uchidama/keras_load_model_viewer.ipynb

## keras_load_model_viewer.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "IMPORT MODULES\n",
    "'''\n",
    "from __future__ import print_function\n",
    "import keras\n",
    "from keras.datasets import mnist\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten\n",
    "from keras.layers import Conv2D, MaxPooling2D\n",
    "from keras import backend as K\n",
    "from keras.models import load_model\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import cm\n",
    "%matplotlib inline\n",
    "\n",
    "import random"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "'''\n",
    "LOAD MNIST DATA\n",
    "'''\n",
    "# input image dimensions\n",
    "img_rows, img_cols = 28, 28\n",
    "\n",
    "# the data, shuffled and split between train and test sets\n",
    "(x_train, y_train), (x_test, y_test) = mnist.load_data()\n",
    "\n",
    "if K.image_data_format() == 'channels_first':\n",
    "    x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)\n",
    "    x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)\n",
    "    input_shape = (1, img_rows, img_cols)\n",
    "else:\n",
    "    x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)\n",
    "    x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)\n",
    "    input_shape = (img_rows, img_cols, 1)\n",
    "\n",
    "x_train = x_train.astype('float32')\n",
    "x_test = x_test.astype('float32')\n",
    "x_train /= 255\n",
    "x_test /= 255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "index: 7703\n",
      "predict ret: [[ 1.  0.  0.  0.  0.  0.  0.  0.  0.  0.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP4AAAD8CAYAAABXXhlaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCl9RcogKX1FyyP8HZBGVgVGozRoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x145baefd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "I think this digit is a  0\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "LOAD MODEL AND PREDICT\n",
    "'''\n",
    "model = load_model('model_mnist_cnn.h5')\n",
    "\n",
    "index = random.randint(0, len(x_test)-1)\n",
    "print(\"index:\", index)\n",
    "\n",
    "ret = model.predict(x_test[index:(index+1)], batch_size=1)   # OK\n",
    "print(\"predict ret:\", ret)\n",
    "\n",
    "np.set_printoptions(precision=5, suppress=True) #指数表示の禁止\n",
    "\n",
    "plt.imshow(x_test[index].reshape(28, 28), cmap=cm.gray_r)\n",
    "plt.show()\n",
    "\n",
    "bestnum = 0.0\n",
    "bestclass = 0\n",
    "for n in [0,1,2,3,4,5,6,7,8,9]:\n",
    "    if bestnum < ret[0][n]:\n",
    "        bestnum = ret[0][n]\n",
    "        bestclass = n\n",
    "\n",
    "print(\"I think this digit is a \", bestclass)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}
	{
	"cells": [
	{
	"cell_type": "code",
	"execution_count": 1,
	"metadata": {
	"collapsed": false
	},
	"outputs": [
	{
	"name": "stderr",
	"output_type": "stream",
	"text": [
	"Using TensorFlow backend.\n"
	]
	}
	],
	"source": [
	"'''\n",
	"IMPORT MODULES\n",
	"'''\n",
	"from __future__ import print_function\n",
	"import keras\n",
	"from keras.datasets import mnist\n",
	"from keras.models import Sequential\n",
	"from keras.layers import Dense, Dropout, Flatten\n",
	"from keras.layers import Conv2D, MaxPooling2D\n",
	"from keras import backend as K\n",
	"from keras.models import load_model\n",
	"\n",
	"import numpy as np\n",
	"\n",
	"import matplotlib.pyplot as plt\n",
	"from matplotlib import cm\n",
	"%matplotlib inline\n",
	"\n",
	"import random"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 2,
	"metadata": {
	"collapsed": false
	},
	"outputs": [],
	"source": [
	"'''\n",
	"LOAD MNIST DATA\n",
	"'''\n",
	"# input image dimensions\n",
	"img_rows, img_cols = 28, 28\n",
	"\n",
	"# the data, shuffled and split between train and test sets\n",
	"(x_train, y_train), (x_test, y_test) = mnist.load_data()\n",
	"\n",
	"if K.image_data_format() == 'channels_first':\n",
	" x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)\n",
	" x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)\n",
	" input_shape = (1, img_rows, img_cols)\n",
	"else:\n",
	" x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)\n",
	" x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)\n",
	" input_shape = (img_rows, img_cols, 1)\n",
	"\n",
	"x_train = x_train.astype('float32')\n",
	"x_test = x_test.astype('float32')\n",
	"x_train /= 255\n",
	"x_test /= 255"
	]
	},
	{
	"cell_type": "code",
	"execution_count": 3,
	"metadata": {
	"collapsed": false
	},
	"outputs": [
	{
	"name": "stdout",
	"output_type": "stream",
	"text": [
	"index: 7703\n",
	"predict ret: [[ 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]\n"
	]
	},
	{
	"data": {
	"image/png": "iVBORw0KGgoAAAANSUhEUgAAAP4AAAD8CAYAAABXXhlaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCl9RcogKX1FyyP8HZBGVgVGozRoAAAAASUVORK5CYII=\n",
	"text/plain": [
	"<matplotlib.figure.Figure at 0x145baefd0>"
	]
	},
	"metadata": {},
	"output_type": "display_data"
	},
	{
	"name": "stdout",
	"output_type": "stream",
	"text": [
	"I think this digit is a 0\n"
	]
	}
	],
	"source": [
	"'''\n",
	"LOAD MODEL AND PREDICT\n",
	"'''\n",
	"model = load_model('model_mnist_cnn.h5')\n",
	"\n",
	"index = random.randint(0, len(x_test)-1)\n",
	"print(\"index:\", index)\n",
	"\n",
	"ret = model.predict(x_test[index:(index+1)], batch_size=1) # OK\n",
	"print(\"predict ret:\", ret)\n",
	"\n",
	"np.set_printoptions(precision=5, suppress=True) #指数表示の禁止\n",
	"\n",
	"plt.imshow(x_test[index].reshape(28, 28), cmap=cm.gray_r)\n",
	"plt.show()\n",
	"\n",
	"bestnum = 0.0\n",
	"bestclass = 0\n",
	"for n in [0,1,2,3,4,5,6,7,8,9]:\n",
	" if bestnum < ret[0][n]:\n",
	" bestnum = ret[0][n]\n",
	" bestclass = n\n",
	"\n",
	"print(\"I think this digit is a \", bestclass)"
	]
	},
	{
	"cell_type": "code",
	"execution_count": null,
	"metadata": {
	"collapsed": true
	},
	"outputs": [],
	"source": []
	}
	],
	"metadata": {
	"kernelspec": {
	"display_name": "Python 3",
	"language": "python",
	"name": "python3"
	},
	"language_info": {
	"codemirror_mode": {
	"name": "ipython",
	"version": 3
	},
	"file_extension": ".py",
	"mimetype": "text/x-python",
	"name": "python",
	"nbconvert_exporter": "python",
	"pygments_lexer": "ipython3",
	"version": "3.5.1"
	}
	},
	"nbformat": 4,
	"nbformat_minor": 0
	}