tensorflow搭建网络学习过程(二分类问题）（1）

序列预处理

首先读取图片目录里各图片的名称存放在train_id.txt文件夹中，（其实我的数据集图片名称及对应标签全都存放在一个文本文档里面）具体代码如下：
我的代码

with open("data/train/train.txt", 'r') as f:
rnames = f.read().splitlines()
for i in range(len(rnames)):
names.append(rnames[i].split(" ")[0])
with open('data/train/train_id.txt','w') as f:
f.writelines(lines+'\n' for lines in names)
f.close()
print(rnames)

这是读取图片目录里各图片的名称存放在train_id.txt文件夹中的代码

import os
import random

image_path="data/train/imgs/"
#你自己的图片文件路径
names=os.list.dir(image_path)
np.random.shuffle(names)
#将图片名称顺序随机打乱
with open('data/train/train_id','w')as f:
f.writelines(lines+'\n' for lines in names)
f.close()
print("txt文件创建成功！")

序列化数据处理

import scipy.misc
import numpy as np
from keras.utils import Sequence, to_categorical
import os

train_images_folder='data/'
valid_images_floder='data/'
train_labels='data/train/train_labels.txt'
valid_labels='data/valid/val_labels.txt'

batch_size=4
H,W=64,64

def read_labellist(labeldir):
with open(labeldir, "r") as f:
labels = f.read().splitlines()
print(labels)
# labels=np.asarray(labels).astype('float32')
# print(labels)
# one_hot_labels=to_categorical(train_label,2)
# print(one_hot_labels)
# batch_y=[]
# for i_batch in range(3):
#     batch_y.append(train_label[i:i+1])
#     i+=1
# print(batch_y)
return labels
# names=read_labellist(images_label)
#读取图片并将其转化为可输入网络的张量
class DataGenSequence(Sequence):
def __init__(self,usage):
self.usage=usage

if usage=='train':
id_file='data/train/train_id.txt'
self.images_folder=train_images_folder
self.labels = read_labellist(train_labels)

else:
id_file='data/valid/val_id.txt'
self.images_folder=valid_images_floder
self.labels = read_labellist(valid_labels)

with open(id_file,'r') as f:
self.names=f.read().splitlines()

def __len__(self):
#计算每一次epoch中数据要分成几批处理
return int(np.ceil(len(self.names) / float(batch_size)))

def __getitem__(self, idx):
i=idx*batch_size
# i=0
length=min(batch_size,(len(self.names)-i))
# length=4
batch_x=np.empty((batch_size,H,W,3),dtype=np.float32)
batch_y =np.empty((batch_size,1,1,2))

for i_batch in range(length):
name=self.names[i]
filename=os.path.join(self.images_folder,name)
img=scipy.misc.imread(filename)
if np.random.random_sample() > 0.5:
img = np.fliplr(img)

x = img / 255.
y = []

batch_x[i_batch,:,:,0:3]=x
y.append(self.labels[i:i+1])

batch_y[i_batch,0,:,:]=to_categorical(y,2)
i+=1
return batch_x,batch_y

def train_gen():
return DataGenSequence('train')

def valid_gen():
return DataGenSequence('valid')

训练集和验证集，这部分等我把整个网络完成后再划分。
训练集，验证集已完成
输入图片维度和标签维度要相同，都是4维。
好了，我今天的博客就到这里了，等我把网络层搭建好后会再放另一篇博客呦！
如果我博客的内容对你有帮助，请给博主点个赞啊！这是我动力的来源。