Time-Series Segmentation Masks


I’m wondering whether there is an equivalent feature of Image Masks for 1-D time-series plots. For error analysis, I would like to compare prediction/ground-truth masks “over top” time-series signals that are logged with plotly html. I’m curious what the best way to log this information with wandb would be.


Hi Tom,

Unfortunately, there is no way of doing this directly in wandb. In this case, you could have 2 different paths.

  • You could either plot everything in plotly yourself first and then upload it to wandb using the Plotly integration.
  • Or you could log the data to wandb and utilize our custom charts. Custom charts are based on VegaLite. They do have multiple examples on their website here that you might find useful and use as a starting point for your Vega charts.

Otherwise, unfortunately, there isn’t a way of accomplishing it natively with preset wandb charts.


1 Like

Hi Tom,

We wanted to follow up with you regarding your support request as we have not heard back from you. Please let us know if we can be of further assistance or if your issue has been resolved.

Weights & Biases

Hi Tom, since we have not heard back from you we are going to close this request. If you would like to re-open the conversation, please feel free to create another thread and refer to this one in the new one!

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.