Pytorch学习2

博主： Xherlock
发布时间：2022 年 08 月 19 日
461 次浏览
暂无评论
6141字数
分类： pytorch

# Pytorch学习2

## transforms

重点：**看官方注释文档**

参考：https://www.cnblogs.com/zhangxiann/p/13570884.html

### ToTensor()

HWC&longrightarrow;CHW，将所有数除以255，将数据**归一化**到【0，1】

补充：**类的使用**

~~~python
class Person:
    def __call__(self, name):
        print('__call__' + ' Hello ' + name)

def hello(self, name):
        print('hello ' + name)

person = Person()
person("zhangsan")
person.hello('lisi')
~~~

输出：

~~~
__call__ Hello zhangsan
hellolisi
~~~

第一个person调用了\_\_call__

第二个person调用了Person类下的hello函数

---

有了上面这个类的使用例子就更容易理解ToTensor的使用方法：

类似于将一个**变量转换为函数**，下面将trans_totensor转换成了类函数来使用，实现将PIL图片形式转换为tensor形式

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

trans_totensor = transforms.ToTensor()
img_tensor = trans_totensor(img)
writer.add_image("Totensor", img_tensor)
writer.close()

~~~

### Normalize()

规范化到[-1, 1]：transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

print(img_tensor[0][0][0])
trans_norm = transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
img_norm = trans_norm(img_tensor)
print(img_norm[0][0][0])
writer.add_image("Normalize", img_norm)

writer.close()
~~~

输出

~~~
<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=768x512 at 0x229C91B9D90>
tensor(0.3137)
tensor(-0.3725)
~~~

![image-20220817161444083.png](http://xherlock.top/usr/uploads/2022/08/4025197193.png)

### Resize()

调整PIL Image对象的尺寸

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

print(img.size)
trans_resize = transforms.Resize((512, 512))
img_resize = trans_resize(img)
print(img_resize.size)
img_resize = trans_totensor(img_resize)

writer.close()
~~~

输出

~~~
<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=768x512 at 0x2C4637341C0>
(768, 512)
(512, 512)
~~~

![image-20220817162812637.png](http://xherlock.top/usr/uploads/2022/08/3561988169.png)

**补充**：

* 单个参数时：按图片长宽比缩放短边到参数
* 恢复原状时：w, h = img.size记录原来的宽和高，resize时需要反过来：transforms.Resize([h, w])
* resize输入的顺序是宽和长（高）

### Compose()

组合多个步骤：eg 先缩放图片（Resize）再转换为tensor形式（ToTensor）

参数为一个列表，数据类型为transforms，形式为[transforms参数1, transforms参数2, ……]

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

# ToTensor
trans_totensor = transforms.ToTensor()
img_tensor = trans_totensor(img)
writer.add_image("Totensor", img_tensor)

# Resize
print(img.size)
trans_resize = transforms.Resize((100, 512))
img_resize = trans_resize(img)
print(img_resize.size)
img_resize = trans_totensor(img_resize)
writer.add_image("Resize", img_resize, 0)

# Compose
trans_resize_2 = transforms.Resize(500)
trans_compose = transforms.Compose([trans_resize_2, trans_totensor])
img_resize_2 = trans_compose(img)
writer.add_image("Resize", img_resize_2, 1)

writer.close()
~~~

### RandomCrop

在一个随机的位置进行裁剪，一个参数表示长宽一样

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

trans_totensor = transforms.ToTensor()
img_tensor = trans_totensor(img)
writer.add_image("Totensor", img_tensor)

# RandomCrop
trans_random = transforms.RandomCrop(512)
trans_compose = transforms.Compose([trans_random, trans_totensor])
for i in range(10):
    img_crop = trans_compose(img)
    writer.add_image("Random", img_crop, i)

writer.close()
~~~

## torchvision数据集的使用

~~~python
import torchvision
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

dataset_transform = torchvision.transforms.Compose([
    torchvision.transforms.ToTensor()
])

# 若根目录不存在数据集，则下载数据集；transforms参数进行transforms处理
train_set = torchvision.datasets.CIFAR10(root='./dataset', train=True, download=True, transform=dataset_transform)
test_set = torchvision.datasets.CIFAR10(root='./dataset', train=True, download=True, transform=dataset_transform)

writer = SummaryWriter("p10")
for i in range(10):
    img, target = test_set[i]
    print(target)
    writer.add_image("test_set", img, i)

writer.close()
~~~

## Dataloader的使用

将自定义的Dataset根据batch_size大小、是否shuffle等封装成一个Batch Size大小的Tensor，用于后面的训练。

参数

- **dataset** ( [*Dataset*](https://pytorch.org/docs/stable/data.html#torch.utils.data.Dataset) ) – 从中加载数据的数据集。
- **batch_size** ( [*int*](https://docs.python.org/3/library/functions.html#int) *,* *optional* ) – 每批要加载多少样本（默认值：`1`）。
- **shuffle** ( [*bool*](https://docs.python.org/3/library/functions.html#bool) *,* *optional* ) – 设置为`True`在每个 epoch 重新洗牌数据（默认值：）`False`。（True相当于随机打乱）
- **num_workers** ( [*int*](https://docs.python.org/3/library/functions.html#int) *,* *optional* ) – 用于数据加载的子进程数。`0`表示数据将在主进程中加载。（默认`0`：）
- **drop_last** ( [*bool*](https://docs.python.org/3/library/functions.html#bool) *,* *optional* ) –`True`如果数据集大小不能被批次大小整除，则设置为丢弃最后一个不完整的批次。如果`False`数据集的大小不能被批大小整除，那么最后一批将更小。（默认`False`：）

