problems with the latest numpy

ondrejkrejci commented 8 months ago

The problem was shown once PBC were applied:

I get it to work with python 3.10.4 and numpy 1.23.5.

It does not work with python 3.11.4 and numpy 1.25.1.:

File "/m/phys/project/sin/jestilj/PPSTM/tests/4NPaG/monolayer_3/PPSTM_simple.py", line 194, in <module>
eigEn, coefs, Ratin = RS.read_AIMS_all(name = files_path + name , geom= files_path + geometry_file, fermi=fermi, orbs = sample_orbs, pbc=pbc, cut_min=cut_min, cut_max=cut_max, cut_at=cut_atoms, lower_atoms=lower_atoms, lower_coefs=lower_coefs);
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/m/phys/project/sin/jestilj/PPSTM/pyPPSTM/ReadSTM.py", line 288, in read_AIMS_all
Ratin, at_num = get_AIMS_geom(geom=geom)
                ^^^^^^^^^^^^^^^^^^^^^^^^
File "/m/phys/project/sin/jestilj/PPSTM/pyPPSTM/ReadSTM.py", line 173, in get_AIMS_geom
Ratin = for_PBC(atoms,lvs)
        ^^^^^^^^^^^^^^^^^^
File "/m/phys/project/sin/jestilj/PPSTM/pyPPSTM/ReadSTM.py", line 128, in for_PBC
assert (lvs != (None or []) ), "Lattice vectors (cell) not specified"
        ^^^^^^^^^^^^^^^^^^^
ValueError: operands could not be broadcast together with shapes (3,3) (0,)

LauriKurki commented 8 months ago

I remember seeing this error previously, I can take a look today if you didn't start yet @ondrejkrejci

ondrejkrejci commented 8 months ago

No I did not, I am ill & having a pedagogy course at the same time.

From: Lauri Kurki @.> Sent: Friday, March 8, 2024 10:29 AM To: Probe-Particle/PPSTM @.> Cc: Krejci Ondrej @.>; Mention @.> Subject: Re: [Probe-Particle/PPSTM] problems with the latest numpy (Issue #22)

I remember seeing this error previously, I can take a look today if you didn't start yet @ondrejkrejcihttps://github.com/ondrejkrejci

- Reply to this email directly, view it on GitHubhttps://github.com/Probe-Particle/PPSTM/issues/22#issuecomment-1985258930, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ADFZAFNIY5DUEEF3OUNOLEDYXFZGBAVCNFSM6AAAAABELGEUGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSOBVGI2TQOJTGA. You are receiving this because you were mentioned.Message ID: @.**@.>>

LauriKurki commented 8 months ago

Get well soon! When you're back, there's a pull request fixing this that requires your review.

Probe-Particle / PPSTM

problems with the latest numpy #22