Cifar 是加拿大政府牵头投资的一个先进科学项目研究所。Hinton、Bengio和他的学生在2004年拿到了 Cifar 投资的少量资金，建立了神经计算和自适应感知项目。这个项目结集了不少计算机科学家、生物学家、电气工程师、神经科学家、物理学家、心理学家，加速推动了 Deep Learning 的进程。从这个阵容来看，DL 已经和 ML 系的数据挖掘分的很远了。Deep Learning 强调的是自适应感知和人工智能，是计算机与神经科学交叉；Data Mining 强调的是高速、大数据、统计数学分析，是计算机和数学的交叉。
Cifar-10 是由 Hinton 的学生 Alex Krizhevsky、Ilya Sutskever 收集的一个用于普适物体识别的数据集。
我们在前面的学习中，使用了MNIST和Fashion-MNIST两个数据集来练习卷积网络的分类，但是这两个数据集都是单通道的灰度图。虽然我们用彩色的几何图形作为例子讲解了卷积网络的基本功能，但是仍然与现实的彩色世界有差距。所以，本节我们将使用Cifar-10数据集来进一步检验一下卷积神经网络的能力。
图18-41是Cifar-10的样本数据。
- airplane，飞机，6000张
- automobile，汽车，6000张
- bird，鸟，6000张
- cat，猫，6000张
- deer，鹿，6000张
- dog，狗，6000张
- frog，蛙，6000张
- horse，马，6000张
- ship，船，6000张
- truck，卡车，6000张
Cifar-10 由60000张32*32的 RGB 彩色图片构成，共10个分类。50000张训练，10000张测试。分为6个文件，5个训练数据文件，每个文件中包含10000张图片，随机打乱顺序，1个测试数据文件，也是10000张图片。这个数据集最大的特点在于将识别迁移到了普适物体，而且应用于多分类（姊妹数据集Cifar-100达到100类，ILSVRC比赛则是1000类）。
import numpy as np
import pandas as pd
import os
import matplotlib.pyplot as plt
import pickle
from sklearn.preprocessing import MinMaxScaler
from keras.utils import to_categorical
from keras.models import Sequential, load_model
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout, Input, Activation, add, BatchNormalization, AveragePooling2D
from keras.applications import ResNet50
from keras.regularizers import l2
names = ["airplane", "car", "bird", "cat", "deer", "dog", "frog", "horse", "ship", "truck"]
def unpickle(file):
with open(file, 'rb') as fo:
dict = pickle.load(fo, encoding='bytes')
return dict
def load_train_data(data_path):
batch = unpickle(data_path + '/data_batch_' + str(1))
x_train = batch[b'data'].reshape((len(batch[b'data']), 3, 32, 32)).transpose(0, 2, 3, 1)
y_train = batch[b'labels']
for i in range(2, 6):
batch = unpickle(data_path + '/data_batch_' + str(i))
x_train = np.concatenate([x_train, batch[b'data'].reshape((len(batch[b'data']), 3, 32, 32)).transpose(0, 2, 3, 1)])
y_train = np.concatenate([y_train, batch[b'labels']])
return x_train, y_train
def process_train_data(x_train, y_train):
minmax = MinMaxScaler()
x_train_rows = x_train.reshape(x_train.shape[0], 32 * 32 * 3)
x_train = minmax.fit_transform(x_train_rows)
x_train = x_train.reshape(x_train.shape[0], 32, 32, 3)
y_train = to_categorical(y_train)
return x_train, y_train
def load_test_data(data_path):
batch = unpickle(data_path + '/test_batch')
x_test = batch[b'data'].reshape((len(batch[b'data']), 3, 32, 32)).transpose(0, 2, 3, 1)
y_test = batch[b'labels']
return x_test, y_test
def process_test_data(x_test, y_test):
minmax = MinMaxScaler()
x_test_rows = x_test.reshape(x_test.shape[0], 32 * 32 * 3)
x_test = minmax.fit_transform(x_test_rows)
x_test = x_test.reshape(x_test.shape[0], 32, 32, 3)
y_test = to_categorical(y_test)
return x_test, y_test
def draw_train_history(history):
plt.figure(1)
# summarize history for accuracy
plt.subplot(211)
plt.plot(history.history['accuracy'])
plt.plot(history.history['val_accuracy'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
# summarize history for loss
plt.subplot(212)
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'validation'], loc='upper left')
plt.show()
def show_result(x, y, y_raw):
fig, ax = plt.subplots(nrows=8, ncols=8, figsize=(11, 11))
for i in range(64):
ax[i // 8, i % 8].imshow(x[i])
if y[i] == y_raw[i]:
ax[i // 8, i % 8].set_title(names[y_raw[i]])
else:
ax[i // 8, i % 8].set_title(names[y_raw[i]]+ "(" + names[y[i]] + ")", fontdict={'color':'r'})
ax[i // 8, i % 8].axis('off')
# endfor
plt.show()
def build_model():
model = ResNet50(weights=None, classes=10, input_shape=(32,32,3))
model.compile(optimizer='rmsprop',
loss='categorical_crossentropy',
metrics=['accuracy'])
return model
if __name__ == "__main__":
x_train, y_train = load_train_data('../data/cifar')
x_train, y_train = process_train_data(x_train, y_train)
x_test, y_test = load_test_data('../data/cifar')
x_test, y_test = process_train_data(x_test, y_test)
print(x_train.shape)
print(y_train.shape)
print(x_test.shape)
print(y_test.shape)
model_path = "cifar/cifar_model_resnet50.h5"
if os.path.exists(model_path):
model = load_model(model_path)
else:
model = build_model()
history = model.fit(x_train, y_train,
batch_size=64,
epochs=1,
validation_split=0.2)
model.save(model_path)
draw_train_history(history)
loss, accuracy = model.evaluate(x_test, y_test)
print("test loss: {}, test accuracy: {}".format(loss, accuracy))
z = model.predict(x_test[0:64])
show_result(x_test[0:64], np.argmax(z, axis=1), np.argmax(y_test, axis=1))
test loss: 1.512419725227356, test accuracy: 0.6747
****Cifar10-Resnet50-keras****