Skip to content

feat: osiris wrapper#279

Open
physicistphil wants to merge 22 commits into
ergodicio:mainfrom
physicistphil:osiris-wrapper
Open

feat: osiris wrapper#279
physicistphil wants to merge 22 commits into
ergodicio:mainfrom
physicistphil:osiris-wrapper

Conversation

@physicistphil

@physicistphil physicistphil commented May 26, 2026

Copy link
Copy Markdown

Todo:

  1. Merge outputs into a single file (xarray) instead of uploading h5
  2. Build reasonable diagnostic / plots (mimicking what we have for vlasov). Right now it's just a few useful plots

physicistphil and others added 15 commits May 26, 2026 13:01
@physicistphil
Initial osiris wrapper commit
5073964
@physicistphil
.gitignore changes
509cad2
@physicistphil
added osiris as a solver
49d583f
@physicistphil
Uploaded original config.yaml instead of translated osiris deck
761a23d
@physicistphil
Aggregate osiris h5 files into an xarrary; save and upload
33ca8a0
@physicistphil
Brought plots up to parity with other solvers
3c6f257
@physicistphil
Fixed issue when multiple datasets were in an h5 file
9f818e8
@physicistphil
More plots added: zoomed omega-k, current, averaged distribution
cb78418 
functions (should probably be moved to another branch). Added density
and temperature profiles. Fixed LaTeX issues
@physicistphil
Modified adept (io.py + plots.py) so the full OSIRIS canned-plot set
704aba8 
regenerates from the saved NetCDF artifacts alone — no rerun, no raw
MS/ tree. Made list_diagnostics/load_series/load_hist_energy dispatch
between the MS/ HDF5 tree and a binary/ NetCDF dir, made field-energy
source-agnostic, and now persist HIST/energy.nc in save_run_datasets.
Added 3 tests
@physicistphil @claude
feat(osiris): NetCDF plot regeneration + plot-set refinements
08a2422 
- regen harness: rebuild the full canned plot set from saved NetCDFs (no rerun)
- f(p) and delta-f lineouts; temperature profile from phase-space Maxwellian fits
- number-density profiles (initial/final/late-mean); 2-panel equal-aspect omega-k
- phase-space & spacetime: space on x-axis, cropped to box, log-contrast floor
- proper-LaTeX titles (prose vs math)

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
@physicistphil
Make srun default instead of mpirun for perlmutter
c6bb017
@physicistphil
Modified distribution function plots; added wave transmission+absorpt…
7528dde 
…ion+reflection plot (should probably be moved to osiris-lpi repo)
@physicistphil
Store single precision floats, add compression (most useful for phase…
d0090ea 
…space)
@physicistphil
Optionally drop t=0 raw chunk (useful for massive raws)
55f662e
@physicistphil
Changed default osiris run directory to ./checkpoints so that the
cbaeabe 
sync-up.sh script doesn't delete ongoing osiris runs
@joglekara

joglekara commented Jun 10, 2026

Copy link
Copy Markdown
Member

is this ready to go?

@physicistphil

physicistphil commented Jun 17, 2026

Copy link
Copy Markdown
Author

Gave it a read through; should be good to go. Idk if the tests are actually all that useful but I left them in ¯_(ツ)_/¯

@physicistphil physicistphil marked this pull request as ready for review June 17, 2026 00:00
@physicistphil @claude
Populate OSIRIS units.yaml from the deck's reference scale
ba10baa 
BaseOsiris.write_units() previously returned {} so OSIRIS runs logged an empty units.yaml. Derive the physical reference scales (wp0, tp0, n0, v0, x0, c_light, beta, box_length, sim_duration) from the deck's simulation.n0 (density) or simulation.omega_p0 (frequency); when both are present, n0 wins, as in OSIRIS. This mirrors the canonical key set the other adept solvers emit so OSIRIS runs are comparable in MLflow.

Adds skin_depth_normalization and skin_depth_normalization_from_frequency to normalization.py. OSIRIS has no single global reference temperature (species carry per-species thermal momenta), so the temperature-dependent keys (T0/nuee/logLambda_ee) are omitted.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
@physicistphil

physicistphil commented Jun 18, 2026

Copy link
Copy Markdown
Author

test-pip-install failure — not caused by this PR

The test-pip-install failure is a pre-existing break on main (started 2026-06-15, runs #287/#288; green on 2026-06-10). This branch just inherited it.

Root cause: unpinned dependency drift between matplotlib and scienceplots.

  • import adept → adept/_lpse2d/helpers.py does import scienceplots
  • scienceplots 2.2.1 (latest) calls plt.style.core.read_style_directory(...) at import time
  • matplotlib 3.11.0 removed the matplotlib.style.core attribute, so the import raises:
AttributeError: module 'matplotlib.style' has no attribute 'core'

pyproject.toml pins neither package, and test-pip-install does a clean, unconstrained install — so it picked up the freshly released matplotlib 3.11.0. The other test jobs use a constrained environment, which is why only this check is red. No fixed scienceplots release exists yet (2.2.1 is latest and still references the removed attribute).

Suggested fix: pin matplotlib<3.11 in pyproject.toml until scienceplots ships a compatible release. Probably best as a small standalone PR against main rather than folded in here.

@joglekara

joglekara commented Jun 21, 2026

Copy link
Copy Markdown
Member

Read through this — looks good to merge from my side. The wrapper architecture is clean, the security surface on the subprocess runner is fine (list-form Popen, no shell), and the synthesized-fixture tests (deck roundtrip / post / plots / units) give real CI coverage. One thing I'd like fixed before it goes in:

Hardcoded local paths in the example configs. configs/osiris/twostream-1d.yaml and twostream-1d-short.yaml both point deck: and binary: at /home/phil/Desktop/pic/..., so the shipped examples won't run for anyone else. Since this PR already adds in-repo example decks under tests/test_osiris/decks/, could you repoint the configs at those (e.g. deck: tests/test_osiris/decks/two-stream-1d) and drop the hardcoded binary: in favor of the OSIRIS_BIN / OSIRIS_BIN_<dim>D env fallback the runner already supports? That makes the examples runnable out of the box.

Otherwise LGTM — nice work on the docstrings and the NetCDF regen path. (Separately: the test-pip-install red is the pre-existing matplotlib/scienceplots break on main, not this PR.)

@physicistphil

physicistphil commented Jun 22, 2026

Copy link
Copy Markdown
Author

Fixed the hardcoded paths. Also updated the docs/ markdown file (osiris-adept-usage.md). Another note: the canned plots were written for 1D sims, so 2D may cause issues (I haven't tried any 2D runs), but I figure we can cross that bridge when we get to it. Otherwise it should be good to go

@joglekara joglekara self-requested a review June 22, 2026 22:05
joglekara
joglekara approved these changes Jun 22, 2026
View reviewed changes

@joglekara joglekara left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🚀

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@joglekara joglekara joglekara approved these changes

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@physicistphil @joglekara