vmax and ticks

[w-yuxuan]

Dear Dr. Tsirkin,
Thank you for building the amazing bandup package!

When the vmax is less than 1, I noticed that the color bar doesn't show the upper range like attached. Is there a setting for it to show fraction like 0.4? I tried adding: --cb_label --round_cb 2 --disable_auto_round_vmin_and_vmax

Can I make the code display the greek letter gamma? I notice the issue only when I have discontinuities like U,K.
I tried // Gamma too. Thank you so much!

[stepan-tsirkin]

Dear @w-yuxuan

First of all, BandUP was developed by @paulovcmedeiros , I just collaborated. Unfortunately, Paulo does not actively maintain the project anymore. Later, I developed a similar tool (but fully in python) banduppy : https://github.com/band-unfolding/banduppy/tree/master/banduppy Could you try that, and say if it works for you? I am able to provide a better support for banduppy.

About v-ticks in colorbar, looks like it is hard-coded to be integers:

bandup/src/python_interface/bandupy/plot.py

Lines 735 to 737 in bd3ff1f

	cb_yticks = np.arange(int(image.norm.vmin), int(image.norm.vmax) + 1, 1)
	cb_ytick_labels = [round(item,abs(args.round_cb)) for item in cb_yticks]

So, you will need to modify line 735 to your taste, smth like

cb_yticks = np.linspace(image.norm.vmin, image.norm.vmax + 1, 5) 

Regarding Greek letters, there should not be a problem with matplotlib : https://matplotlib.org/stable/users/explain/text/mathtext.html Did you try to give it as "$\Gamma$"

Please, tell me if this works.

Note: banduppy also has a plotting routine, but much simpler (less options, but easier to understand and modify)

https://github.com/band-unfolding/banduppy/blob/160a60257201c708a6f68809ba13c2a8dd64c34e/banduppy/unfolding_path.py#L165-L222

[w-yuxuan]

thank you so so much for pointing me to the specific line!! Unfortunately it didn't work but "!G " worked. The vmax ticks line worked very well and this is what I ended up to make it even better: cb_yticks = np.linspace(image.norm.vmin, image.norm.vmax, 5, endpoint=True).
Thanks for your quick support again!!