tensorboard学习日记：add_image的使用
from torch.utils.tensorboard import SummaryWriter
#利用numpy.array(),对PIL图片进行转换
import numpy as ny
from PIL import Image
#如果想要查看这个类，ctrl+鼠标点击
#
#这里使用SummaryWriter创建一个tensorboard文件
writer = SummaryWriter("logs")#这个咱们只需要给他一个文件的名称 即folder location
image_path = "D:\PyCharm Community Edition 2021.3.1\\tensorboard\data\\train\\ants_image\\0013035.jpg"
img_PIL = Image.open(image_path)#创建PIL的图片类
# print(type(img_PIL))在控制台查询类型
image_array = ny.array(img_PIL) #转成也就是add_image img_tensor所需要的参数类型
# 开始向add_image添加参数
writer.add_image("test",image_array,1)
# """Add image data to summary.
#
# Note that this requires the ``pillow`` package.
#
# Args:
# tag (string): Data identifier 起个名字
# img_tensor (torch.Tensor, numpy.array, or string/blobname): Image data 添加图片，本次采取的是numpy.array的类型
# global_step (int): Global step value to record
#
# 在经过以上的将PIL的图片转成numpy.array之后我们便满足了add_image函数的参数类型要求，但此时运行仍然不行，原因出在add_image使用numpy.array类型时需要添加说明：
def add_image(self, tag, img_tensor, global_step=None, walltime=None, dataformats='CHW'):
"""Add image data to summary.
Note that this requires the ``pillow`` package.
Args:
tag (string): Data identifier
img_tensor (torch.Tensor, numpy.array, or string/blobname): Image data
global_step (int): Global step value to record
walltime (float): Optional override default walltime (time.time())
seconds after epoch of event
Shape:
img_tensor: Default is :math:`(3, H, W)`. You can use ``torchvision.utils.make_grid()`` to
convert a batch of tensor into 3xHxW format or call ``add_images`` and let us do the job.
Tensor with :math:`(1, H, W)`, :math:`(H, W)`, :math:`(H, W, 3)` is also suitable as long as
corresponding ``dataformats`` argument is passed, e.g. ``CHW``, ``HWC``, ``HW``.意思便是img_tensor的形状有要求，而默认格式是（3，H，W）即通道(channel)为3，H为高度，W为宽度
我们在控制台显示其形状，并不是默认格式，所以需要添加额外说明
Examples::
from torch.utils.tensorboard import SummaryWriter
import numpy as np
img = np.zeros((3, 100, 100))
img[0] = np.arange(0, 10000).reshape(100, 100) / 10000
img[1] = 1 - np.arange(0, 10000).reshape(100, 100) / 10000
img_HWC = np.zeros((100, 100, 3))
img_HWC[:, :, 0] = np.arange(0, 10000).reshape(100, 100) / 10000
img_HWC[:, :, 1] = 1 - np.arange(0, 10000).reshape(100, 100) / 10000
writer = SummaryWriter()
writer.add_image('my_image', img, 0)
# If you have non-default dimension setting, set the dataformats argument.
writer.add_image('my_image_HWC', img_HWC, 0, dataformats='HWC')
writer.close()此时说明文档也给了我们例子，因此我们在写的时候需要在后面添加 dataformats='HWC'
from torch.utils.tensorboard import SummaryWriter
#利用numpy.array(),对PIL图片进行转换
import numpy as ny
from PIL import Image
#如果想要查看这个类，ctrl+鼠标点击
#
#这里使用SummaryWriter创建一个tensorboard文件
writer = SummaryWriter("logs")#这个咱们只需要给他一个文件的名称 即folder location
image_path = "D:\PyCharm Community Edition 2021.3.1\\tensorboard\data\\train\\ants_image\\0013035.jpg"
img_PIL = Image.open(image_path)#创建PIL的图片类
# print(type(img_PIL))在控制台查询类型
image_array = ny.array(img_PIL) #转成也就是add_image img_tensor所需要的参数类型
# 开始向add_image添加参数
writer.add_image("test",image_array,1,dataformats='HWC')
# """Add image data to summary.
#
# Note that this requires the ``pillow`` package.
#
# Args:
# tag (string): Data identifier 起个名字
# img_tensor (torch.Tensor, numpy.array, or string/blobname): Image data 添加图片，本次采取的是numpy.array的类型
# global_step (int): Global step value to record
#
#
for i in range(100):
writer.add_scalar("y=x",i,i)
# #add。scalar作用为将值存入到tensorboard中
# """Add scalar data to summary.简单来说就是添加数据
#
# Args:
# tag (string): Data identifier数据标识符 简单来说就是咱给其结果图取一个名字
# scalar_value (float or string/blobname): Value to save 传入的数值（也就是结果图的y轴值）
# global_step (int): Global step value to record 也就是结果图的x轴记录训练的步数
#例如writer.add("train_loss",loss,(epoch*epoch_size+iteration)) 分别表示给其存入到tensorboard取一个名称/存入的值变量/以及tensorboard的横坐标
writer.close()
#range的使用
# for i in range(1, 4):
# print(i)
# #设置停顿三秒
# time.sleep(3)
# #代表1到10，间隔2,不包含10
# for i in range(1, 10, 2):
# print(i)
# time.sleep(3)
# #代表0到5，不包含5
# for i in range(5):
# print(i)
运行代码之后打开tensorboard
添加第二张图片，
我们改变图片的地址和步数step=2，再次运行代码
