PyTorch Tensorboard中的投影仪选项卡为空

PyTorch Tensorboard中的投影仪选项卡为空,pytorch,tensorboard,Pytorch,Tensorboard,我试图将每个历元的嵌入可视化。但投影仪页面显示了这一点- 这是这一页上的文字- No checkpoint was found. Probable causes: No checkpoint has been saved yet. Please refresh the page periodically. You are not saving any checkpoint. To save your model, create a tf.train.Saver and sa

我试图将每个历元的嵌入可视化。但投影仪页面显示了这一点-

这是这一页上的文字-

No checkpoint was found.

Probable causes:

    No checkpoint has been saved yet. Please refresh the page periodically.
    You are not saving any checkpoint. To save your model, create a tf.train.Saver and save your model periodically by calling saver.save(session, LOG_DIR/model.ckpt, step).

If you’re new to using TensorBoard, and want to find out how to add data and set up your event files, check out the README and perhaps the TensorBoard tutorial.

If you think TensorBoard is configured properly, please see the section of the README devoted to missing data problems and consider filing an issue on GitHub.
为了简化事情,我使用以下方法在我的训练循环中生成一个嵌入;还是不行

for e in epochs:
    .
    .
    writer.add_embedding(torch.randn(100, 5), global_step=e)

我在这里遗漏了什么?

在遇到同样的错误后,我尝试运行一个用于嵌入的示例脚本,但突然它成功了

我把范围缩小到:在添加嵌入之前或之后向Tensorboard添加任何图像都可以使其工作。像这样:

writer.add_image('image',np.one((3,3,3)),0)


不确定原因。

您是否尝试重新启动张力板?您的浏览器控制台是否出现错误?是的,我已尝试重新启动控制台。甚至删除了
/runs
目录。我已经用浏览器控制台的屏幕截图更新了这个问题。请检查一下。作为参考,图形/标量/直方图/图像都可以使用。只有嵌入没有。我试过了。不幸的是,这对我来说不起作用。
import torch;输入numpy作为np;从torch.utils.tensorboard导入SummaryWriter;进口张力板;输入张量流;tensorflow.io.gfile=tensorboard.compat.tensorflow_stub.io.gfile;writer=SummaryWriter();writer.add_image('image',np.one((3,3,3)),0);writer.add_嵌入(torch.randn(100,5),global_step=0)
重新启动tensorboard后,尝试运行此MWE并选择投影仪