Bug report: Application crashed with NotExistentAttributeError

[unkcpz]

magnetism metal bandstructure calculation

The input structure is O.cif with bandstructure on and pdos on.

#======================================================================

# CRYSTAL DATA

#----------------------------------------------------------------------

data_VESTA_phase_1


_pd_phase_name                         'O                                     '
_cell_length_a                         4.27163
_cell_length_b                         4.27682
_cell_length_c                         4.57148
_cell_angle_alpha                      90
_cell_angle_beta                       66.72128
_cell_angle_gamma                      90
_symmetry_space_group_name_H-M         'P 1'
_symmetry_Int_Tables_number            1

loop_
_symmetry_equiv_pos_as_xyz
   'x, y, z'

loop_
   _atom_site_label
   _atom_site_occupancy
   _atom_site_fract_x
   _atom_site_fract_y
   _atom_site_fract_z
   _atom_site_thermal_displace_type
   _atom_site_B_iso_or_equiv
   _atom_site_type_symbol
   O1       1.0    0.93782      0            0.14667     Biso  1.000  O
   O2       1.0    0.06218      0            0.85333     Biso  1.000  O
   O3       1.0    0.43782      0.50000      0.14667     Biso  1.000  O
   O4       1.0    0.56218      0.50000      0.85333     Biso  1.000  O

Automated report

This issue was created with the app's automated bug reporting feature.
Attached to this issue is the full traceback as well as an environment
fingerprint that contains information about the operating system as well as all
installed libraries.

Additional comments (optional):

Example: I submitted a band structure calculation for Silica.

Attachments

Traceback

~/apps/quantum-espresso/aiidalab_qe/node_view.py in __init__(self, node, **kwargs)
    460 
    461         self.result_tabs.observe(on_selected_index_change, "selected_index")
--> 462         self._update_view()
    463 
    464         super().__init__(

~/apps/quantum-espresso/aiidalab_qe/node_view.py in _update_view(self)
    488                 and "band_structure" in self.node.outputs
    489             ):
--> 490                 self._show_band_structure()
    491                 self._results_shown.add("band_structure")
    492 

~/apps/quantum-espresso/aiidalab_qe/node_view.py in _show_band_structure(self)
    499 
    500     def _show_band_structure(self):
--> 501         data = export_data(self.node)
    502         bands_data = data.get("bands", None)
    503         dos_data = data.get("dos", None)

~/apps/quantum-espresso/aiidalab_qe/node_view.py in export_data(work_chain_node)
    170 def export_data(work_chain_node):
    171     return dict(
--> 172         bands=export_bands_data(work_chain_node), dos=export_pdos_data(work_chain_node)
    173     )
    174 

~/apps/quantum-espresso/aiidalab_qe/node_view.py in export_pdos_data(work_chain_node)
    130         pdos_orbitals = []
    131 
--> 132         for orbital, pdos, energy in work_chain_node.outputs.projections.get_pdos():
    133             orbital_data = orbital.get_orbital_dict()
    134             kind_name = orbital_data["kind_name"]

/opt/conda/lib/python3.7/site-packages/aiida/orm/utils/managers.py in __getattr__(self, name)
    147             raise NotExistentAttributeError(
    148                 f"Node<{self._node.pk}> does not have an {prefix} with link label '{name}'"
--> 149             ) from exception
    150 
    151     def __contains__(self, key):

NotExistentAttributeError: Node<429> does not have an output with link label 'projections'

