List view

Optionally release memory during solving process

enhancementNew feature or requestNew feature or request

Status: Open.

#219 In PyPSA/linopy;

· FabianHofmann opened on Jan 18, 2024

Support for sparse xarrays

data-model

Status: Open.

#248 In PyPSA/linopy;

· staadecker opened on Mar 7, 2024

perf: default integer arrays to int32 for ~25% memory reduction

performanceThis improves performance while not (meaningfully) altering behaviour for usersThis improves performance while not (meaningfully) altering behaviour for users

Status: Open (in progress).

PyPSA/linopynumber 566

#566 In PyPSA/linopy;

· FBumann opened on Feb 2, 2026

@/dot against a sparse matrix densifies the result to full _term

performanceThis improves performance while not (meaningfully) altering behaviour for usersThis improves performance while not (meaningfully) altering behaviour for users

Status: Open.

#748 In PyPSA/linopy;

· FBumann opened on Jun 4, 2026

merge of ragged expressions is the peak allocation in PyPSA model builds

performanceThis improves performance while not (meaningfully) altering behaviour for usersThis improves performance while not (meaningfully) altering behaviour for users

Status: Open.

#749 In PyPSA/linopy;

· FBumann opened on Jun 4, 2026

groupby-sum pads every group to the largest group size

performanceThis improves performance while not (meaningfully) altering behaviour for usersThis improves performance while not (meaningfully) altering behaviour for users

Status: Open.

#745 In PyPSA/linopy;

· FBumann opened on Jun 3, 2026

Benchmark comparison vs polar-high (Polars-backed LP/MIP eDSL on HiGHS) — thoughts?

performanceThis improves performance while not (meaningfully) altering behaviour for usersThis improves performance while not (meaningfully) altering behaviour for users

Status: Open.

#740 In PyPSA/linopy;

· MaykThewessen opened on Jun 2, 2026

Umbrella: long-format / sparse _term kernel (dense-_term memory cluster)

data-model

performanceThis improves performance while not (meaningfully) altering behaviour for usersThis improves performance while not (meaningfully) altering behaviour for users

Status: Open.

#756 In PyPSA/linopy;

· FBumann opened on Jun 4, 2026

Multi-key groupby([names]).sum() densifies to a dense cartesian grid (memory)

performanceThis improves performance while not (meaningfully) altering behaviour for usersThis improves performance while not (meaningfully) altering behaviour for users

Status: Open.

#757 In PyPSA/linopy;

· FBumann opened on Jun 4, 2026