一、Python学习两大道具#

1. dir()工具#

• 作用：支持打开package，看到里面的工具函数

• 示例：

  (1) 输出torch库包含的函数

Copy
<span class="token builtin">dir</span><span class="token punctuation">(</span>torch<span class="token punctuation">)</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span></span>
<span class="token builtin">dir</span><span class="token punctuation">(</span>torch<span class="token punctuation">)</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span></span>
dir(torch)

(2) 输出torch.AVG函数中的参数

Copy
<span class="token builtin">dir</span><span class="token punctuation">(</span>torch<span class="token punctuation">.</span>AVG<span class="token punctuation">)</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span></span>
<span class="token builtin">dir</span><span class="token punctuation">(</span>torch<span class="token punctuation">.</span>AVG<span class="token punctuation">)</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span></span>
dir(torch.AVG)

2. help()工具#

• 作用：说明书，查看库中函数某个参数的说明或使用方法

• 示例：

  (1) 输出torch库中AVG函数的AVG参数使用方法

Copy
<span class="token builtin">help</span><span class="token punctuation">(</span>torch<span class="token punctuation">.</span>AVG<span class="token punctuation">.</span>AVG<span class="token punctuation">)</span> 
<span aria-hidden="true" class="line-numbers-rows"><span></span></span>
<span class="token builtin">help</span><span class="token punctuation">(</span>torch<span class="token punctuation">.</span>AVG<span class="token punctuation">.</span>AVG<span class="token punctuation">)</span> 
<span aria-hidden="true" class="line-numbers-rows"><span></span></span>
help(torch.AVG.AVG) 

二、Pytorch读取图像数据#

0. 写在前面：#

（1）PIL库中Image函数的基本使用方法#

• PIL的安装：win+r → cmd → 选择环境 → pip install Pillow

• 使用方法：

Copy
<span class="token keyword">from</span> PIL <span class="token keyword">import</span> Image
  
<span class="token comment">#选择图像路径</span>
<span class="token comment">#注意：在复制图像路径后，在windows环境下，需要将\变为\\</span>
img_path<span class="token operator">=</span><span class="token string">"E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train\\ants\\0013035.jpg"</span>
  
<span class="token comment">#打开并标识给定的图像文件。</span>
<span class="token comment">#image.open()是一个懒惰的操作；此函数可识别文件，但文件保持打开状态，直到尝试处理数据(或调用load()方法)，才会从文件中读取实际图像数据。</span>
img<span class="token operator">=</span>image<span class="token punctuation">.</span><span class="token builtin">open</span><span class="token punctuation">(</span>img_path<span class="token punctuation">)</span>
  
<span class="token comment">#输出图像大小</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>img<span class="token punctuation">.</span>size<span class="token punctuation">)</span>
  
<span class="token comment">#打开图像(指的是直接根据该地址打开一个窗口显示这个图像)</span>
img<span class="token punctuation">.</span>show<span class="token punctuation">(</span><span class="token punctuation">)</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
<span class="token keyword">from</span> PIL <span class="token keyword">import</span> Image
  
<span class="token comment">#选择图像路径</span>
<span class="token comment">#注意：在复制图像路径后，在windows环境下，需要将\变为\\</span>
img_path<span class="token operator">=</span><span class="token string">"E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train\\ants\\0013035.jpg"</span>
  
<span class="token comment">#打开并标识给定的图像文件。</span>
<span class="token comment">#image.open()是一个懒惰的操作；此函数可识别文件，但文件保持打开状态，直到尝试处理数据(或调用load()方法)，才会从文件中读取实际图像数据。</span>
img<span class="token operator">=</span>image<span class="token punctuation">.</span><span class="token builtin">open</span><span class="token punctuation">(</span>img_path<span class="token punctuation">)</span>
  
<span class="token comment">#输出图像大小</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>img<span class="token punctuation">.</span>size<span class="token punctuation">)</span>
  
<span class="token comment">#打开图像(指的是直接根据该地址打开一个窗口显示这个图像)</span>
img<span class="token punctuation">.</span>show<span class="token punctuation">(</span><span class="token punctuation">)</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
from PIL import Image
  
