v2.6.2

Increased performance of "Propensities" collective variable via numba.

v2.6.1

Fixed bugs:

• For slow CNVMs, the number of snapshots in the output could be less than the specified len_output. This is fixed by supplementing the snapshots with duplicates.

v2.6.0

new features:

• function to sample states given target opinion shares (sponet.states.sample_states_target_shares)
• pair approximation for the CNVM with 2 opinions

v2.5.1

Patch: new input argument max_chunk_size to reduce memory consumption of estimating the steady state

v2.5.0

Added function that automatically samples the collective variable steady state.

v2.4.0

What's Changed

• Optimize datatypes of network states by @manuelosos in #5

v2.3.4

Support for Python 3.12

v2.3.3

Small patch that improves the printing of the progression in sample_many_runs.

v2.3.2

Fixes:

• If len_output=None in a simulate call, the output trajectory should not contain subsequent identical states. This is fixed in this patch.

v2.3.1

Patch:

• also print remaining time in sample_many_runs when verbose=True