Description

FlynnOConnell

opened on May 14, 2025

Issue body actions

This does not work:

iw = fpl.ImageWidget(data=[data_raw, data_reg, mean_sub], names=["Raw", "Registered", "Reg, Mean-Sub"])
iw.show()

...
    158     if subplot_names.size != n_subplots:
--> 159         raise ValueError(
    160             f"must provide same number of subplot `names` as specified by shape, extents, or rects: {n_subplots}"

ValueError: must provide same number of subplot `names` as specified by shape, extents, or rects: 4

This is fixed by specifying figure shape:

iw = fpl.ImageWidget(data=[data_raw, data_reg, mean_sub], names=["Raw", "Registered", "Reg, Mean-Sub"], figure_shape=(1, 3))
iw.show()

This is confusing and took me a while to figure out

Activity

1. kushalkolar commented on May 14, 2025

   

   kushalkolar

   on May 14, 2025 via email

   Member

   More actions

   I ran into this for a minute yesterday too 😂. Any ideas for a generalizable solution?

   …

   On Wed, May 14, 2025, 11:27 Flynn ***@***.***> wrote: *FlynnOConnell* created an issue (fastplotlib/fastplotlib#818) <#818> This does not work: iw = fpl.ImageWidget(data=[data_raw, data_reg, mean_sub], names=["Raw", "Registered", "Reg, Mean-Sub"])iw.show() ... 158 if subplot_names.size != n_subplots:--> 159 raise ValueError( 160 f"must provide same number of subplot `names` as specified by shape, extents, or rects: {n_subplots}" ValueError: must provide same number of subplot `names` as specified by shape, extents, or rects: 4 This is fixed by specifying figure shape: iw = fpl.ImageWidget(data=[data_raw, data_reg, mean_sub], names=["Raw", "Registered", "Reg, Mean-Sub"], figure_shape=(1, 3))iw.show() This is confusing and took me a while to figure out — Reply to this email directly, view it on GitHub <#818>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACHXXRCSTUJGADNTBBMRYOL26NOHDAVCNFSM6AAAAAB5DWSNK2VHI2DSMVQWIX3LMV43ASLTON2WKOZTGA3DGNJRGY4TQNA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>
2. FlynnOConnell commented on May 14, 2025

   

   FlynnOConnell

   on May 14, 2025

   CollaboratorAuthor

   More actions

   I naively expect a single row when I don't specify a figure shape. A default that makes sense to me would be up to 4 or 5 plots in a row, any more than that, optimize to be a square. Something like this?

   n = n_subplots
   if n <= 5:
       rows, cols = 1, n
   else:
       cols = int(ceil(sqrt(n)))
       rows = int(ceil(n / cols))
3. kushalkolar commented on May 14, 2025

   

   kushalkolar

   on May 14, 2025

   Member

   More actions

   That gets very messy fast. I think a better way here would be to verify the number of data arrays and names matches, and to pad names with empties if there are blank subplots for the given shape. A user can still specify an arbitrary shape which has more subplots than they need, ex: (2, 3) to view 5 arrays.
4. clewis7

   closed this as completedin #828on May 20, 2025

Issue actions