Error running tardis_kromer_plotter.generate_plot()

Question

Error running tardis_kromer_plotter.generate_plot()

Closed this issue 7 years ago · 11 comments

Following instructions.md:

import tardis
import tardis_kromer_plot as tkp
import tardis_minimal_model as tmm

config = tardis.yaml_load("tardis_example_carsus.yml")
mdl = tardis.run_tardis(config)
minmodel = tmm.minimal_model(mode="virtual")
minmodel.from_interactive(mdl)


# initialise the plotting tool
plotter = tkp.tardis_kromer_plotter(minmodel, mode="virtual")
# generate plot
plotter.generate_plot(xlim=(3000,1e4), twinx=True)

The last statement results in the following error:

Traceback (most recent call last):
  File "<stdin>", line 2, in <module>
  File "/Users/lshingles/Dropbox/GitHub/tardisanalysis/tardis_kromer_plot.py", line 364, in generate_plot
    self._generate_emission_part()
  File "/Users/lshingles/Dropbox/GitHub/tardisanalysis/tardis_kromer_plot.py", line 408, in _generate_emission_part
    histtype="stepfilled", normed=True, weights=weights)
  File "/usr/local/lib/python2.7/site-packages/matplotlib/__init__.py", line 1898, in inner
    return func(ax, *args, **kwargs)
  File "/usr/local/lib/python2.7/site-packages/matplotlib/axes/_axes.py", line 6156, in hist
    raise ValueError('weights should have the same shape as x')
ValueError: weights should have the same shape as x

The config file should match the example config, except that the atomic database points to a database I've made with carsus.

tardis_example_carsus.yml.txt

Answer 1 · 2017-10-13T07:48:44.000Z

@lukeshingles we updated and fixed several problems. Are you using the newest versions of both analysis and tardis

Answer 2 · 2017-10-13T08:25:23.000Z

Yes, latest master branch of both. Can you reproduce the error?

Here's the atomic database (23MB):
https://www.dropbox.com/s/bwffn033offnmjq/kurucz_latest_o_to_ni.h5.zip?dl=0

Answer 3 · 2017-10-13T08:31:52.000Z

@lukeshingles Can you reproduce the error with tardis example or is it only with your database?

Answer 4 · 2017-10-13T08:40:09.000Z

The atomic database in tardis_example.tar.gz doesn't run with the latest version of tardis. Is there a new example database for me to use?

Answer 5 · 2017-10-13T08:49:13.000Z

The latest version is in tardis-refdata

Answer 6 · 2017-10-13T09:04:07.000Z

@lukeshingles this is seemingly the problem

[<Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>, <Quantity [] Angstrom>

I'm looking further into it - any idea why it would find no non-interacting packets?

Answer 7 · 2017-10-13T09:10:46.000Z

the problem is in tardis_minimal_model

Answer 8 · 2017-10-13T09:45:59.000Z

@lukeshingles @yeganer mode='real' works. The capturing of the precise information for virtual was disabled as it lead to a memory leak for large numbers of virtual packages. I think it should be able to enable this.

Answer 9 · 2017-10-13T11:55:47.000Z

So compiling with enabled virtual logging solved the issue?

Answer 10 · 2017-10-13T13:30:12.000Z

If that is the case, there should be a warning, at least when trying to access model.runner.virtual_packet_energies because I made a safeguard there.

Answer 11 · 2017-10-13T13:45:44.000Z

Ok, once I reverted my astropy version from 2 to 1.3 and used mode='real', it runs without error. Thanks!