f93f661c5c0>"]},"metadata":{"tags":[]},"execution_count":30},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"code","metadata":{"id":"kDaZM4wjsx96","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":266},"outputId":"79aafa78-94e9-495d-9bb9-0bfc9cf9fa48","executionInfo":{"status":"ok","timestamp":1569906638654,"user_tz":-540,"elapsed":1507,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["%matplotlib inline\n","import matplotlib.pyplot as plt\n","plt.figure(figsize=(10,4))\n","\n","for i in range(0, 10):\n","  plt.subplot(2,5,i+1)\n","  plt.imshow(mnist2.train.images[i].reshape(28,28))"],"execution_count":53,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAlMAAAD5CAYAAAADWbBJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n","text/plain":["<Figure size 720x288 with 10 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"code","metadata":{"id":"GqB0uR2Lv8oY","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":315},"outputId":"e902594a-cecc-4ad4-8f9e-dcb7493dc34d","executionInfo":{"status":"ok","timestamp":1569907303493,"user_tz":-540,"elapsed":1564,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["import matplotlib.pyplot as plt\n","\n","figure = plt.figure()\n","figure.set_size_inches(12,5)\n","\n","axes = [ ]\n","for i in range(1,11):\n","  axes.append(figure.add_subplot(2,5,i))\n","  \n","for i in range(10):\n","  axes[i].matshow(X_train[i])"],"execution_count":62,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAsMAAAEqCAYAAADwL12VAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASC0GwwAAAEgtBsMAAABIrf8fLIIYh6mGOn0AAAAASUVORK5CYII=\n","text/plain":["<Figure size 864x360 with 10 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"markdown","metadata":{"id":"sGvysI4FzHNV","colab_type":"text"},"source":["## 3"]},{"cell_type":"code","metadata":{"id":"4wpY4xh1mFaq","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":55},"outputId":"0c112b90-6f87-4edd-ff68-c3a2055b35ff","executionInfo":{"status":"ok","timestamp":1569905236224,"user_tz":-540,"elapsed":661,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["from keras.datasets import mnist\n","\n","(X_train, y_train), (X_test, y_test) = mnist.load_data()\n","print(X_train.shape, X_test.shape)\n","print(y_train.shape, y_test.shape)"],"execution_count":40,"outputs":[{"output_type":"stream","text":["(60000, 28, 28) (10000, 28, 28)\n","(60000,) (10000,)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"jC_w74J6pdH6","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":55},"outputId":"4fd2729f-191d-49b6-fa5a-74d30a9cbf3e","executionInfo":{"status":"ok","timestamp":1569907380246,"user_tz":-540,"elapsed":470,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["# 데이터 변환하기\n","print(X_train.shape)\n","X_train = X_train.reshape(60000, 28*28)\n","print(X_train.shape)"],"execution_count":64,"outputs":[{"output_type":"stream","text":["(60000, 28, 28)\n","(60000, 784)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"GIrBNfGmw4L4","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":36},"outputId":"a2061e82-d09f-405c-fb4f-772fb2e170c9","executionInfo":{"status":"ok","timestamp":1569907400107,"user_tz":-540,"elapsed":464,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["X_test = X_test.reshape(10000, 28*28)\n","print(X_test.shape)"],"execution_count":66,"outputs":[{"output_type":"stream","text":["(10000, 784)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"JfUu5FWwxQIJ","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":73},"outputId":"fe262624-f01f-42ac-df1a-f56c58435c04","executionInfo":{"status":"ok","timestamp":1569907560421,"user_tz":-540,"elapsed":492,"user":{"displayName":"백서연","photoUrl":"","userId":"13608049772715071568"}}},"source":["## one-hot 인코딩\n","import numpy as np\n","\n","# train data\n","print(y_train.shape)\n","y_train_hot = np.eye(10)[y_train]\n","print(y_train_hot.shape)\n","\n","# test data\n","y_test_hot = np.eye(10)[y_test]\n","print(y_test_hot.shape)"],"execution_count":70,"outputs":[{"output_type":"stream","text":["(60000,)\n","(60000, 10)\n","(10000, 10)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"1BkAKj0Rx1Cp","colab_type":"code","colab":{}},"source":[""],"execution_count":0,"outputs":[]}]}