【土堆pytorch实战】P7-8 tensorboard使用

P7 tensorboard使用

1. 画出简单函数图

import numpy as np
from PIL import Image
from torch.utils.tensorboard import SummaryWriter
writer=SummaryWriter("logs")

for i in range(100): #画出y=2x的图像
    writer.add_scalar("y=2x",2*i,i)  # tag 图的标题 y轴 x轴
writer.close()

terminal中输入tensorboard --logdir=logs
如下图示，点击蓝色6006地址即可查看

2. 查看图片

import torch
import numpy as np
from PIL import Image
from torch.utils.tensorboard import SummaryWriter
writer=SummaryWriter("logs")

image_path="./hymenoptera_data/train/ants_image/6240329_72c01e663e.jpg"
img_PIL=Image.open(image_path)
img_array=np.array(img_PIL)
print(type(img_array))  #numpy 类型
print(img_array.shape)

writer.add_image("test",img_array,1,dataformats="HWC")
writer.close()

add_image中确保图像类型是numpy或者tensor类型