#选择图像路径
#注意：在复制图像路径后，在windows环境下，需要将\变为\\
img_path="E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train\\ants\\0013035.jpg"
  
#打开并标识给定的图像文件。
#image.open()是一个懒惰的操作；此函数可识别文件，但文件保持打开状态，直到尝试处理数据(或调用load()方法)，才会从文件中读取实际图像数据。
img=image.open(img_path)
  
#输出图像大小
print(img.size)
  
#打开图像(指的是直接根据该地址打开一个窗口显示这个图像)
img.show()

（2）os库中函数的基本使用方法#

• os.path.join(path1, path2, … ,pathn)：合并路径，在一定程度上可以避免因python语法问题，造成的\t或\n之类的错误。

  比如说，图片路径为：”hymenoptera_data\hymenoptera_data\train“，如果直接输入路径，那么会出现以下结果：

Copy
path<span class="token operator">=</span><span class="token string">"hymenoptera_data\hymenoptera_data\train"</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>path<span class="token punctuation">)</span>
  
<span class="token triple-quoted-string string">'''
[Run]
hymenoptera_data\hymenoptera_data    rain
'''</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
path<span class="token operator">=</span><span class="token string">"hymenoptera_data\hymenoptera_data\train"</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>path<span class="token punctuation">)</span>
  
<span class="token triple-quoted-string string">'''
[Run]
hymenoptera_data\hymenoptera_data    rain
'''</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
path="hymenoptera_data\hymenoptera_data\train"
print(path)
  
'''
[Run]
hymenoptera_data\hymenoptera_data    rain
'''

此时的处理方法有两种：

方法一：在\后面加个\

Copy
path<span class="token operator">=</span><span class="token string">"hymenoptera_data\\hymenoptera_data\\train"</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>path<span class="token punctuation">)</span>
<span class="token triple-quoted-string string">'''
[Run]
hymenoptera_data\hymenoptera_data\train
'''</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
path<span class="token operator">=</span><span class="token string">"hymenoptera_data\\hymenoptera_data\\train"</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>path<span class="token punctuation">)</span>

<span class="token triple-quoted-string string">'''
[Run]
hymenoptera_data\hymenoptera_data\train
'''</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
path="hymenoptera_data\\hymenoptera_data\\train"
print(path)

'''
[Run]
hymenoptera_data\hymenoptera_data\train
'''

方法二：os.path.join

Copy
<span class="token keyword">import</span> os
path1<span class="token operator">=</span><span class="token string">"hymenoptera_data\hymenoptera_data"</span>
path2<span class="token operator">=</span><span class="token string">"train"</span>
path<span class="token operator">=</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>path1<span class="token punctuation">,</span>path2<span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>path<span class="token punctuation">)</span>
  
<span class="token triple-quoted-string string">'''
[Run]
hymenoptera_data\hymenoptera_data\train
'''</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
<span class="token keyword">import</span> os
path1<span class="token operator">=</span><span class="token string">"hymenoptera_data\hymenoptera_data"</span>
path2<span class="token operator">=</span><span class="token string">"train"</span>
path<span class="token operator">=</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>path1<span class="token punctuation">,</span>path2<span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>path<span class="token punctuation">)</span>
  
<span class="token triple-quoted-string string">'''
[Run]
hymenoptera_data\hymenoptera_data\train
'''</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
import os
path1="hymenoptera_data\hymenoptera_data"
path2="train"
path=os.path.join(path1,path2)
print(path)
  
'''
[Run]
hymenoptera_data\hymenoptera_data\train
'''

• os.listdir(path)：将path中包含的图片名称变为一个列表。

  比如说，路径为”hymenoptera_data\hymenoptera_data\train\ants“的文件夹中有0013035.jpg、5650366_e22b7e1065.jpg、6240329_72c01e663e.jpg三张图片，那么此时有：

