Custom plot based on a default one

Hi,
I’m finding really difficult to follow the documentation about custom plots. I am doing reinforcement learning and using the default plots that group the runs, to compare the rewards, etc. What would be the best way to replicate a default plot and change something about it? In fact, I just want to add a horizontal line (my baseline) but doing it by adding an calculated expression removes the default information.
Many thanks!

Hi @jorgecsilva, you could add something like the following to your Vega spec:

"mark": "rule",
      "encoding": {
        "y": {"value": 10},
        "size": {"value": 2},
        "color": {"field": "symbol"}

In this example 10 would be the Y-value for the line.

If your question is more around how the custom chart query works I’d be happy to go into that more as well. One thing I want to note is that our default charts are not written in Vega so there is no 1:1 way to just copy the code into a custom chart and using the examples on the Vega website is probably the best way to get started creating your own grouped charts.
Thank you,
Nate

Hi @jorgecsilva, I just wanted to follow up and see if you were still looking for help with this or if the above answered your question?

Thank you,
Nate

Hi Nate

Many thanks for your help, and sorry for my late answer! By now, my approach is to use the api to download the results and plot with matplotlib locally. The only reason for that is not being able to replicate a default plot and just add a horizontal line. So, it would be really great if the documentation could cover this kind scenario. To familiarize myself with the vega syntax would, I guess, take longer than what I’m doing now.
Thanks again!
Jorge

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