Environment fingerprint

eNpdWFtv47gV_iuB-7oSbPkSO2-LFH0oZtppM9OiKIoFJdE2bYpkSCqOspj_3nOTrCySF30kD8_1O4f-ffGmYzLeLZ5WvyyCVfno
Y7d4-n2hYnM2WTe5j3rx9N_FblObvPhlsfgf7Bvy2bvfpqOLdflYHmDxjvxptaqKH3Xvcv_w8vXbw_dz__AXXT-
sH1brp-3yabt9-PH9-aFaVtXiJ0hUzVWddKKbjWlV0Xi8d7Eqd-UWJDPoQzad-
cCFZbkpl9NCSLpvPcG7cj3Br70CBTqdQtQpedJ0W1bTekopFDcfr0frb3S6Kpf1alrP3ttE-Pou1aoaoGpVruB_BhbaOtr7B_gV1
a0qFP0ZppubszKODP20ejPtSedU1CqxG9Yi1BfBXBVAu3IvRyBir2Sa7LHtRXcDqXIQRHe1aUjOI7vTJVOdc2fFxSQohNbERMim3
CCCOfCmi6DyWbxD4uKp8V2wOpNl5ZKPx5N3VdEcj4bds_4jXNTGtcadEq-LP2Ik169G4WQvWHJgj6fBNcVJOx1V9pE2ruhczvGTo
BpSqFHWjmFEqIlDyCROAONUHHx0PksKoZm11apBAzclR6H2V30m0zaULPVrsHJiVZVrVLOOPlsT2MuPdOoZhOhn73L0dtxLqja4M
KYSSKRLGx2zYVctMZeW5R5R9h58s2qQHbHVmc6xmxFJOhcOoq4s1EKUGPBVZ90Zd_SfFGtMbHpO49UjObmxprkCAF9kIH0XElQu4
KUUyrjkjuZUHI3VUmTLaS0F4xypQalPC961ipQ-sBD8LqTIi7NyrYU8kHyUDSDf1xfAtmDMjjB_Nbrpc9ZRtmJGUEz9KapwHmZZ3
wyNHZXgKLaqMVnPErzVdX8iz6zAuBUhQDOSV1sJfquPfdLte2fFhbwRGKRR4ptKtv75otzJEwB_GK_WJSbH2a62tzfTSAHBP8ZZQ
0Vql6WwKuEy3SljizcIa3tPdjIa9oLZ3rgsdIRY1msuFXLhUaWsguHwbGkHhst6ijS6iXZBfs5Scc9gakr9HnzMUoiozLFn6keH8
jby-Zv5YOxAiNbO6iya4ibTOsXGoxDToVRr6qLTWYFVnBUrzuX7KtCz72ODDQACUZH2UF14QJhQSmfJC1cdnbbEgjsKEGCQvKTZv
nxkQOIAabMczyGEbNJnI8Q-icxgy4zkALnf_Chd6A4W-j1r1-pWTMcjF92a0V_4_VfjLqpiZ3AYL4QU0MT07PKLr8EHIoiA5B0FW
4q7EizqIx-T74Tlrmb8dunBDh0LKEzIGHIPyKzmK9xWN9IZBJ51HG5FdMXV3O4MtxEgefsm9cgZZVW8QmkDKcWR9VAVzLuJLlZsw
1d19WLmjj7jtfU3J-k5Qn14Ucexs6CSncrIv-wjbODLT2BhHLCJXERu7kzKvSB70ryDtB9GroXoPBIWdQEMEcd62I-6d6HjwhybU
pdOSF7iG7qit9m0psmSsHjM1UBj4BzGOC9djQOVykIweNTplAvqa8ZLs2U44zzwTobzEONO9s3oG1nN10Lj0jXwIoV2Qznp-k44D
UxDwFs9OX6DTqWpSbXioR01mhG7p4q4hmiMmZopumKtAwwROavaagnXksDuNZDC4zfwfJLk2ArgG1AH3JyEeHljVJ25MndytWAKe
QlZRd_5fJ-_mFxCzUzNk08AztINN0dBTqmf1XCA9mR1go6pJYtRp2_G8rS3F9r9ZlwefcPHwjjB0Cem2UD3ipVWXE58TogfvPRM-
owe6O6s-3R371QYuBgyjZZXkyW2Fbk4IS3NDUxD48OpKnhuEWdhSEMeQA7wZZr1-TA0ykTOqYpL59vwbI7__De5iNMMNvmk8mxUB
mQqX6hTltTC4CwT44GdNJywYyXhdrkQMhvYlLOHynVHW78OL__4IgPompC_qWcryc4n_x60e3n5QkelzsOAenDWrWUICAMgKU8-F
McMqYMUH8mKLX0BxkkyKLA4YnvoIVr8Wo2pxSvJ9idzHGTkGBegDVzSPKeH7C-N6oLkoWBRDfIwkIMfZMpqNPg_v379QpI3oswHD
enVKBbG_zG7aYKODcrBfprjMKPXqKlHqpH7dnsCX3tgET5c8XQzYgWNm7MJblrovFAYOzEem_Vh_yh3kdheddqWg5LZB-th_RkvG
6ZhLLFqWipwqRinPjgIGYxXpMZAjhcwX0eeKSseqAEXzuKZLml9DSPlHqh_v0B3rb5Hlc4zlWHw7cM0vhxwriuDTxm5a7la7j9tK
VLTEU-yI9NZW5w6z6qbJWI6g3f73rSSSiTC4ByMDZbHq0fKxGTeuXdxjadwmpr2ji6AlP_VYkeewrf5hBY_pqFjzeNEerVUefIWI
S2ziki0QtYjGKHTFBiw9frAFY_wMDI_zmLAzT2838euR5B2OAMPxHVcJsDD8EBQ8lZeVQKmPL3x2LwM803RO4MTcisvUIShPFQbw
L1NtlKmtN2PKcOFhN9GGih_R7mUKTa_th2R0la-ozLZ8vixlXGKMOA5nWaUDt8Qw99o-sKfGpI8jWjtA4YOxYWO373DRxY8r7MhG
WuZO_pwhNorJMAjR_UR0qNes9LMjP9SUHvDrP_cVG7OrT-NpmKAbw1MTtP7GMN60zAIWC8NT34lAFA78KW8f5fyArmdtbZjTtAVN
IbBu9-1aXp1ilxeCjoaj6OHNGN8Z9yXk6sn58x-77jB7J5HfRCA9800Pq7kmTZA7s0eZR8mBBKBBPPz5_8B5x0blA==

