0.2.2
Features
-
An
allow_unary
flag (False by default
) has been added to all methods. -
A
set_metadata
flag has been added so that node and mutation metadata can be
omitted, saved (default), or overwritten even if this requires changing the schema. -
An environment variable
TSDATE_ENABLE_NUMBA_CACHE
can be set to cache JIT
compiled code, speeding up loading time (useful when testing). -
The time taken for running tsdate is now recorded in the provenance data
-
Historical samples (sample nodes with time > 0) are now allowed in the
tree sequence.
Documentation
- Various fixes in documentation, including documenting returned fits.
Breaking changes
-
The
return_posteriors
argument has been removed and replaced withreturn_fit
.
An instance of one of two previously internal classes,ExpectationPropagation
andBeliefPropagation
, are now returned whenreturn_fit=True
, and posteriors can
be obtained usingfit.node_posteriors()
. -
Topology-only dating (setting
mutation_rate=None
) has been removed for tree sequences
of more than one tree, as tests have found that span-weighting the conditional coalescent
causes substantial bias. -
The
trim_telomeres
parameter in thetsdate.preprocess_ts()
function has been renamed
toerase_flanks
, to matchtinfer.preprocess()
. The previous name is kept as a
deprecated alias.