Line plot for bands doesn't work

[MRedies]

I am trying to plot a bandstructure with lines instead of dots. I tried to use the option line_plot=True documented here:

https://masci-tools.readthedocs.io/en/latest/module_guide/code.html?highlight=plot_bands#masci_tools.vis.plot_methods.plot_bands

I did it in this jupyter notebook, but I'll try to copy it a swell
line_plot.zip

from masci_tools.io.parsers.hdf5 import HDF5Reader
from masci_tools.io.parsers.hdf5.recipes import FleurBands
from masci_tools.vis.fleur import plot_fleur_bands

#Read in data
with HDF5Reader('banddos.hdf') as h5reader:
   data, attributes = h5reader.read(recipe=FleurBands)



#Plot the data
#Notice that you get the axis object of this plot is returned
#if you want to make any special additions
ax = plot_fleur_bands(data, attributes, limits={'y': (-3,5)}, line_plot=True)

results in

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-25-afd0580a8c3a> in <module>
     12 #Notice that you get the axis object of this plot is returned
     13 #if you want to make any special additions
---> 14 ax = plot_fleur_bands(data, attributes, limits={'y': (-3,5)}, line_plot=True)

~/.local/lib/python3.8/site-packages/masci_tools/vis/fleur.py in plot_fleur_bands(bandsdata, bandsattributes, spinpol, only_spin, bokeh_plot, weight, **kwargs)
    276     else:
    277         if spinpol:
--> 278             fig = plot_spinpol_bands(bandsdata['kpath'],
    279                                      bandsdata['eigenvalues_up'],
    280                                      bandsdata['eigenvalues_down'],

~/.local/lib/python3.8/site-packages/masci_tools/vis/parameters.py in ensure_consistency(*args, **kwargs)
     77 
     78             try:
---> 79                 res = func(*args, **kwargs)
     80             except Exception:
     81                 plotter_object.remove_added_parameters()

~/.local/lib/python3.8/site-packages/masci_tools/vis/plot_methods.py in plot_spinpol_bands(kpath, bands_up, bands_dn, size_data, show_spin_pol, special_kpoints, e_fermi, xlabel, ylabel, title, saveas, markersize_min, markersize_scaling, scale_color, **kwargs)
   1965         plot_params.set_defaults(default_type='function', sub_colormap=(0.15, 1.0))
   1966 
-> 1967     ax = multi_scatter_plot([kpath, kpath], [bands_up, bands_dn],
   1968                             size_data=size_data,
   1969                             xlabel=xlabel,

~/.local/lib/python3.8/site-packages/masci_tools/vis/parameters.py in ensure_consistency(*args, **kwargs)
     77 
     78             try:
---> 79                 res = func(*args, **kwargs)
     80             except Exception:
     81                 plotter_object.remove_added_parameters()

~/.local/lib/python3.8/site-packages/masci_tools/vis/plot_methods.py in multi_scatter_plot(xdata, ydata, size_data, color_data, xlabel, ylabel, title, saveas, axis, **kwargs)
    476             plot_kw.pop('color')
    477 
--> 478         res = ax.scatter(x, y=y, s=size, c=color, **plot_kw, **kwargs)
    479         if plot_kw.get('label', None) is not None and color is not None:
    480             if isinstance(color, (list, np.ndarray, pd.Series)):

/usr/local/anaconda3/lib/python3.8/site-packages/matplotlib/__init__.py in inner(ax, data, *args, **kwargs)
   1359     def inner(ax, *args, data=None, **kwargs):
   1360         if data is None:
-> 1361             return func(ax, *map(sanitize_sequence, args), **kwargs)
   1362 
   1363         bound = new_sig.bind(ax, *args, **kwargs)

/usr/local/anaconda3/lib/python3.8/site-packages/matplotlib/axes/_axes.py in scatter(self, x, y, s, c, marker, cmap, norm, vmin, vmax, alpha, linewidths, edgecolors, plotnonfinite, **kwargs)
   4595                 )
   4596         collection.set_transform(mtransforms.IdentityTransform())
-> 4597         collection.update(kwargs)
   4598 
   4599         if colors is None:

/usr/local/anaconda3/lib/python3.8/site-packages/matplotlib/artist.py in update(self, props)
   1060                     func = getattr(self, f"set_{k}", None)
   1061                     if not callable(func):
-> 1062                         raise AttributeError(f"{type(self).__name__!r} object "
   1063                                              f"has no property {k!r}")
   1064                     ret.append(func(v))

AttributeError: 'PathCollection' object has no property 'line_plot'

[janssenhenning]

Oh that's unfortunate. The default documentation in readthedocs points to the docs of the develop branch. We should probably change that.
You had unlucky timing, since I only merged this feature today :D
If you wanna use it now you'll have to use the newest develop branch

[MRedies]

I did a python3 -m pip install --upgrade masci-tools, but this doesn't seem to get me to the development version.

Can I get the development version through pip?

[janssenhenning]

No there's no pre-release for this feature yet, which you could install through pip.

You'll have to clone the git repository, checkout the develop branch and install it with calling pip install -e . in the repository

[MRedies]

Are you sure this feature works correctly? I pulled newest version from github, where the tests pass:

Requirement already satisfied: zipp>=0.5 in /Users/alzeer/.local/lib/python3.6/site-packages (from importlib-metadata; python_version < "3.8"->SQLAlchemy<2.0.0,>=1.3.23->mendeleev->masci-tools==0.5.0)
Installing collected packages: masci-tools
  Found existing installation: masci-tools 0.4.8
    Uninstalling masci-tools-0.4.8:
      Successfully uninstalled masci-tools-0.4.8
  Running setup.py develop for masci-tools
Successfully installed masci-tools
bash-4.4$ pip3 show masci-tools
Name: masci-tools
Version: 0.5.0
Summary: Tools for Materials science. Vis contains wrappers of matplotlib functionality to visualize common material science data. Plus wrappers of visualisation for aiida-fleur workflow nodes
Home-page: https://github.com/JuDFTteam/masci-tools
Author: Jens Broeder
Author-email: [email protected]
License: MIT License, see LICENSE.txt file.
Location: /Users/alzeer/masci-tools
Requires: numpy, scipy, matplotlib, h5py, pandas, ase, bump2version, future, lxml, more-itertools, seaborn, deepdiff, humanfriendly, mendeleev
bash-4.4$ git rev-parse HEAD
84354df6022cd11519cfd5efca417eb87ccff3c3

and I still get the initial error.

[janssenhenning]

Yes, there is a test which is literally the code block you have 🤔
Did you restart the jupyter notebook after installing the develop version?

[MRedies]

I restarted the Kernel, which usually works for changes in my modules

[janssenhenning]

And the kernel corresponds to the correct python executable, where you installed the develop version?
Can you do the following without error import masci_tools.vis.data?

[MRedies]

No. This module seems unknown. I had to try a bit a few combinations untill I found a pip, that would let me install things without superuser.

[MRedies]

I don't know how jupyter selects their python kernel. Maybe you can help Mahmood next time when you are in Jülich.

[janssenhenning]

Yes I'll be happy to help