SVHN的简单实现

以下代码只是博主的学习笔记，由于为了稍稍符合读者需求，经过了一些修改，可能在修改过程中出现一些小bug，见谅

首先介绍一下SVHN（Street View House Number）Dateset 来源于谷歌街景门牌号码，这里提供一下地址，本代码只需要下载格式2中的train和test.mat即可，格式2中所有数字均已调整为固定的32 x 32像素分辨率，原始字符边框在适当的尺寸上扩展为方形窗口，因此将它们的大小调整为32 x 32像素不会引起纵横比失真。然而，这种预处理将一​​些分散注意力的数字引入到感兴趣数字的侧面。加载.mat文件会创建2个变量：X是包含图像的4-D矩阵，而y是类标签的向量。

话不多说，进入正题，下面我写了两份代码，分别是生成tfl模型和h5模型的，其中生成的tfl模型是对其他博主的源码稍稍修改后的复用，地址也放在这了

---

数据处理

数据集含有两个变量 X 代表图像， 训练集 X 的 shape 是  （32，32，3，73257） 也就是(width, height, channels, samples), tensorflow 的张量需要 （samples, width, height, channels），所以需要转换一下，由于直接调用 cifar 10 的网络模型，数据只需要先做个归一化，所有像素除于255就 OK，另外原始数据 0 的标签是 10，这里要转化成 0，并提供 one_hot 编码。

下面的代码建立一个py文件保存，以供后续训练代码调用

#!/usr/bin/env python2
# -*- coding: utf-8 -*-
import scipy.io as sio
import matplotlib.pyplot as plt
import numpy as np

image_size = 32
num_labels = 10


def display_data():
    print('loading Matlab data...')
    train = sio.loadmat('Data/train_32x32.mat')
    data = train['X']
    label = train['y']
    for i in range(10):
        plt.subplot(2, 5, i + 1)
        plt.title(label[i][0])
        plt.imshow(data[..., i])
        plt.axis('off')
    plt.show()


def load_data(one_hot=False):
    train = sio.loadmat('Data/train_32x32.mat')
    test = sio.loadmat('Data/test_32x32.mat')

    train_data = train['X']
    train_label = train['y']
    test_data = test['X']
    test_label = test['y']

    train_data = np.swapaxes(train_data, 0, 3)
    train_data = np.swapaxes(train_data, 2, 3)
    train_data = np.swapaxes(train_data, 1, 2)
    test_data = np.swapaxes(test_data, 0, 3)
    test_data = np.swapaxes(test_data, 2, 3)
    test_data = np.swapaxes(test_data, 1, 2)

    test_data = test_data / 255.
    train_data = train_data / 255.

    for i in range(train_label.shape[0]):
        if train_label[i][0] == 10:
            train_label[i][0] = 0

    for i in range(test_label.shape[0]):
        if test_label[i][0] == 10:
            test_label[i][0] = 0

    if one_hot:
        train_label = (np.arange(num_labels) == train_label[:, ]).astype(np.float32)
        test_label = (np.arange(num_labels) == test_label[:, ]).astype(np.float32)

    return train_data, train_label, test_data, test_label


if __name__ == '__main__':
    load_data(one_hot=True)
    display_data()

 

---

tflearn训练及模型保存

'''
tflearn训练代码
'''

from __future__ import division, print_function, absolute_import

import tflearn
from tflearn.data_utils import shuffle, to_categorical
from tflearn.layers.core import input_data, dropout, fully_connected
from tflearn.layers.conv import conv_2d, max_pool_2d
from tflearn.layers.estimator import regression
from tflearn.data_preprocessing import ImagePreprocessing
from tflearn.data_augmentation import ImageAugmentation

# Data loading and preprocessing
import transData as SVHN  # 这里换成你自己的数据处理的文件名称
X, Y, X_test, Y_test = SVHN.load_data(one_hot = True)
X, Y = shuffle(X, Y)

# Real-time data preprocessing
img_prep = ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()


# Convolutional network building
network = input_data(shape=[None, 32, 32, 3],
                     data_preprocessing=img_prep)
network = conv_2d(network, 32, 3, activation='relu')
network = max_pool_2d(network, 2)
network = conv_2d(network, 64, 3, activation='relu')
network = conv_2d(network, 64, 3, activation='relu')
network = max_pool_2d(network, 2)
network = fully_connected(network, 512, activation='relu')
network = dropout(network, 0.5)
network = fully_connected(network, 10, activation='softmax')
network = regression(network, optimizer='adam',
                     loss='categorical_crossentropy',
                     learning_rate=0.001)

