How to use tensorboar in pytorch

1

so im using the pytorch-nighty and wanna see how to use tensorboard in pytorch
I am following steps here:
https://pytorch.org/docs/stable/tensorboard.html
everything works okay

when i reach to this line:
tensorboard --logdir=runs
I have no clue what should i do? how should i visualize?
if i just have that in my code it will give the error that:
SyntaxError: can't assign to operator
can anyone give me a suggestion? it should not be difficult I just cannot figure how to do the visualization

2

In your Python script, you would have to create a SummaryWriter and log some values/images/etc. using it as described in the code example in the docs.

In another terminal you then start TensorBoard using tensorboard --logdir=runs with additional arguments if needed.
This will create the tensorboard process, so that you can inspect the saved logs in your browser (default: localhost:6006).

3

You can check TensorboadX Documentation. It works perfectly and there are some examples to guide you through this.

4

I’m currently experimenting with the new implementation of Tensorboard in Torch.Utilis and so far it works fine.
The only problem I have is that I have to restart the process every time to update the scalar graphs (maybe also the other Graph-Types, but I did not test them).
I see, that Tensorboard detect new log files, but it do not show the data until i kill and restart the process.
Is there a possability to fix this, right now?