MNIST数据集分类.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from keras.datasets import mnist\n",
    "from keras.utils import np_utils\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "# from keras.optimizers import SGD\n",
    "from tensorflow.keras.optimizers import SGD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_shape: (60000, 28, 28)\n",
      "y_shape: (60000,)\n"
     ]
    }
   ],
   "source": [
    "# 载入数据\n",
    "(x_train,y_train),(x_test,y_test) = mnist.load_data()\n",
    "# (60000,28,28)\n",
    "print('x_shape:',x_train.shape)\n",
    "# (60000)\n",
    "print('y_shape:',y_train.shape)\n",
    "# (60000,28,28)->(60000,784)\n",
    "x_train = x_train.reshape(x_train.shape[0],-1)/255.0\n",
    "x_test = x_test.reshape(x_test.shape[0],-1)/255.0\n",
    "# 换one hot格式\n",
    "y_train = np_utils.to_categorical(y_train,num_classes=10)\n",
    "y_test = np_utils.to_categorical(y_test,num_classes=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\_SoftWare\\Miniconda3\\lib\\site-packages\\keras\\optimizer_v2\\optimizer_v2.py:355: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "# 创建模型，输入784个神经元，输出10个神经元\n",
    "model = Sequential([\n",
    "    Dense(units=10, input_dim=784, bias_initializer='one', activation='softmax')\n",
    "])\n",
    "\n",
    "# 定义优化器，loss function，训练过程中计算准确率\n",
    "sgd = SGD(lr=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(\n",
    "    optimizer=sgd,\n",
    "    loss='mse',\n",
    "    metrics=['accuracy'],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "1875/1875 [==============================] - 3s 958us/step - loss: 0.0388 - accuracy: 0.7635\n",
      "Epoch 2/10\n",
      "1875/1875 [==============================] - 2s 903us/step - loss: 0.0204 - accuracy: 0.8813\n",
      "Epoch 3/10\n",
      "1875/1875 [==============================] - 2s 924us/step - loss: 0.0177 - accuracy: 0.8927\n",
      "Epoch 4/10\n",
      "1875/1875 [==============================] - 2s 993us/step - loss: 0.0164 - accuracy: 0.8991\n",
      "Epoch 5/10\n",
      "1875/1875 [==============================] - 2s 962us/step - loss: 0.0156 - accuracy: 0.9035\n",
      "Epoch 6/10\n",
      "1875/1875 [==============================] - 2s 866us/step - loss: 0.0150 - accuracy: 0.9072\n",
      "Epoch 7/10\n",
      "1875/1875 [==============================] - 2s 875us/step - loss: 0.0146 - accuracy: 0.9090\n",
      "Epoch 8/10\n",
      "1875/1875 [==============================] - 2s 920us/step - loss: 0.0143 - accuracy: 0.9107\n",
      "Epoch 9/10\n",
      "1875/1875 [==============================] - 2s 880us/step - loss: 0.0140 - accuracy: 0.9124\n",
      "Epoch 10/10\n",
      "1875/1875 [==============================] - 2s 904us/step - loss: 0.0138 - accuracy: 0.9137\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x2338344d520>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 训练模型\n",
    "model.fit(x_train,y_train,batch_size=32,epochs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 0s 1ms/step - loss: 0.0130 - accuracy: 0.9182\n",
      "test loss 0.013032983057200909\n",
      "accuracy 0.9182000160217285\n"
     ]
    }
   ],
   "source": [
    "# 评估模型\n",
    "loss,accuracy = model.evaluate(x_test,y_test)\n",
    "\n",
    "print('test loss',loss)\n",
    "print('accuracy',accuracy)\n",
    "\n",
    "# 保存模型\n",
    "# model.save('model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 保存参数，载入参数\n",
    "model.save_weights('MNIST数据集分类.h5')\n",
    "model.load_weights('MNIST数据集分类.h5')\n",
    "# 保存网络结构，载入网络结构\n",
    "from keras.models import model_from_json\n",
    "json_string = model.to_json()\n",
    "model = model_from_json(json_string)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{\"class_name\": \"Sequential\", \"config\": {\"name\": \"sequential\", \"layers\": [{\"class_name\": \"InputLayer\", \"config\": {\"batch_input_shape\": [null, 784], \"dtype\": \"float32\", \"sparse\": false, \"ragged\": false, \"name\": \"dense_input\"}}, {\"class_name\": \"Dense\", \"config\": {\"name\": \"dense\", \"trainable\": true, \"batch_input_shape\": [null, 784], \"dtype\": \"float32\", \"units\": 10, \"activation\": \"softmax\", \"use_bias\": true, \"kernel_initializer\": {\"class_name\": \"GlorotUniform\", \"config\": {\"seed\": null}}, \"bias_initializer\": {\"class_name\": \"Ones\", \"config\": {}}, \"kernel_regularizer\": null, \"bias_regularizer\": null, \"activity_regularizer\": null, \"kernel_constraint\": null, \"bias_constraint\": null}}]}, \"keras_version\": \"2.6.0\", \"backend\": \"tensorflow\"}\n"
     ]
    }
   ],
   "source": [
    "print(json_string)"
   ]
  }
 ],
 "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": "3.9.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}