~~~python
import torchvision
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

# 准备数据集
test_data = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor())

test_loader = DataLoader(dataset=test_data, batch_size=4, shuffle=False, num_workers=0, drop_last=True)

writer = SummaryWriter("dataloader")

for epoch in range(2):
    step = 0
    for data in test_loader:
        imgs, targets = data
        writer.add_images("Epoch: {}".format(epoch), imgs, step)
        step += 1

writer.close()
~~~

最后修改：2022 年 08 月 19 日

如果觉得我的文章对你有用，请随意赞赏

发表评论取消回复

评论 *

私密评论

名称 *

🎲

邮箱 *

地址

Reverse（十一）
浏览次数: 2629
Win11新机鼓捣记录
浏览次数: 2382
栈~数制转换
浏览次数: 1604
C++学习
浏览次数: 1603
Python实现文件管理系统（重命名和删除）
浏览次数: 1575

顶点软件
你写得非常清晰明了，让我很容易理解你的观点。
mars
博主你好，请问怎么知道的alloc函数属于Buffer下的方法...
lj
该评论仅登录用户及评论双方可见
Mr.Guan
:@(赞一个)
Typecho
欢迎加入 Typecho 大家族

CSS1
浏览次数: 529
pytorch学习4
浏览次数: 479
修改搜索引擎已访问的超链接颜色
浏览次数: 580
Vue（三）
浏览次数: 490
Pytorch学习1
浏览次数: 506

Pytorch学习2

Xherlock • 2022 年 08 月 19 日

# Pytorch学习2

## transforms

重点：**看官方注释文档**

参考：https://www.cnblogs.com/zhangxiann/p/13570884.html

### ToTensor()

HWC&longrightarrow;CHW，将所有数除以255，将数据**归一化**到【0，1】

补充：**类的使用**

~~~python
class Person:
    def __call__(self, name):
        print('__call__' + ' Hello ' + name)

def hello(self, name):
        print('hello ' + name)

person = Person()
person("zhangsan")
person.hello('lisi')
~~~

输出：

~~~
__call__ Hello zhangsan
hellolisi
~~~

第一个person调用了\_\_call__

第二个person调用了Person类下的hello函数

---

有了上面这个类的使用例子就更容易理解ToTensor的使用方法：

类似于将一个**变量转换为函数**，下面将trans_totensor转换成了类函数来使用，实现将PIL图片形式转换为tensor形式

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

trans_totensor = transforms.ToTensor()
img_tensor = trans_totensor(img)
writer.add_image("Totensor", img_tensor)
writer.close()

~~~

### Normalize()

规范化到[-1, 1]：transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

print(img_tensor[0][0][0])
trans_norm = transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
img_norm = trans_norm(img_tensor)
print(img_norm[0][0][0])
writer.add_image("Normalize", img_norm)

writer.close()
~~~

输出

~~~
<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=768x512 at 0x229C91B9D90>
tensor(0.3137)
tensor(-0.3725)
~~~

![image-20220817161444083.png](http://xherlock.top/usr/uploads/2022/08/4025197193.png)

### Resize()

调整PIL Image对象的尺寸

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

print(img.size)
trans_resize = transforms.Resize((512, 512))
img_resize = trans_resize(img)
print(img_resize.size)
img_resize = trans_totensor(img_resize)

writer.close()
~~~

输出

~~~
<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=768x512 at 0x2C4637341C0>
(768, 512)
(512, 512)
~~~

![image-20220817162812637.png](http://xherlock.top/usr/uploads/2022/08/3561988169.png)

**补充**：

### Compose()

组合多个步骤：eg 先缩放图片（Resize）再转换为tensor形式（ToTensor）

参数为一个列表，数据类型为transforms，形式为[transforms参数1, transforms参数2, ……]

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

# ToTensor
trans_totensor = transforms.ToTensor()
img_tensor = trans_totensor(img)
writer.add_image("Totensor", img_tensor)

# Compose
trans_resize_2 = transforms.Resize(500)
trans_compose = transforms.Compose([trans_resize_2, trans_totensor])
img_resize_2 = trans_compose(img)
writer.add_image("Resize", img_resize_2, 1)

writer.close()
~~~

### RandomCrop

在一个随机的位置进行裁剪，一个参数表示长宽一样

~~~python
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

writer = SummaryWriter("logs")
img = Image.open("hymenoptera_data/train/ants/0013035.jpg")
print(img)

trans_totensor = transforms.ToTensor()
img_tensor = trans_totensor(img)
writer.add_image("Totensor", img_tensor)

writer.close()
~~~

## torchvision数据集的使用

~~~python
import torchvision
from PIL import Image
from torchvision import transforms
from torch.utils.tensorboard import SummaryWriter

dataset_transform = torchvision.transforms.Compose([
    torchvision.transforms.ToTensor()
])

writer = SummaryWriter("p10")
for i in range(10):
    img, target = test_set[i]
    print(target)
    writer.add_image("test_set", img, i)

writer.close()
~~~

## Dataloader的使用

将自定义的Dataset根据batch_size大小、是否shuffle等封装成一个Batch Size大小的Tensor，用于后面的训练。

参数

~~~python
import torchvision
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

# 准备数据集
test_data = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor())

test_loader = DataLoader(dataset=test_data, batch_size=4, shuffle=False, num_workers=0, drop_last=True)

writer = SummaryWriter("dataloader")

for epoch in range(2):
    step = 0
    for data in test_loader:
        imgs, targets = data
        writer.add_images("Epoch: {}".format(epoch), imgs, step)
        step += 1

writer.close()
~~~

发表评论 取消回复

Pytorch学习2

发表评论取消回复