Copy
<span class="token keyword">import</span> os
path<span class="token operator">=</span><span class="token string">"hymenoptera_data\\hymenoptera_data\\train\\ants"</span>
path_list<span class="token operator">=</span>os<span class="token punctuation">.</span>listdir<span class="token punctuation">(</span>path<span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>path_list<span class="token punctuation">)</span>
  
<span class="token triple-quoted-string string">'''
[Run]
['0013035.jpg', '5650366_e22b7e1065.jpg', '6240329_72c01e663e.jpg']
'''</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
<span class="token keyword">import</span> os
path<span class="token operator">=</span><span class="token string">"hymenoptera_data\\hymenoptera_data\\train\\ants"</span>
path_list<span class="token operator">=</span>os<span class="token punctuation">.</span>listdir<span class="token punctuation">(</span>path<span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>path_list<span class="token punctuation">)</span>
  
<span class="token triple-quoted-string string">'''
[Run]
['0013035.jpg', '5650366_e22b7e1065.jpg', '6240329_72c01e663e.jpg']
'''</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
import os
path="hymenoptera_data\\hymenoptera_data\\train\\ants"
path_list=os.listdir(path)
print(path_list)
  
'''
[Run]
['0013035.jpg', '5650366_e22b7e1065.jpg', '6240329_72c01e663e.jpg']
'''

1. Dataset#

• 作用：提供一种方式去获取数据及其label

• 功能：

  • 如何获取每一个数据及其label

  • 告诉我们总共有多少数据（作用：神经网络要对同一个数据迭代多次，只有当我们知道总共有多少个数据，训练的时候我们才知道要训练多少次，才能把这个数据集迭代完然后进行下一次的迭代）

• 详解：

  • getitem：获取数据对应的label

  • len：返回数据的size

使用示例：

数据下载地址：百度网盘 请输入提取码 (baidu.com)

