使用 Tensorboard 在一张图中绘制多个图 [英] Plot multiple graphs in one plot using Tensorboard

问题描述

我正在使用带有 Tensorflow 后端的 Keras.我的工作涉及在我的数据集上比较多个模型(如 Inception、VGG、Resnet 等)的性能.我想在一张图中绘制多个模型的训练精度.我正在尝试在 Tensorboard 中执行此操作，但它不起作用.

I am using Keras with Tensorflow backend. My work involves comparing the performances of several models such as Inception, VGG, Resnet etc on my dataset. I would like to plot the training accuracies of several models in one graph. I am trying to do this in Tensorboard, but it is not working.

有没有办法使用 Tensorboard 在一个图中绘制多个图形，或者有其他方法可以做到这一点?

Is there a way of plotting multiple graphs in one plot using Tensorboard or is there some other way I can do this?

谢谢

推荐答案

如果您使用的是 SummaryWriter 来自 tensorboardX 或 pytorch 1.2，您有一个名为 add_scalars:

If you are using the SummaryWriter from tensorboardX or pytorch 1.2, you have a method called add_scalars:

这样称呼它:

my_summary_writer.add_scalars(f'loss/check_info', {
    'score': score[iteration],
    'score_nf': score_nf[iteration],
}, iteration)

它会像这样显示:

小心 add_scalars 会干扰你的运行组织:它会向这个列表添加多个条目(从而造成混乱):

Be careful that add_scalars will mess with the organisation of your runs: it will add mutliple entries to this list (and thus create confusion):

我建议您这样做:

my_summary_writer.add_scalar(f'check_info/score',    score[iter],    iter)
my_summary_writer.add_scalar(f'check_info/score_nf', score_nf[iter], iter)

这篇关于使用 Tensorboard 在一张图中绘制多个图的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！