Questions and Answers

0 Like 0 Dislike

youanaa s ragheb

How were the material properties in the file “Band gaps from Strehlow and Cook” generated using “MAST_ML” .... How were all the elements in the “MAST_ML” library linked to the elements in “Band gaps from Strehlow and Cook” to create the file named "g

Report abuse

Chosen Answer

  1. 0 Dislike

    Benjamin Afflerbach

    It looks like your question got cut off. If you're looking for exactly how things are setup in MASTML you can check the documentation here: https://mastmldocs.readthedocs.io/en/latest/

    The code has been updated since this materials was put together, but if you check out tutorial 3 here: https://mastmldocs.readthedocs.io/en/latest/0_5_tutorials.html#mast-ml-tutorials

    that should give a good example of how everything works.

    Reply Report abuse

    Please login to answer the question.

    1. 0 Like 0 Dislike

      youanaa s ragheb

      First of all thank you for replaying me

      yes my question cut off mine was:

      "\ How were the material properties in the file “Band gaps from Strehlow and Cook” generated using “MAST_ML” .... How were all the elements in the “MAST_ML” library linked to the elements in “Band gaps from Strehlow and Cook” to create the file named "generated_features/generated_features.csv"  /"

      I already saw tutorial no.3, but i think it was very general so i didn't git it or apply it on the project as you did

      maybe cause i'm still a beginner so please could you recommend me a simpler tutorial.

      Reply Report abuse

      Please login to answer the question.

      1. 0 Dislike

        Benjamin Afflerbach

        Ok I think I see. The way the elements are linked is by matching the name of the elements from the composition column "Al2O3" for example. That string gets read in using the Pymatgen package which parses the string for us. Pymatgen identifies that Aluminum and Oxygen are in that composition and then we use a lookup table for all of the properties we have access to. on the mastml github for example (https://github.com/uw-cmg/MAST-ML/tree/master/mastml/magpie) you can see the .table files one for each property ordered by atomic number. So mastml just searches through all of those files for Aluminum and Oxygen properties and adds them to the generated features file.

        I'm not sure of a simpler tutorial off the top of my head unfortunately. If you are still feeling confused and want to talk through in more detail you are welcome to join our help-desk that we run each week (https://sites.google.com/wisc.edu/informatics-help-desk).

        Report abuse

0 Responses

No other responses made.