For a 0.14.0 release #362
Merged
For a 0.14.0 release #362
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Tweak behaviour of localmodeltypes to resolve #356
* Add a check to enuser that fitted and transforming features have the same category levels in OneHotEncoder * Add a test case to check transform function throw error when detect level mismatch * Add a test case for `ingore` parameter that was missin
Code re-organization and registry improvements
Check onehotencoding category level
Update model registry
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
(breaking) Change
@loadto return a model type, instead ofmodel instance. New usage example:
Tree = @load DecisionTreeClassifier; tree = Tree()or, in one line,tree = (@load DecisionTreeClassifier)(). Improve the reliability of@loadwhen called outside global scope (e.g., from function) andeliminate need for
scope=...flag. Drop experimental@loadcodemacro as essentially redundant (Change @load to return a type instead of instance. Addition of interactive version, @iload #358 @ablaom)
(experimental) Add interactive version
@load, called@iload,to prompt user for permission to automatically add packages to
user's current environment (automatic in non-interactive version
with
add=trueflag) and to display choices for models withambiguous names (resolved with
pkg=...flag in non-interactivecase) (Add an interactive version of @load #348 Change @load to return a type instead of instance. Addition of interactive version, @iload #358 @ablaom)
(new models) Add
MultitargetLinearRegressorandMultitargetRidgeRegressorfromMultivariatStatsto the registry(multitarget functionality was previously bundled into the models
LinearRegressorandRidgeRegressorin versions ofMLJMultivariateStatsInterfacepreceding 0.2). This will bebreaking for existing users who use
LinearRegressorandRidgeRegressoron multitarget problems.(new models) Add the new models
MultitargetKNNClassifierandMultitargetKNNRegressorprovided byNearestNeighborModelsto themodel registry (see below)
(code re-organization) Drop NearestNeighbors interface (last
Requires.jl-lazily loaded interface) in favour of registration of
new package
NearestNeighborModels.jl. This
is mildly breaking for general users who refer explicitly to the
package_nameorload_pathtraits ofKNNRegressororKNNClassifier(or corresponding entries in the model registry) asthese have changed to reflect the change. It will also break some
users' environments, generally resolved by adding
NearestNeighborModels(Disintegration of MLJModels #244, @OkonSamuel @ablaom).(improvement) Have
OneHotEncodercheck that each categoricalcolumn in the input passed to
transformhas a pool matching thatencountered during
fit(Check onehotencoding category level #360 @i-aki-y)