Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add ML coupling #14

Draft
wants to merge 22 commits into
base: datawave_ml
Choose a base branch
from
Draft

Add ML coupling #14

wants to merge 22 commits into from

Commits on Aug 19, 2024

  1. Add variable for the convect_dp_gw net filepath to be read from namel… 

    …ist.
    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    fd6e669 View commit details
    Browse the repository at this point in the history

  2. WIP Adapt gw_drag.F90 to import ftorch and read in a net from file. T… 

    …ODO: Move reading net to init routine.
    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    bed301e View commit details
    Browse the repository at this point in the history

  3. Move reading of NN to initialisation routine for gw_drag.

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    7968fa9 View commit details
    Browse the repository at this point in the history

  4. Add a gw_final() subroutine to destroy NN instance.

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    b3ac3e9 View commit details
    Browse the repository at this point in the history

  5. Remove spurious test code.

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    16b53ac View commit details
    Browse the repository at this point in the history

  6. Update source to use latest version of FTorch following API Changes.

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    dfd451d View commit details
    Browse the repository at this point in the history

  7. WIP add gw_ml file with starting point for a routine to run the ML pa… 

    …rameterisation via FTorch.

* Implements reshaping of the data to the form expected by the net as a starting point.
  * Consider using torch layout argument to avoid transposition at a
    later date.

* Does not implement any normalisation of features as we need to discuss
  how to do this on a chunked grid when data is provided for the global
  grid.

* Not currently called from anywhere.
    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    0d334db View commit details
    Browse the repository at this point in the history

  8. Add normalisation file to namelist variables, read, and distribute.

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    01950bc View commit details
    Browse the repository at this point in the history

  9. Create gw_ml initialisation routine to read in net and normalisation … 

    …variables from file.
    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    fa35aed View commit details
    Browse the repository at this point in the history

  10. Add finalisation routine to gw_ml to destroy the net.

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    ca19081 View commit details
    Browse the repository at this point in the history

  11. Add normalisation routine for inputs to net.

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    e6c77c8 View commit details
    Browse the repository at this point in the history

  12. Add denormalisation routine for the gw_convect_dp net.

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    4d352a5 View commit details
    Browse the repository at this point in the history

  13. BugFix: Remove redundant net variable

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    8f9723c View commit details
    Browse the repository at this point in the history

  14. Couple gw_drag to the ML scheme

    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    db3dd29 View commit details
    Browse the repository at this point in the history

  15. Bugfix: Update ML switches to be logicals when calling gw_drag init a… 

    …nd finalize routines.
    @jatkinson1000
    jatkinson1000 committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    be8bea2 View commit details
    Browse the repository at this point in the history

Commits on Sep 9, 2024

  1. [DEBUG: Run net with ones as the input to compare to same operation u… 

    …sing pure python (Same result obtained).]
    @jatkinson1000
    jatkinson1000 committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    7795a76 View commit details
    Browse the repository at this point in the history

  2. Revert changes to debug commit using ones as net input.

    @jatkinson1000
    jatkinson1000 committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    bebcf7c View commit details
    Browse the repository at this point in the history

  3. Add output for U and V tendencies from GW from Net and Beres schemes.

    @jatkinson1000
    jatkinson1000 committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    de3b447 View commit details
    Browse the repository at this point in the history

Commits on Oct 15, 2024

  1. Set tendencies not predicted by the net to 0.0 for the ml scheme rath… 

    …er than running the physics scheme beforehand.
    @jatkinson1000
    jatkinson1000 committed Oct 15, 2024
    Configuration menu
    Copy the full SHA
    c724d24 View commit details
    Browse the repository at this point in the history

  2. Add iulog to gw_ml to write outputs to diagnostics in init scheme.

    @jatkinson1000
    jatkinson1000 committed Oct 15, 2024
    Configuration menu
    Copy the full SHA
    83c6cfa View commit details
    Browse the repository at this point in the history

  3. Add code to produce output required by the testcase.

    @jatkinson1000
    jatkinson1000 committed Oct 15, 2024
    Configuration menu
    Copy the full SHA
    03b1db6 View commit details
    Browse the repository at this point in the history

  4. Update rhoi output to be ncol rather than pcol.

    @jatkinson1000
    jatkinson1000 committed Oct 15, 2024
    Configuration menu
    Copy the full SHA
    b65912e View commit details
    Browse the repository at this point in the history