传统弱校HFUT的蒟蒻,真相只有一个

python常用转换numpy和PIL互转,tensor和numpy互转

Deeplearning中常用转换速查

 

1、numpy和PIL互转

from PIL import Image
import numpy as np
import cv2
img = cv2.imread('image.jpg')
np.size(img,0) #0,1,2
print(type(img))

#numpy to PIL
pil_img= Image.fromarray(img)
print(type(pil_img))

#PIL to numpy
np_img= np.array(pil_img)
print(type(np_img))

2、tensor(pytorch)和numpy互转

import torch
import numpy as np

#tensor to numpy
torch_img = torch.ones(5)
print(torch_img)
torch_img+=1
np_img = torch_img.numpy()
print(torch_img, np_img)

#numpy to tensor
np_img = np.ones(9)
print(np_img)
torch_img = torch.from_numpy(np_img)
np_img+=1
print(np_img, torch_img)

 

 

 

 

#tensor to numpy
posted @ 2021-09-03 11:04  未名亚柳  阅读(1185)  评论(0编辑  收藏  举报