提取码：zsh8

（1）读取、简单处理图像数据

Copy
<span class="token keyword">from</span> torch<span class="token punctuation">.</span>utils<span class="token punctuation">.</span>data <span class="token keyword">import</span> Dataset
<span class="token keyword">from</span> PIL <span class="token keyword">import</span> Image
<span class="token keyword">import</span> os
<span class="token keyword">class</span> <span class="token class-name">MyData</span><span class="token punctuation">(</span>Dataset<span class="token punctuation">)</span><span class="token punctuation">:</span>
    <span class="token comment">#参数说明：</span>
    <span class="token comment">#root_dir:数据集的路径，如"E:\Desktop\hymenoptera_data\hymenoptera_data\train"(最好还是都加上\\，像其中的\train由于\t的存在会导致错误)</span>
    <span class="token comment">#label_dir:数据的标签，如"ants"</span>
    <span class="token keyword">def</span> <span class="token function">__init__</span><span class="token punctuation">(</span>self<span class="token punctuation">,</span>root_dir<span class="token punctuation">,</span>label_dir<span class="token punctuation">)</span><span class="token punctuation">:</span>
        self<span class="token punctuation">.</span>root_dir<span class="token operator">=</span>root_dir
        self<span class="token punctuation">.</span>label_dir<span class="token operator">=</span>label_dir
        <span class="token comment"># 使用os.path.join的方法是为了避免\和\\错误的问题</span>
        <span class="token comment"># self.path='E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train\\ants'</span>
        self<span class="token punctuation">.</span>path<span class="token operator">=</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>self<span class="token punctuation">.</span>root_dir<span class="token punctuation">,</span>self<span class="token punctuation">.</span>label_dir<span class="token punctuation">)</span>
        <span class="token comment">#将文件夹内所有图片的地址变成一个列表，并按先后顺序排列</span>
        self<span class="token punctuation">.</span>img_path<span class="token operator">=</span>os<span class="token punctuation">.</span>listdir<span class="token punctuation">(</span>self<span class="token punctuation">.</span>path<span class="token punctuation">)</span>
    <span class="token comment">#输入：图像对应的索引</span>
    <span class="token comment">#返回：idx索引对应的打开并标识过后的图像文件img；图像对应的标签label</span>
    <span class="token keyword">def</span> <span class="token function">__getitem__</span><span class="token punctuation">(</span>self<span class="token punctuation">,</span>idx<span class="token punctuation">)</span><span class="token punctuation">:</span>
        img_name<span class="token operator">=</span>self<span class="token punctuation">.</span>img_path<span class="token punctuation">[</span>idx<span class="token punctuation">]</span>    <span class="token comment">#根据索引idx，读取列表self.img_path中的图像名称</span>
        img_item_path<span class="token operator">=</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>self<span class="token punctuation">.</span>root_dir<span class="token punctuation">,</span>self<span class="token punctuation">.</span>label_dir<span class="token punctuation">,</span>img_name<span class="token punctuation">)</span>   <span class="token comment">#读取该索引对应图像的存储路径</span>
        img<span class="token operator">=</span>Image<span class="token punctuation">.</span><span class="token builtin">open</span><span class="token punctuation">(</span>img_item_path<span class="token punctuation">)</span>  <span class="token comment">#打开并标识给定的图像文件</span>
        label<span class="token operator">=</span>self<span class="token punctuation">.</span>label_dir    <span class="token comment">#label=数据标签,在该数据集中为ant或bee</span>
        <span class="token keyword">return</span> img<span class="token punctuation">,</span>label
    <span class="token comment">#返回图像数据集的长度，也就是说所读取的数据集中有多少张图片</span>
    <span class="token keyword">def</span> <span class="token function">__len__</span><span class="token punctuation">(</span>self<span class="token punctuation">)</span><span class="token punctuation">:</span>
        <span class="token keyword">return</span> <span class="token builtin">len</span><span class="token punctuation">(</span>self<span class="token punctuation">.</span>img_path<span class="token punctuation">)</span>
root_dir<span class="token operator">=</span><span class="token string">"E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train"</span>
ants_label_dir<span class="token operator">=</span><span class="token string">"ants"</span>
bees_label_dir<span class="token operator">=</span><span class="token string">"bees"</span>
ants_dataset<span class="token operator">=</span>MyData<span class="token punctuation">(</span>root_dir<span class="token punctuation">,</span>ants_label_dir<span class="token punctuation">)</span>
bees_dataset<span class="token operator">=</span>MyData<span class="token punctuation">(</span>root_dir<span class="token punctuation">,</span>ants_label_dir<span class="token punctuation">)</span>
img<span class="token punctuation">,</span>label<span class="token operator">=</span>ants_dataset<span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>img<span class="token punctuation">,</span>label<span class="token punctuation">)</span>   <span class="token comment">#根据索引，获取标识过后的图像img，以及对应的标签</span>
img<span class="token punctuation">.</span>show<span class="token punctuation">(</span><span class="token punctuation">)</span>         <span class="token comment">#显示对应索引的图像</span>
train_dataset<span class="token operator">=</span>ants_dataset<span class="token operator">+</span>bees_dataset    <span class="token comment">#将两个数据集进行拼接</span>
<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token builtin">len</span><span class="token punctuation">(</span>ants_dataset<span class="token punctuation">)</span><span class="token punctuation">,</span><span class="token builtin">len</span><span class="token punctuation">(</span>bees_dataset<span class="token punctuation">)</span><span class="token punctuation">,</span><span class="token builtin">len</span><span class="token punctuation">(</span>train_dataset<span class="token punctuation">)</span><span class="token punctuation">)</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
<span class="token keyword">from</span> torch<span class="token punctuation">.</span>utils<span class="token punctuation">.</span>data <span class="token keyword">import</span> Dataset
<span class="token keyword">from</span> PIL <span class="token keyword">import</span> Image
<span class="token keyword">import</span> os

