遗失宠物的智能寻找-红鲤鱼

1 遗失宠物的智能寻找/好用的宠物识别-红鲤鱼/README.md

@@ -0,0 +1,10 @@
+# 赛题一：遗失宠物的智能寻找
+
+伴随着人们物质生活水平的提高，越来越多的家庭选择饲养小动物，它们不仅是家庭的宠物，很多人也将他们视作家庭成员之一，给予百般宠爱和呵护。然而即便如此，意外还是难免发生——宠物的意外走失给很多宠物家庭带来了困扰，满大街发传单、贴广告不仅浪费时间，成效也甚微。
+
+- 作品介绍：使用cnn网络，使得宠物的识别准确率提高。
+- 作品截图：见文件夹
+- 安装、编译指南：使用ipynb训练和识别，便于使用和训练，并且适合在多种平台上使用
+- 团队介绍：林君瑶 13101581895
+- 使用到的 AWS 技术：得益于AmazonSageMaker提供的高效的GPU服务，使得在训练cnn网络可以完成较大规模的网络训练，并且较快的速度带来了极佳的使用体验
+

1 遗失宠物的智能寻找/好用的宠物识别-红鲤鱼/main.ipynb

@@ -0,0 +1,138 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy\n",
+    "import pandas\n",
+    "from PIL import Image\n",
+    "from keras import backend as K\n",
+    "from keras.utils import np_utils\n",
+    "\n",
+    "def load_data(dataset_path):\n",
+    "    img = Image.open(dataset_path)\n",
+    "    img_ndarray = numpy.asarray(img, dtype='float64') / 256\n",
+    "    print(img_ndarray.shape)\n",
+    "    faces = numpy.empty((400,57,47))\n",
+    "    for row in range(20):\n",
+    "        for column in range(20):\n",
+    "            faces[row * 20 + column] = img_ndarray[row * 57:(row + 1) * 57, column * 47:(column + 1) * 47]\n",
+    "    label = numpy.empty(400)\n",
+    "    for i in range(40):\n",
+    "        label[i * 10:i * 10 + 10] = i\n",
+    "        label = label.astype(numpy.int)\n",
+    "        label = np_utils.to_categorical(label, 40) \n",
+    "\n",
+    "\n",
+    "    train_data = numpy.empty((320, 57,47))\n",
+    "    train_label = numpy.empty((320,40))\n",
+    "    valid_data = numpy.empty((40, 57,47))\n",
+    "    valid_label = numpy.empty((40,40))\n",
+    "    test_data = numpy.empty((40, 57,47))\n",
+    "    test_label = numpy.empty((40,40)) \n",
+    "\n",
+    "    for i in range(40):\n",
+    "        train_data[i * 8:i * 8 + 8] = faces[i * 10:i * 10 + 8]\n",
+    "        train_label[i * 8:i * 8 + 8] = label[i * 10:i * 10 + 8]\n",
+    "        valid_data[i] = faces[i * 10 + 8]\n",
+    "        valid_label[i] = label[i * 10 + 8]\n",
+    "        test_data[i] = faces[i * 10 + 9]\n",
+    "        test_label[i] = label[i * 10 + 9]\n",
+    "\n",
+    "    return [(train_data, train_label), (valid_data, valid_label),(test_data, test_label)]\n",
+    "\n",
+    "\n",
+    "if __name__ == '__main__':\n",
+    "    [(train_data, train_label), (valid_data, valid_label), (test_data, test_label)] = load_data('olivettifaces.gif')\n",
+    "    oneimg = train_data[0]*256\n",
+    "    print(oneimg)\n",
+    "    im = Image.fromarray(oneimg)\n",
+    "    im.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "np.random.seed(1337)\n",
+    "from keras.models import Sequential\n",
+    "from keras.layers import Dense, Activation, Flatten\n",
+    "from keras.layers import Conv2D, MaxPooling2D,AveragePooling2D\n",
+    "from PIL import Image\n",
+    "import FaceData\n",
+    "\n",
+    "batch_size = 128 \n",
+    "nb_classes = 40 \n",
+    "epochs = 600 \n",
+    "img_rows, img_cols = 57, 47\n",
+    "nb_filters = 32 \n",
+    "pool_size = (2, 2) \n",
+    "kernel_size = (5, 5)\n",
+    "input_shape = (img_rows, img_cols,1)\n",
+    "\n",
+    "[(X_train, Y_train), (X_valid, Y_valid),(X_test, Y_test)] =FaceData.load_data('olivettifaces.gif')\n",
+    "\n",
+    "X_train=X_train[:,:,:,np.newaxis] \n",
+    "X_valid=X_valid[:,:,:,np.newaxis]\n",
+    "X_test=X_test[:,:,:,np.newaxis]\n",
+    "\n",
+    "print('样本数据集的维度：', X_train.shape,Y_train.shape)\n",
+    "print('测试数据集的维度：', X_test.shape,Y_test.shape)\n",
+    "\n",
+    "\n",
+    "model = Sequential()\n",
+    "model.add(Conv2D(6,kernel_size,input_shape=input_shape,strides=1))\n",
+    "model.add(AveragePooling2D(pool_size=pool_size,strides=2)) \n",
+    "model.add(Conv2D(12,kernel_size,strides=1))\n",
+    "model.add(AveragePooling2D(pool_size=pool_size,strides=2))\n",
+    "model.add(Flatten())\n",
+    "model.add(Dense(nb_classes))\n",
+    "model.add(Activation('sigmoid'))\n",
+    "\n",
+    "\n",
+    "model.compile(loss='categorical_crossentropy',optimizer='adadelta',metrics=['accuracy'])\n",
+    "\n",
+    "model.fit(X_train, Y_train, batch_size=batch_size, epochs=epochs,verbose=1, validation_data=(X_test, Y_test))\n",
+    "\n",
+    "score = model.evaluate(X_test, Y_test, verbose=0)\n",
+    "print('Test score:', score[0])\n",
+    "print('Test accuracy:', score[1])\n",
+    "\n",
+    "y_pred = model.predict(X_test)\n",
+    "y_pred = y_pred.argmax(axis=1) \n",
+    "for i in range(len(y_pred)):\n",
+    "    oneimg = X_test[i,:,:,0]*256\n",
+    "    im = Image.fromarray(oneimg)\n",
+    "    im.show()\n",
+    "    print('第%d只狗识别为第%d个狗'%(i,y_pred[i]))"
+   ]
+  }
+ ],
+ "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": "3.8.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}