Releases: microsoft/SparseSC
Releases · microsoft/SparseSC
v0.2.0
- Added tools too to use
fit_fast()
with large datasets. This includes thesample_frac
option toMTLassoCV_MatchSpace_factory()
to estimate the match space on a subset of the observations. It also includesfit_fast()
optionavoid_NxN_mats
which will avoid making large matrices (at the expense of only returning the Synthetic controltargets
andtargets_aux
and not the full weight matrix) - Added logging in
fit_fast
via theverbose
numerical option. This can help identify out-of-memory errors. - Added a pseudo-Doubly robust match space maker
D_LassoCV_MatchSpace_factory
. It apporptions some of the normalized variable V weight to those variables that are good predictors of treatment. This should only be done if there are many treated units so that one can reasonably model this relationship. - Extracted the Azure Batch configuration to a dedicated Batch-Config package
v0.1.0
Initial release.