search

ComputationalThermodynamics / MAGEMin

The parallel Mineral Assemblage Gibbs Energy Minimization package
GNU General Public License v3.0
71 stars 17 forks source link

A problem about MAGEMin_EquilibriumPath #44

Closed CHNxzr closed 1 year ago

CHNxzr commented 2 years ago

Hi, what can I do to debug?

reply30 reply31

boriskaus commented 2 years ago

that's a question for @kellertobs

kellertobs commented 2 years ago

I tested the composition and PT-path as in the screen grab provided and can't reproduce the error. In my case the data read routine runs through with no problem. Please make sure you've updated your executable to v.1.2.7. If the error persists, can you please post the exact version of MAGEMin_EquilibriumPath.mlx that gives the error?

kellertobs commented 2 years ago

Having said that, the same test case brought up another error later when the data is written to CSV files. It's an important fix that I'm going to create a pull request for in MAGEMin_dev. The error occurs when a once stable phase disappears later along the PT path. Some of the data arrays need to be padded with zeros to avoid inconsistent array dimensions when writing the data to CSV. @boriskaus or @NicolasRiel, this fix should be updated in the main distribution directly but does not require an update to the MAGEMin executable. I'll tag you in the pull request once it's ready.

kellertobs commented 2 years ago

pull request for the fix is ready to be merged

CHNxzr commented 2 years ago

I'm sure I have updated for the latest version. But the problem still exist. reply32

boriskaus commented 2 years ago

It works fine on my system as well. Does the default computation work for you (not using your own chemistry)?

NicolasRiel commented 2 years ago

Could you try again, I now pushed the fix from Tobias on MAGEMin repo. I also tried your composition on my machine (On windows) and it works. Just to double check, which mode do you use here?

CHNxzr commented 2 years ago

It works fine on my system as well. Does the default computation work for you (not using your own chemistry)?

Yes!

boriskaus commented 2 years ago

in that case, can you give us more details on which parameters you exactly changed, starting from the default parameters (all changes please).

CHNxzr commented 2 years ago

Could you try again, I now pushed the fix from Tobias on MAGEMin repo. I also tried your composition on my machine (On windows) and it works. Just to double check, which mode do you use here?

Closed system.

CHNxzr commented 2 years ago

in that case, can you give us more details on which parameters you exactly changed starting from the default parameters (all changes please).

ok! ![Uploading reply33.png…]()

CHNxzr commented 2 years ago

reply34 reply35 reply33

CHNxzr commented 2 years ago

Exactly I can receive the "_matlab_output" txt , it seems that there is just a wrong with printing.

kellertobs commented 2 years ago

I just ran it again with your exact composition and P-T path and it runs through with no error.

boriskaus commented 1 year ago

since we have no news on this, I'm closing the issue