By submitting this issue I confirm that I am aware that this information can
potentially be used to determine what kind of calculation was performed at the
time of error.

[mbercx]

Indeed, the current approach for obtaining the PDOS data does not consider spin-polarised calculations. In this case the output port is not just projections, but there are two (projections_up, and projections_down):

https://github.com/aiidateam/aiida-quantumespresso/blob/e3268c573dd543f63b3770bf9902a5c43dd2bec0/aiida_quantumespresso/calculations/projwfc.py#L53-L54

I don't think @dou-du's widget can currently elegantly plot up and down states (also related to #104). I think the best approach for fixing this issue for now is to add the up and down PDOS together.

[unkcpz]

Thanks @mbercx.

I don't think @dou-du's widget can currently elegantly plot up and down states (also related to #104).

Yes, I was checking this one. So this issue is caused by qe-app. But I find the QeWorkChain also failed. Let me check the error message and paste it here.

[unkcpz]

  File "/opt/conda/lib/python3.7/site-packages/plumpy/process_states.py", line 231, in execute
    result = self.run_fn(*self.args, **self.kwargs)
  File "/opt/conda/lib/python3.7/site-packages/aiida/engine/processes/workchains/workchain.py", line 214, in _do_step
    finished, stepper_result = self._stepper.step()
  File "/opt/conda/lib/python3.7/site-packages/plumpy/workchains.py", line 299, in step
    finished, result = self._child_stepper.step()
  File "/opt/conda/lib/python3.7/site-packages/plumpy/workchains.py", line 250, in step
    return True, self._fn(self._workchain)
  File "/home/aiida/.local/lib/python3.7/site-packages/aiidalab_qe_workchain/__init__.py", line 329, in results
    "projections", self.ctx.workchain_pdos.outputs.projwfc__projections
  File "/opt/conda/lib/python3.7/site-packages/aiida/engine/processes/process.py", line 355, in out
    return super().out(output_port, value)
  File "/opt/conda/lib/python3.7/site-packages/plumpy/processes.py", line 81, in func_wrapper
    return func(self, *args, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/plumpy/processes.py", line 1276, in out
    raise ValueError(msg)
ValueError: Error validating output 'projections' for port 'outputs': Unexpected ports {'result': 'projections'}, for a non dynamic namespace

[mbercx]

@unkcpz indeed, that's the same error I encounter.

Working on a fix now. Seems the energy arrays for projections_up and projections_down are _slightly different:

np.all(np.isclose(proj_down.get_pdos()[0][1], proj_up.get_pdos()[0][1], rtol=0, atol=1e-4))

returns False, but:

np.all(np.isclose(proj_down.get_pdos()[0][1], proj_up.get_pdos()[0][1], rtol=0, atol=1e-3))

returns True. I think for now this isn't a massive issue, but I'm slightly surprised they aren't the same.

[mbercx]

Note that the DOS will also be split up in dos_spin_up and dos_spin_down arrays, both stored in a single XyData. Again I would suggest just adding the two together, until @dou-du's widget is able to plot up vs down states.

[mbercx]

Another thing to note here: If we really want to represent the results properly, perhaps we should check if the end result is actually magnetic. If not, we should have a non-magnetic representation.

[mbercx]

So, I can't plot both the up and down states with the current widget-bandsplot, since it adds the PDOS together by orbital type (s, p, d, f), which is based on the first character in the value corresponding to the "orbital" key, which represents the orbital name:

https://github.com/osscar-org/widget-bandsplot/blob/8cece372479063de3e7a63877a823d2cfb784fc7/widget_bandsplot/bandsplot.py#L78-L89

So for now we can't distinguish up and down states in the widget.

[unkcpz]

@mbercx Thanks for checking this.

Another thing to note here: If we really want to represent the results properly, perhaps we should check if the end result is actually magnetic. If not, we should have a non-magnetic representation.

I personally will not like this feature. It is too much intelligent to guess and 'correct' what the user wants to do. Maybe they explicitly gonna check the input structure is no magnetism. This is quite normal I guess even if the structure contains the magnetic elements they still reveal the difference between spin up and spin down.