Efficient, object-oriented programming on the building blocks of machine learning. Successor of mlr.

• We started writing a book manual, but it is still in early stages.
• Reference Manual
• Extension packages
• useR!2019 talks
• Blog about mlr and mlr3

remotes::install_github("mlr-org/mlr3")

library(mlr3)
set.seed(1)

# create learning task
task_iris = TaskClassif$new(id = "iris", backend = iris, target = "Species")
task_iris

## <TaskClassif:iris> (150 x 5)
## * Target: Species
## * Properties: multiclass
## * Features (4):
##   - dbl (4): Petal.Length, Petal.Width, Sepal.Length, Sepal.Width

# load learner and set hyperparamter
learner = lrn("classif.rpart", cp = 0.01)

# train/test split
train_set = sample(task_iris$nrow, 0.8 * task_iris$nrow)
test_set = setdiff(seq_len(task_iris$nrow), train_set)

# train the model
learner$train(task_iris, row_ids = train_set)

# predict data
prediction = learner$predict(task_iris, row_ids = test_set)

# calculate performance
prediction$confusion

##             truth
## response     setosa versicolor virginica
##   setosa         11          0         0
##   versicolor      0         12         1
##   virginica       0          0         6

measure = msr("classif.acc")
prediction$score(measure)

## classif.acc 
##   0.9666667

# automatic resampling
resampling = rsmp("cv", folds = 3L)
rr = resample(task_iris, learner, resampling)

## INFO  [21:11:08.322] Applying learner 'classif.rpart' on task 'iris' (iter 1/3) 
## INFO  [21:11:08.350] Applying learner 'classif.rpart' on task 'iris' (iter 2/3) 
## INFO  [21:11:08.364] Applying learner 'classif.rpart' on task 'iris' (iter 3/3)

rr$score(measure)

##             task task_id               learner    learner_id
##           <list>  <char>                <list>        <char>
## 1: <TaskClassif>    iris <LearnerClassifRpart> classif.rpart
## 2: <TaskClassif>    iris <LearnerClassifRpart> classif.rpart
## 3: <TaskClassif>    iris <LearnerClassifRpart> classif.rpart
##        resampling resampling_id iteration prediction classif.acc
##            <list>        <char>     <int>     <list>       <num>
## 1: <ResamplingCV>            cv         1     <list>        0.92
## 2: <ResamplingCV>            cv         2     <list>        0.92
## 3: <ResamplingCV>            cv         3     <list>        0.94

rr$aggregate(measure)

## classif.acc 
##   0.9266667

mlr was first released to CRAN in 2013. Its core design and architecture date back even further. The addition of many features has led to a feature creep which makes mlr hard to maintain and hard to extend. We also think that while mlr was nicely extensible in some parts (learners, measures, etc.), other parts were less easy to extend from the outside. Also, many helpful R libraries did not exist at the time mlr was created, and their inclusion would result in non-trivial API changes.

• Only the basic building blocks for machine learning are implemented in this package.
• Focus on computation here. No visualization or other stuff. That can go in extra packages.
• Overcome the limitations of R’s S3 classes with the help of R6.
• Embrace R6, clean OO-design, object state-changes and reference semantics. This might be less “traditional R”, but seems to fit mlr nicely.
• Embrace data.table for fast and convenient data frame computations.
• Combine data.table and R6, for this we will make heavy use of list columns in data.tables.
• Be light on dependencies. mlr3 requires the following packages:
  • backports: Ensures backward compatibility with older R releases. Developed by members of the mlr team. No recursive dependencies.
  • checkmate: Fast argument checks. Developed by members of the mlr team. No extra recursive dependencies.
  • mlr3misc Miscellaneous functions used in multiple mlr3 extension packages. Developed by the mlr team. No extra recursive dependencies.
  • paradox: Descriptions for parameters and parameter sets. Developed by the mlr team. No extra recursive dependencies.
  • R6: Reference class objects. No recursive dependencies.
  • data.table: Extension of R’s data.frame. No recursive dependencies.
  • digest: Hash digests. No recursive dependencies.
  • lgr: Logging facility. No extra recursive dependencies.
  • Metrics: Package which implements performance measures. No recursive dependencies.
  • mlbench: A collection of machine learning data sets. No dependencies.
• Reflections: Objects are queryable for properties and capabilities, allowing you to program on them.
• Additional functionality that comes with extra dependencies:
  • For parallelization, mlr3 utilizes the future and future.apply packages.
  • To capture output, warnings and exceptions, evaluate and callr can be used.

mlr-outreach holds all outreach activities related to mlr and mlr3.

mlr3 talk at useR! 2019 conference in Toulouse, France: