The best (in my opinion) way of solving this problem is by going around it and using create_profile function. I really like this method because it offers great customisability

Here is a snippet of my code that illustrates this:

fig, ax = plt.subplots(2, 3, figsize=(8, 5), sharex='all', sharey='all')

for i in range(2):

for j in range(3):

ad = ds.all_data() #here I normally go over selecting the required data

bins_dense = np.logspace(-29, -21, num=80)

bins_vz = np.linspace(-1000, 1000, num=80)

logs = dict(velocity_cylindrical_z=False, density=True)

units = dict(velocity_cylindrical_z='km/s')

profile = yt.create_profile(gal_dense, ['density', 'velocity_cylindrical_z'], fields='cell_mass', weight_field=None, units=units, logs=logs, override_bins={'density': bins_dense, 'velocity_cylindrical_z': bins_vz})

prof = profile['cell_mass']

im = ax[i, j].pcolormesh(profile.x, profile.y, prof.d.T, cmap='arbre', norm = matplotlib.colors.LogNorm(vmin=1e0, vmax=1e4))

ax[i, j].set_xscale('log')

ax[i, j].set_xlim(1e-29, 1e-21)

ax[i, j].set_ylim(-1000, 1000)

Here I want to plot density on x-axis and velocity_cylindrical_z on y-axis, colour-coded by cell_mass

You can select whether you want axes to be log or linear, bin size and limits and the units (used in the bins)

Then you select the data in prof = profile['cell_mass’] and just plot it using regular subplots so you can set aspect playing with figsize or ax[i].set_aspect('equal’) or fig.subplots_adjust()

I use this method extensively so don’t hesitate to contact me if you have any questions :)

On 25 Jul 2024, at 13:27, HyeonYong Kim <gusdyd398@snu.ac.kr> wrote:

Dear YT user,

I'm trying to draw a multipanel Phaseplot in yt. There is no problem with drawing the plot, but the problem is the size of figure.

As shown in my figure (https://drive.google.com/file/d/1IM_slIts1lgtQ95m-sKHf7EncIM2ZvoO/view?usp=drive_link) or yt document here (https://yt-project.org/doc/cookbook/complex_plots.html#multipanel-with-phaseplot), the images are compressed, not the square shape.

I tried fig = plt.figure(figsize = (2,8)) but it doesn't work and the set_figure_size function is only for the size of the figure on the longest axis. I've searched the YT document, it seems that there is no function that adjusts the width or height of the figure.

Each image comes out as a square if I draw the multipanel with Projectionplot, it doesn't work with Phaseplot.
Is there any way to handle the size of the figures or axesgrid in YT?

Thanks in advance.
Hyeonyong
_______________________________________________
yt-users mailing list -- yt-users@python.org
To unsubscribe send an email to yt-users-leave@python.org
https://mail.python.org/mailman3/lists/yt-users.python.org/
Member address: akermannia@gmail.com