empty INFO_ITER

[Yue Gu]

Dear developers:
I compiled DMFTwDFT based on Python/2.7.9 and Wannier90/2.1.0. No error was reported during compilation. However, there are some problems:

(1) when I tried to use the command "DMFT.py - h", the following error occurred:
Traceback (most recent call last):
File "/WORK/bin/g-vasp/DMFTwDFT-master/bin/DMFT.py", line 18, in
from INPUT import *
ImportError: No module named INPUT
Then, I copied the INPUT.py of example to the directory /DMFTwDFT-master/bin/, and then the error disappears. Is this operation correct?

(2) Following https://dmftwdft-project.github.io/DMFTwDFT/tutorials.html, I try to run the example-LaNiO3_vasp. However, the program will stop soon, with empty INFO_ITER and such information in "DMFT.error":

Traceback (most recent call last):
  File "~/DMFTwDFT-master/bin/RUNDMFT.py", line 464, in <module>
    p, pC, pD, it, itt, para_com, DMFT.mu, Ed, DMFT.Vdc, args.hf
  File "~/DMFTwDFT-master/bin/IMP_SOLVER.py", line 216, in RUN_CTQMC
    shutil.copy2("ctqmc.log", "ctqmc.log." + str(itt) + "." + str(it))
  File "/WORK/app/Python/2.7.9/lib/python2.7/shutil.py", line 130, in copy2
    copyfile(src, dst)
  File "/WORK/app/Python/2.7.9/lib/python2.7/shutil.py", line 82, in copyfile
    with open(src, 'rb') as fsrc:
IOError: [Errno 2] No such file or directory: 'ctqmc.log'

How can I solve this problem?
Many thanks!

[Vijay Singh]

Dear Yue,

Thank you for using DMFTwDFT code. 

I see that the impurity directory "imp0" has not been generated in the running folder.

Please ensure that the CTQMC - a third-party software - has been compiled properly. Please see the instructions here - http://hauleweb.rutgers.edu/tutorials/Tutorial0.html. Also, make sure you have linked the GSL lib while running the code - which is essential for the ctqmc.

If you think all is okay, you can share your input file with me. We shall try to run it here and let you know if we find something.

I'll be looking forward to hearing from you.

Thank you.

Sincerely,

Vijay

[Yue Gu]

Dear Vijay,

Thank you for your quick reply.

Following your suggestion, I found that the error of CTQMC comes from the unsuccessful loading of libmpigc4.so.4. So I reload the MPI/Intel module. 

However, a new problem appeared: the "atomd" module has an error (see the atomd.log) during the compilation. 

The good news is that the CTQMC has successfully compiled, and the example in  http://hauleweb.rutgers.edu/tutorials/Tutorial0.html can also run successfully.

My compilation is based on (1)Python/2.7.9  (2) LAPACK/3.5.0  (3) BLAS/3.5.0 (4) fftw/3.3.5 (5) gsl/1.16-icc.

Look forward to your reply. Thank you!

Sincerely,

Yue Gu

[Uthpala Herath]

Dear Yue, 

In addition to Vijay's response, to address the problem with, "ImportError: No module named INPUT", you can set "export PYTHONPATH= ". This clears the current PYTHONPATH and allows for the running of DMFT.py even when you don't have INPUT.py in the /bin directory.

Let us know if you still have issues regarding the ctqmc part.

Best,

Uthpala

--
You received this message because you are subscribed to the Google Groups "DMFTwDFT" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dmftwdft+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dmftwdft/6519648b-f69b-403e-9428-e98e7633d355n%40googlegroups.com.

--

Uthpala Herath

PhD Candidate

Department of Physics and Astronomy

West Virginia University

Morgantown, WV 26505

Tel. (304) 216-2535

Email: ukh...@mix.wvu.edu

Website: https://uthpalaherath.github.io/ 

[Yue Gu]

Dear  Uthpala,

Thank you very much for your reply! I think the ctqmc part has been successfully compiled now.
Do you have any suggestions for my error of ATOMD part. which is as shown above?

Best,

Yue Gu

[Singh, Vijay Ramdin]

Very good, very nice Yue, at least you have the ctqmc running.

I would like to mention that atom.d py only supports python 2 versions at the moment, which is also one of the files we access Haule's EDMFT code.

Since I see in your log file that you are accessing the Python 3 libraries. For example here:

-I/tmp/tmpv_59b9d2/src.linux-x86_64-3.7 -I/WORK/th_sz_xlzhou_1/.local/lib/python3.7/site-packages/numpy/core/include

Can you please make sure that all the lib assigned in make.sys or setup.py access  pthon2?

Also,  make sure - weave -  is properly imported which is a necessary module of atomd.py. Note: Python (version 2) with numpy and scipy as well as weave (the latter became a stand-alone package, separate from scipy, after scipy 0.14 version).

"import weave

#import scipy.weave as weave

from numpy import linalg"

Let me know how it goes.

I'll be looking forward to hearing from you.

Thank you.

Sincerely,

Vijay

---

From: dmft...@googlegroups.com <dmft...@googlegroups.com> on behalf of Yue Gu <boyg...@gmail.com>
Sent: Wednesday, March 2, 2022 7:26 AM
To: DMFTwDFT <dmft...@googlegroups.com>
Subject: Re: empty INFO_ITER

 

--
You received this message because you are subscribed to the Google Groups "DMFTwDFT" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dmftwdft+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/dmftwdft/c1fedd49-3c32-461b-be86-12d0e2a1c498n%40googlegroups.com.

[Yue Gu]

Dear Vijay,

Thank you for your suggestion.

After forcibly specifing default Python as Python2.7, the error of atomd is no longer reported.
However, when I run the example of LaNiO3, there is a new error about MPI, as shown in ksum_error and slurm.out.
Does this error comes from my MPI module itself (MPI/Intel/IMPI/4.1.3.048) or my input settings?

Many thanks!

Best,

Yue Gu