<span class="token keyword">class</span> <span class="token class-name">MyData</span><span class="token punctuation">(</span>Dataset<span class="token punctuation">)</span><span class="token punctuation">:</span>
    <span class="token comment">#参数说明：</span>
    <span class="token comment">#root_dir:数据集的路径，如"E:\Desktop\hymenoptera_data\hymenoptera_data\train"(最好还是都加上\\，像其中的\train由于\t的存在会导致错误)</span>
    <span class="token comment">#label_dir:数据的标签，如"ants"</span>
    <span class="token keyword">def</span> <span class="token function">__init__</span><span class="token punctuation">(</span>self<span class="token punctuation">,</span>root_dir<span class="token punctuation">,</span>label_dir<span class="token punctuation">)</span><span class="token punctuation">:</span>
        self<span class="token punctuation">.</span>root_dir<span class="token operator">=</span>root_dir
        self<span class="token punctuation">.</span>label_dir<span class="token operator">=</span>label_dir

        <span class="token comment"># 使用os.path.join的方法是为了避免\和\\错误的问题</span>
        <span class="token comment"># self.path='E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train\\ants'</span>
        self<span class="token punctuation">.</span>path<span class="token operator">=</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>self<span class="token punctuation">.</span>root_dir<span class="token punctuation">,</span>self<span class="token punctuation">.</span>label_dir<span class="token punctuation">)</span>

        <span class="token comment">#将文件夹内所有图片的地址变成一个列表，并按先后顺序排列</span>
        self<span class="token punctuation">.</span>img_path<span class="token operator">=</span>os<span class="token punctuation">.</span>listdir<span class="token punctuation">(</span>self<span class="token punctuation">.</span>path<span class="token punctuation">)</span>

    <span class="token comment">#输入：图像对应的索引</span>
    <span class="token comment">#返回：idx索引对应的打开并标识过后的图像文件img；图像对应的标签label</span>
    <span class="token keyword">def</span> <span class="token function">__getitem__</span><span class="token punctuation">(</span>self<span class="token punctuation">,</span>idx<span class="token punctuation">)</span><span class="token punctuation">:</span>
        img_name<span class="token operator">=</span>self<span class="token punctuation">.</span>img_path<span class="token punctuation">[</span>idx<span class="token punctuation">]</span>    <span class="token comment">#根据索引idx，读取列表self.img_path中的图像名称</span>
        img_item_path<span class="token operator">=</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>self<span class="token punctuation">.</span>root_dir<span class="token punctuation">,</span>self<span class="token punctuation">.</span>label_dir<span class="token punctuation">,</span>img_name<span class="token punctuation">)</span>   <span class="token comment">#读取该索引对应图像的存储路径</span>
        img<span class="token operator">=</span>Image<span class="token punctuation">.</span><span class="token builtin">open</span><span class="token punctuation">(</span>img_item_path<span class="token punctuation">)</span>  <span class="token comment">#打开并标识给定的图像文件</span>
        label<span class="token operator">=</span>self<span class="token punctuation">.</span>label_dir    <span class="token comment">#label=数据标签,在该数据集中为ant或bee</span>
        <span class="token keyword">return</span> img<span class="token punctuation">,</span>label

    <span class="token comment">#返回图像数据集的长度，也就是说所读取的数据集中有多少张图片</span>
    <span class="token keyword">def</span> <span class="token function">__len__</span><span class="token punctuation">(</span>self<span class="token punctuation">)</span><span class="token punctuation">:</span>
        <span class="token keyword">return</span> <span class="token builtin">len</span><span class="token punctuation">(</span>self<span class="token punctuation">.</span>img_path<span class="token punctuation">)</span>

root_dir<span class="token operator">=</span><span class="token string">"E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train"</span>
ants_label_dir<span class="token operator">=</span><span class="token string">"ants"</span>
bees_label_dir<span class="token operator">=</span><span class="token string">"bees"</span>
ants_dataset<span class="token operator">=</span>MyData<span class="token punctuation">(</span>root_dir<span class="token punctuation">,</span>ants_label_dir<span class="token punctuation">)</span>
bees_dataset<span class="token operator">=</span>MyData<span class="token punctuation">(</span>root_dir<span class="token punctuation">,</span>ants_label_dir<span class="token punctuation">)</span>

img<span class="token punctuation">,</span>label<span class="token operator">=</span>ants_dataset<span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>img<span class="token punctuation">,</span>label<span class="token punctuation">)</span>   <span class="token comment">#根据索引，获取标识过后的图像img，以及对应的标签</span>
img<span class="token punctuation">.</span>show<span class="token punctuation">(</span><span class="token punctuation">)</span>         <span class="token comment">#显示对应索引的图像</span>

