[Re] [Ten Year Challenge] Oscillating chemical reactions and phase separation simulated by molecular dynamics

[kneth]

Original article: https://aip.scitation.org/doi/abs/10.1063/1.475237

@article{Geisshirt:1997,
  author  = {Kenneth Geisshirt and Eigil L. Praestgaard and Soren Toxvaerd},
  title   = {Oscillating chemical reactions and phase separation simulated by molecular dynamics},
  journal = {Journal of Chemical Physics},
  volume  = {107},
  number  = {22},
  pages   = {9406--9412},
  year    = {1997}
}

PDF URL: https://github.com/kneth/rescience-ten-year-challenge-paper/blob/master/article.pdf
Metadata URL: https://github.com/kneth/rescience-ten-year-challenge-paper/blob/master/metadata.yaml
Code URL: https://github.com/kneth/MDreac/tree/master/oldsrc

Scientific domain: non-equilibrium thermodynamics, chemical kinetics
Programming language: Fortran-77
Suggested editor:

[kneth]

Declaration of participation: ReScience/ten-years#1 (comment)

[rougier]

Thanks for your submission, we'll assign an editor soon.

@khinsen Can you find an editor (or edit yourself) for this submission?

[pdebuyl]

Hi @rougier I can edit.

[rougier]

Thanks @pdebuyl.

[pdebuyl]

Just a notice to let you know that "reviewer finding" is well in process.

[pdebuyl]

@fhoefling , can you review this submission ?

[fhoefling]

I started with the review. Some first feedback on the software:

A detailed guide of how to proceed is missing and should be given. Yet, building the programme went smoothly, just by installing gfortran and running make in oldsrc.

Next, I started the script phd.sh from the protocols folder. On a Xeon E3-1241 @ 3.50GHz, the script was running for more than 5 hours on a single core and created almost 8 GB of data. Is this the intended behaviour? If so they user should be alerted. No further output was created.

I've got the following Python error repeatedly:

File "../../../diffcoeff.py", line 8, in <module>
from sklearn.linear_model import LinearRegression
ModuleNotFoundError: No module named 'sklearn'

Please use the linear regression functionality from a more basic library. Machine learning approaches were not needed in the original work.

The provided README is rather brief. It mentions some means to run the programme in parallel. Is this still possible today?

[pdebuyl]

@kneth even though the review is still ongoing, you may already take into account the first comments of @fhoefling

[fhoefling]

Actually, I can only continue with the reviewing after the aforementioned issues have been solved. Currently, the programme does not output data. I apologise that I have not communicated this more clearly.