1

I'm plotting several metrics on a single Tensorboard scalar plot using writer.add_scalars(). If possible, I'd like to set some of these metrics to use a dashed line rather than a solid line to help in visually distinguishing some important subgroupings in the data. Is it possible to use a dashed line in Tensorboard for certain data? I can't seem to find anything on this. Ideally, I could pass a mapping from names to linestyles so that I can add everything in a single dictionary.

Edit: In case it's relevant, I'm using PyTorch.

Jigsaw
  • 405
  • 5
  • 11
  • Similar issue was described [here](https://stackoverflow.com/questions/37146614/tensorboard-plot-training-and-validation-losses-on-the-same-graph) and also find solution provided using [Keras](https://stackoverflow.com/questions/47877475/keras-tensorboard-plot-train-and-validation-scalars-in-a-same-figure/48393723#48393723), Let us know if this helps. Thanks –  Dec 17 '21 at 01:52
  • I might be missing something, but I see a solution for how to plot several metrics on a single graph at those links, but nothing about changing the linestyle (i.e., dashed or solid lines). The use case is that I have a few different validation sets I'm plotting metrics for. I'd like to have the training set metrics plotted using a solid line, and the metrics for each validation set plotted using dashed lines, to help visually distinguish the training and validation performance. Is using dashed lines supported by Tensorboard? – Jigsaw Dec 17 '21 at 02:03

0 Answers0