train_dataset<span class="token operator">=</span>ants_dataset<span class="token operator">+</span>bees_dataset    <span class="token comment">#将两个数据集进行拼接</span>
<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token builtin">len</span><span class="token punctuation">(</span>ants_dataset<span class="token punctuation">)</span><span class="token punctuation">,</span><span class="token builtin">len</span><span class="token punctuation">(</span>bees_dataset<span class="token punctuation">)</span><span class="token punctuation">,</span><span class="token builtin">len</span><span class="token punctuation">(</span>train_dataset<span class="token punctuation">)</span><span class="token punctuation">)</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
from torch.utils.data import Dataset
from PIL import Image
import os

class MyData(Dataset):
    #参数说明：
    #root_dir:数据集的路径，如"E:\Desktop\hymenoptera_data\hymenoptera_data\train"(最好还是都加上\\，像其中的\train由于\t的存在会导致错误)
    #label_dir:数据的标签，如"ants"
    def __init__(self,root_dir,label_dir):
        self.root_dir=root_dir
        self.label_dir=label_dir

        # 使用os.path.join的方法是为了避免\和\\错误的问题
        # self.path='E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train\\ants'
        self.path=os.path.join(self.root_dir,self.label_dir)

        #将文件夹内所有图片的地址变成一个列表，并按先后顺序排列
        self.img_path=os.listdir(self.path)

    #输入：图像对应的索引
    #返回：idx索引对应的打开并标识过后的图像文件img；图像对应的标签label
    def __getitem__(self,idx):
        img_name=self.img_path[idx]    #根据索引idx，读取列表self.img_path中的图像名称
        img_item_path=os.path.join(self.root_dir,self.label_dir,img_name)   #读取该索引对应图像的存储路径
        img=Image.open(img_item_path)  #打开并标识给定的图像文件
        label=self.label_dir    #label=数据标签,在该数据集中为ant或bee
        return img,label

    #返回图像数据集的长度，也就是说所读取的数据集中有多少张图片
    def __len__(self):
        return len(self.img_path)

root_dir="E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train"
ants_label_dir="ants"
bees_label_dir="bees"
ants_dataset=MyData(root_dir,ants_label_dir)
bees_dataset=MyData(root_dir,ants_label_dir)

img,label=ants_dataset[0]
print(img,label)   #根据索引，获取标识过后的图像img，以及对应的标签
img.show()         #显示对应索引的图像

train_dataset=ants_dataset+bees_dataset    #将两个数据集进行拼接
print(len(ants_dataset),len(bees_dataset),len(train_dataset))

（2）存储图像数据

首先新建一个文件夹，在该文件夹中：.txt文件名表示.jpg的图片名称，.txt文件中存储的数据为对应图像的label。用下面代码将文件名、label实现写入：

Copy
<span class="token keyword">import</span> os
root_dir<span class="token operator">=</span><span class="token string">"E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train"</span>
target_dir<span class="token operator">=</span><span class="token string">"ants"</span>
img_path<span class="token operator">=</span>os<span class="token punctuation">.</span>listdir<span class="token punctuation">(</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>root_dir<span class="token punctuation">,</span>target_dir<span class="token punctuation">)</span><span class="token punctuation">)</span>  <span class="token comment">#将target_dir文件夹中的图片名称存到img_path的列表中</span>
label<span class="token operator">=</span>target_dir  <span class="token comment">#标签为target_dir(根据实际情况读取标签)</span>
out_dir<span class="token operator">=</span><span class="token string">"ants_label"</span>   <span class="token comment">#输出图片的地址</span>
<span class="token keyword">for</span> i <span class="token keyword">in</span> img_path<span class="token punctuation">:</span>  <span class="token comment">#遍历每一张图片的名字</span>
    file_name<span class="token operator">=</span>i<span class="token punctuation">.</span>split<span class="token punctuation">(</span><span class="token string">".jpg"</span><span class="token punctuation">)</span><span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span>   <span class="token comment">#除去.jpg后缀，取出图片名字</span>
    <span class="token keyword">with</span> <span class="token builtin">open</span><span class="token punctuation">(</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>root_dir<span class="token punctuation">,</span>out_dir<span class="token punctuation">,</span><span class="token string">"{}.txt"</span><span class="token punctuation">.</span><span class="token builtin">format</span><span class="token punctuation">(</span>file_name<span class="token punctuation">)</span><span class="token punctuation">)</span><span class="token punctuation">,</span><span class="token string">'w'</span><span class="token punctuation">)</span> <span class="token keyword">as</span> f<span class="token punctuation">:</span>   <span class="token comment">#以图片名命名.txt文件</span>
        f<span class="token punctuation">.</span>write<span class="token punctuation">(</span>label<span class="token punctuation">)</span>  <span class="token comment">#将相应图片的标签，写入.txt文件中</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
<span class="token keyword">import</span> os

