Plotting arguments

[0u812]

Some arguments of the old plotting function (ExtendedRoadRunner.plot) such as log had to be removed because they are not supported by both Plotly and Matplotlib. However, some arguments can be added back:

• grid
• linewidth
• color

These need to be tested in both Matplotlib and Plotly.

Also, for plotting names versus IDs, the following should be used: There should be a switch on the plot function which decides if

• id (default behavior), e.g., glc
• name, e.g., D-glucose
• both, e.g., glc (D-glucose) if name set, glc otherwise

[kirichoi]

Just to add on the list that might be worth considering:

• color
• loc

[0u812]

I think Plotly doesn't support loc, but color could be added provided that it is specified in a uniform way.

[0u812]

Actually it does seem to support legend locations: https://plot.ly/python/legend/. Still, needs to be specified in a uniform way for mpl vs Plotly.

[0u812]

It might be worth leaving the legend placement alone for now because you can have inset legends etc.

[kirichoi]

What we would like to do is to make sure that every single code in the document is functional. What I suggest is to accept loc and apply it on mpl but ignore it in Plotly if adding that feature is too time-consuming.

[0u812]

For the plotting functions, it would be nice to have a uniform interface that works regardless of whether the backend is mpl or plotly, so having parameters that only work in one is a limitation. I updated the examples from the docs to use the uniform interface.

[hsauro]

Agreed, make sure it works with code in the document.

H

On Fri, Sep 15, 2017 at 4:55 PM Kiri Choi notifications@github.com wrote:

What we would like to do is to make sure that every single code in the document is functional. What I suggest is to accept loc and apply it on mpl but ignore it in Plotly if adding that feature is too time-consuming.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/sys-bio/tellurium/issues/214#issuecomment-329928863, or mute the thread https://github.com/notifications/unsubscribe-auth/ABAZDlUWKjymAUDpMhfmEzGformQ0quzks5siw4zgaJpZM4PZn5G .

[matthiaskoenig]

The plot functionality should not be cut down because one of the plotting frameworks is not supporting some features. In such cases a warning should be provided that the respective feature is currently not supported by the given plotting engine. Especially with the additional plotting features coming in SED-ML L1V4 the full set of arguments must be transferred to the plotting engines. It is not a solution to cut down to the smallest common denominator.

[hsauro]

Agreed.

Herbert

On Sun, Oct 1, 2017 at 11:56 PM Matthias König notifications@github.com wrote:

The plot functionality should not be cut down because one of the plotting frameworks is not supporting some features. In such cases a warning should be provided that the respective feature is currently not supported by the given plotting engine. Especially with the additional plotting features coming in SED-ML L1V4 the full set of arguments must be transferred to the plotting engines. It is not a solution to cut down to the smallest common denominator.

— You are receiving this because you commented.

Reply to this email directly, view it on GitHub https://github.com/sys-bio/tellurium/issues/214#issuecomment-333454116, or mute the thread https://github.com/notifications/unsubscribe-auth/ABAZDsSxV07C-nY_34kEffPPkwUM6toQks5soIj5gaJpZM4PZn5G .