Releases: msamsami/wnb
Releases · msamsami/wnb
v0.8.0
What's Changed
- feat: add support for laplace distribution by @msamsami in #47
- feat: add docstring examples, improve error weights and class prior initialization by @msamsami in #48
- feat: add
var_smoothing
parameter toGaussianWNB
by @msamsami in #49 - fix: make
distributions
param positional-or-keyword inGeneralNB
by @msamsami in #50
Full Changelog: v0.7.0...v0.8.0
🔧 GeneralNB's distributions
parameter is now positional-or-keyword, meaning that both styles below work:
clf = GeneralNB([D.NORMAL, D.POISSON])
clf = GeneralNB(distributions=[D.NORMAL, D.POISSON])
v0.7.0
What's Changed
- feat: add parameter constraints and validations, add
_fit_context
decorator by @msamsami in #44 - feat: add support for specifying distributions as a list of tuples by @msamsami in #45
- doc: minor update in getting started section of RAEDME by @msamsami in #46
🎉 Now you specify GeneralNB
's distributions as a list of tuples:
from wnb import GeneralNB, Distribution as D
gnb = GeneralNB(
distributions=[
(D.NORMAL, "col1"),
(D.BERNOULLI, "col2"),
(D.EXPONENTIAL, ["col3", "col4"])
]
)
Full Changelog: v0.6.0...v0.7.0