root_dir<span class="token operator">=</span><span class="token string">"E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train"</span>
target_dir<span class="token operator">=</span><span class="token string">"ants"</span>
img_path<span class="token operator">=</span>os<span class="token punctuation">.</span>listdir<span class="token punctuation">(</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>root_dir<span class="token punctuation">,</span>target_dir<span class="token punctuation">)</span><span class="token punctuation">)</span>  <span class="token comment">#将target_dir文件夹中的图片名称存到img_path的列表中</span>
label<span class="token operator">=</span>target_dir  <span class="token comment">#标签为target_dir(根据实际情况读取标签)</span>
out_dir<span class="token operator">=</span><span class="token string">"ants_label"</span>   <span class="token comment">#输出图片的地址</span>
<span class="token keyword">for</span> i <span class="token keyword">in</span> img_path<span class="token punctuation">:</span>  <span class="token comment">#遍历每一张图片的名字</span>
    file_name<span class="token operator">=</span>i<span class="token punctuation">.</span>split<span class="token punctuation">(</span><span class="token string">".jpg"</span><span class="token punctuation">)</span><span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span>   <span class="token comment">#除去.jpg后缀，取出图片名字</span>
    <span class="token keyword">with</span> <span class="token builtin">open</span><span class="token punctuation">(</span>os<span class="token punctuation">.</span>path<span class="token punctuation">.</span>join<span class="token punctuation">(</span>root_dir<span class="token punctuation">,</span>out_dir<span class="token punctuation">,</span><span class="token string">"{}.txt"</span><span class="token punctuation">.</span><span class="token builtin">format</span><span class="token punctuation">(</span>file_name<span class="token punctuation">)</span><span class="token punctuation">)</span><span class="token punctuation">,</span><span class="token string">'w'</span><span class="token punctuation">)</span> <span class="token keyword">as</span> f<span class="token punctuation">:</span>   <span class="token comment">#以图片名命名.txt文件</span>
        f<span class="token punctuation">.</span>write<span class="token punctuation">(</span>label<span class="token punctuation">)</span>  <span class="token comment">#将相应图片的标签，写入.txt文件中</span>
<span aria-hidden="true" class="line-numbers-rows"><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span><span></span></span>
import os

root_dir="E:\\Desktop\\hymenoptera_data\\hymenoptera_data\\train"
target_dir="ants"
img_path=os.listdir(os.path.join(root_dir,target_dir))  #将target_dir文件夹中的图片名称存到img_path的列表中
label=target_dir  #标签为target_dir(根据实际情况读取标签)
out_dir="ants_label"   #输出图片的地址
for i in img_path:  #遍历每一张图片的名字
    file_name=i.split(".jpg")[0]   #除去.jpg后缀，取出图片名字
    with open(os.path.join(root_dir,out_dir,"{}.txt".format(file_name)),'w') as f:   #以图片名命名.txt文件
        f.write(label)  #将相应图片的标签，写入.txt文件中

2. Dataloder#

• 作用：为后面的网络提供不同的数据形式（对其中几条数据进行打包）