v1.1.0

ISOKANN v1.1.0

Diff since v1.0.0

Merged pull requests:

• Girsanov 2 (#37) (@axsk)
• fix negative steps in reactionpath (#38) (@MrKresse)

Closed issues:

• IsoForce seems to be bugged for T != 1 (#3)
• replace vec_to_coords by reinterpret (#5)
• Girsanov for Overdamped Langevin (#6)
• Alignment of molecules (#8)
• VGV adaptive sampling (#12)
• Clean up MOR (#13)
• Release 1.0 (#17)
• Change default parameters (#27)
• New Features systems (#28)
• Validation loss logger (#30)
• Saving and Loading state / OpenMMSimulation (#31)
• getxs, getcoords etc. inconsistencies (#39)
• Reactive Path uses unaligned distances (#41)
• Alignment is not working properly (#43)

v1.0.0

ISOKANN v1.0.0

Diff since v0.1.0

Merged pull requests:

• BioStructures.jl compat (#25) (@jgreener64)
• Fix: set_numpy_state missing positional argument (#26) (@joramkuntze)

v0.1.0

ISOKANN v0.1.0

Merged pull requests:

• Adding support for small molecule/protein system, adding options for waters/solvent, typo correction (#22) (@csecker)
• Making openff.toolkit and openmmforcefields imports conditional, fixing addwater for protein only system (#23) (@csecker)

Closed issues:

• Switch to using Flux.DataLoader (#2)
• Save momentum in Langevin simulations (#4)
• Change ys dimensions as to reflect batch in last dimension? (#7)
• Clean up VGV (#10)
• Look into save_reactive_coord (#11)
• add SimulationData type (#14)
• Remove the old IsoRun (#15)
• Move Molly support to Extension (#16)
• pairwise distance feature cutoff parameter (#20)

v0.1.0-rc

update README