Bump pyrcel from 1.3.5 to 2.0.0 in /examples

[dependabot]

Bumps pyrcel from 1.3.5 to 2.0.0.

Release notes

Sourced from pyrcel's releases.

pyrcel v2.0.0

This is a major release. The numerical core has been rewritten from scratch on JAX + diffrax, replacing the numba + Assimulo/SUNDIALS stack from v1. The public interface is largely unchanged; see the migration guide for the handful of breaking changes.

Added

• Differentiable integration — the full parcel model is now differentiable end-to-end via jax.grad. Gradients flow through the ODE solve, the equilibration step, and the activation diagnostics.
• Batched ensemble runs — run_updraft_ensemble and smax_nact_ensemble run an array of independent parcel solves in a single compiled jax.vmap kernel with no Python loop overhead.
• GPU acceleration — pass device="gpu" to ParcelModel or set JAX_PLATFORM_NAME=gpu; install with pip install "pyrcel[gpu]".
• ModelOutput — ParcelModel.run() returns a structured ModelOutput object with .to_pandas(), .to_polars(), .to_xarray(), .to_netcdf(), .to_csv(), and .to_parquet().
• Short-circuit run modes — run(mode="smax") returns the peak supersaturation as a plain float (ideal for the differentiable path); run(mode="nd") returns the activated droplet number concentration.
• InterpolatedUpdraft — a JAX-traceable, vmap-safe time-varying updraft specified by (ts, Vs) arrays. Replaces bare callables from v1.
• JAX-native activation parameterizations — ARG2000 and MBN2014 are now differentiable callable classes in pyrcel.activation; the underlying implementations are fully jax.jit- and jax.grad-compatible.
• Hermite cubic S_max refinement — S_max is now located via a single-kernel Hermite cubic interpolant of the terminal supersaturation trajectory, giving a grid-independent peak estimate without a second ODE solve.
• console=True mode — pretty-prints setup tables, phase timers (with first-compile detection), trajectory sample, and activation summary without adding any overhead to the compiled solve.
• live=True mode — streams a z/T/S step table during integration, replicating the interactive output of the v1 CVode integrator.
• pyrcel.legacy subpackage — the original NumPy/SciPy implementations of thermodynamics and activation functions are preserved here for cross-validation; they are not part of the v2 numerical path.
• MkDocs/Material documentation site — full rebuild with API reference (mkdocstrings), migration guide, numerical methods description, worked examples, and GPU setup guide.
• Python 3.11, 3.12, and 3.13 support.
• PEP 561 type annotations throughout the public API, validated with pyrefly.

Changed

... (truncated)

Changelog

Sourced from pyrcel's changelog.

[2.0.0] — unreleased

This is a major release. The numerical core has been rewritten from scratch on JAX + diffrax, replacing the numba + Assimulo/SUNDIALS stack from v1. The public interface is largely unchanged; see the migration guide for the handful of breaking changes.

Added

• Differentiable integration — the full parcel model is now differentiable end-to-end via jax.grad. Gradients flow through the ODE solve, the equilibration step, and the activation diagnostics.
• Batched ensemble runs — run_updraft_ensemble and smax_nact_ensemble run an array of independent parcel solves in a single compiled jax.vmap kernel with no Python loop overhead.
• GPU acceleration — pass device="gpu" to ParcelModel or set JAX_PLATFORM_NAME=gpu; install with pip install "pyrcel[gpu]".
• ModelOutput — ParcelModel.run() returns a structured ModelOutput object with .to_pandas(), .to_polars(), .to_xarray(), .to_netcdf(), .to_csv(), and .to_parquet().
• Short-circuit run modes — run(mode="smax") returns the peak supersaturation as a plain float (ideal for the differentiable path); run(mode="nd") returns the activated droplet number concentration.
• InterpolatedUpdraft — a JAX-traceable, vmap-safe time-varying updraft specified by (ts, Vs) arrays. Replaces bare callables from v1.
• JAX-native activation parameterizations — ARG2000 and MBN2014 are now differentiable callable classes in pyrcel.activation; the underlying implementations are fully jax.jit- and jax.grad-compatible.
• Hermite cubic S_max refinement — S_max is now located via a single-kernel Hermite cubic interpolant of the terminal supersaturation trajectory, giving a grid-independent peak estimate without a second ODE solve.
• console=True mode — pretty-prints setup tables, phase timers (with first-compile detection), trajectory sample, and activation summary without adding any overhead to the compiled solve.
• live=True mode — streams a z/T/S step table during integration, replicating the interactive output of the v1 CVode integrator.
• pyrcel.legacy subpackage — the original NumPy/SciPy implementations of thermodynamics and activation functions are preserved here for cross-validation; they are not part of the v2 numerical path.
• MkDocs/Material documentation site — full rebuild with API reference (mkdocstrings), migration guide, numerical methods description, worked examples, and GPU setup guide.
• Python 3.11, 3.12, and 3.13 support.
• PEP 561 type annotations throughout the public API, validated with pyrefly.

Changed

... (truncated)

Commits

• 977e909 chore: fix pyproject.toml deprecations and MANIFEST.in for v2.0
• b856b5e ci: add PyPI trusted publishing workflow
• 1b643b6 docs: add CHANGELOG.md for v2.0.0 release
• 94fea68 feat(v2): merge pyrcel v2.0 JAX/diffrax implementation into master
• 2ecfe5f chore(merge): resolve modify/delete conflicts — keep v2 deletions of legacy R...
• 3b7ca2c refactor(integrator): single-kernel Hermite cubic S_max (closes #78, #79) (#81)
• 2500b93 fix: post-hoc Hermite cubic S_max finder + live/no-terminate NaN fix (#80)
• d1c202c docs: convert RST math directives to MkDocs arithmatex syntax
• 1fa2f01 docs(sweep): update sensitivity_sweep.md for 3-column figure
• 142457b feat(sweep): larger fonts, denser ticks, horizontal bottom colorbars
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.07%. Comparing base (f942453) to head (ef221b5).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1907   +/-   ##
=======================================
  Coverage   87.07%   87.07%           
=======================================
  Files         420      420           
  Lines       10751    10751           
=======================================
  Hits         9361     9361           
  Misses       1390     1390           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.