# Train using classifier
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(X, Y, n_epoch=15, shuffle=True, validation_set=(X_test, Y_test),
          show_metric=True, batch_size=96, run_id='svhn_cnn')

model.save("./models/model1/svhnCNN.tfl")

keras训练及模型保存

'''
由tflearn训练代码修改的keras训练代码
'''

from __future__ import division, print_function, absolute_import

from tflearn.data_utils import shuffle
from tflearn.data_preprocessing import ImagePreprocessing

import keras
from keras import Sequential
from keras.layers import Conv2D, MaxPooling2D, Dropout, Flatten, Dense
import matplotlib.pyplot as plt

# 数据加载和预处理
import transData as SVHN  # 这里换成你自己的数据处理的文件名称
X, Y, X_test, Y_test = SVHN.load_data(one_hot = True)
X, Y = shuffle(X, Y)

# 实时数据预处理
epochs = 15
batch_size = 96
img_prep = ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()

input_shape = (32, 32, 3)

# 卷积网络搭建
model = Sequential()
model.add(Conv2D(filters=32, kernel_size=(3, 3), activation='relu',
                 input_shape=input_shape))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(filters=64, kernel_size=(3, 3), activation='relu'))
model.add(Conv2D(filters=64, kernel_size=(3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())
model.add(Dense(units=512, activation='relu'))
model.add(Dropout(rate=0.5))
model.add(Dense(units=10, activation='softmax'))
model.summary()
model.compile(loss=keras.losses.categorical_crossentropy,
              optimizer=keras.optimizers.Adadelta(),
              metrics=['accuracy'])

# 训练模型
model.fit(X, Y,
          batch_size=batch_size,
          epochs=epochs,
          verbose=1,
          validation_data=(X_test, Y_test))



# 保存
model.save("svhnCNN.h5")

---

模型加载

tflearn

'''
加载tflearn模型
'''
import tflearn
from tflearn.data_utils import shuffle, to_categorical
from tflearn.layers.core import input_data, dropout, fully_connected
from tflearn.layers.conv import conv_2d, max_pool_2d, global_avg_pool
from tflearn.layers.estimator import regression
from tflearn.layers.normalization import batch_normalization
from tflearn.data_preprocessing import ImagePreprocessing
from tflearn.data_augmentation import ImageAugmentation
import numpy as np
from load_input import load_test_data
import cv2 as cv


X_test, Y_test = load_test_data()
X_test, Y_test = shuffle(X_test, Y_test)

# Real-time data preprocessing
img_prep = ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()


# Convolutional network building
network = input_data(shape=[None, 32, 32, 3],
                     data_preprocessing=img_prep)
network = conv_2d(network, 32, 3, activation='relu')
network = max_pool_2d(network, 2)
network = conv_2d(network, 64, 3, activation='relu')
network = conv_2d(network, 64, 3, activation='relu')
network = max_pool_2d(network, 2)
network = fully_connected(network, 512, activation='relu')
network = dropout(network, 0.5)
network = fully_connected(network, 10, activation='softmax')
network = regression(network, optimizer='adam',
                     loss='categorical_crossentropy',
                     learning_rate=0.001)

# Train using classifier
model = tflearn.DNN(network, tensorboard_verbose=0)


model.load("./models/model1/svhnCNN.tfl")


img = cv.imread('你的测试图片路径')
img = cv.resize(img, (32, 32))
img = img.reshape(1, 32, 32, 3)
img = img.astype('float32') / 255.0  # 归一化
results = model.predict(img)
print(results)
result = results[0].argmax()
print(result)

keras

'''
加载keras模型
'''
import cv2
import numpy as np
from keras import models
import matplotlib.pyplot as plt

def transMNIST(path, size=(32, 32)):
    img = cv2.imread(path)
    img = cv2.imread(path)
    img = cv2.resize(img, size)
    # cv2.imshow('img', img)
    # cv2.waitKey(0)
    img = img.reshape(1, 32, 32, 3)
    # print(img)
    return img


def predict(imgData):
    my_mnist_model = models.load_model('models/model2/svhnCNN.h5')
    # print(my_mnist_model.summary())
    img = imgData.astype('float32') / 255.0  # 归一化
    results = my_mnist_model.predict(img)
    result_number = []
    result = results[0].argmax()
    print(result)
    # for result in results:
    #     result_number.append(np.argmax(result))
    return result


path = '你的路径'  
imgData = transMNIST(path)
results = predict(imgData)