ResultData
Internal object to store results in list of data.tables, arranged in a star schema. It is advised to not directly work on this data structure as it may be changed in the future without further warnings.
The main motivation of this data structure is the necessity to avoid storing duplicated R6 objects. While this is usually no problem in a single R session, serialization via serialize() (which is used in save()/saveRDS() or during parallelization) leads to objects with unreasonable memory requirements.
# table overview print(ResultData$new()$data)
data: (list())
List of `data.table::data.table()`, arranged in a star schema. Do not operate directly on this list.
task_type: (character(1))
Returns the task type of stored objects, e.g. `"classif"` or `"regr"`. Returns `NULL` if the ResultData is empty.
new()Creates a new instance of this R6 class. An alternative construction method is provided by as_result_data().
ResultData$new(data = NULL, data_extra = NULL, store_backends = TRUE)
data: (data.table::data.table()) | NULL)
Do not initialize this object yourself, use `as_result_data()` instead.
data_extra: (list())
Additional data to store. This can be used to store additional information for each iteration.
store_backends: (logical(1))
If set to `FALSE`, the backends of the Task s provided in `data` are removed.
uhashes()Returns all unique hashes (uhash values) of all included ResampleResult s.
ResultData$uhashes(view = NULL)
view: character(1)
Single `uhash` to restrict the results to.
character().
iterations()Returns the number of recorded iterations / experiments.
ResultData$iterations(view = NULL)
view: character(1)
Single `uhash` to restrict the results to.
integer(1).
tasks()Returns a table of included Task s.
ResultData$tasks(view = NULL)
view: character(1)
Single `uhash` to restrict the results to.
data.table() with columns "task_hash" (character()) and "task" (Task ).
learners()Returns a table of included Learner s.
ResultData$learners(view = NULL, states = TRUE, reassemble = TRUE)
view: character(1)
Single `uhash` to restrict the results to.
states: (logical(1))
If `TRUE`, returns a learner for each iteration/experiment in the ResultData object. If `FALSE`, returns an exemplary learner (without state) for each ResampleResult .
reassemble: (logical(1))
Reassemble the learners, i.e. re-set the `state` and the hyperparameters which are stored separately before returning the learners.
data.table() with columns "learner_hash" (character()) and "learner" (Learner ).
learner_states()Returns a list of states of included Learner s without reassembling the learners.
@return list of list()
ResultData$learner_states(view = NULL)
view: character(1)
Single `uhash` to restrict the results to.
resamplings()Returns a table of included Resampling s.
ResultData$resamplings(view = NULL)
view: character(1)
Single `uhash` to restrict the results to.
data.table() with columns "resampling_hash" (character()) and "resampling" (Resampling ).
predictions()Returns a list of Prediction objects.
ResultData$predictions(view = NULL, predict_sets = "test")
view: character(1)
Single `uhash` to restrict the results to.
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
list() of Prediction .
prediction()Returns a combined Prediction objects.
ResultData$prediction(view = NULL, predict_sets = "test")
view: character(1)
Single `uhash` to restrict the results to.
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
Prediction .
data_extra()Returns additional data stored.
ResultData$data_extra(view = NULL)
view: character(1)
Single `uhash` to restrict the results to.
data.table::data.table().
combine()Combines multiple ResultData objects, modifying self in-place.
ResultData$combine(rdata)
rdata: (ResultData ).
self (invisibly).
sweep()Updates the ResultData object, removing rows from all tables which are not referenced by the fact table anymore. E.g., can be called after filtering/subsetting the fact table.
ResultData$sweep()
Modified self (invisibly).
marshal()Marshals all stored learner models. This will do nothing to models that are already marshaled.
ResultData$marshal(...)
...: (any)
Additional arguments passed to `marshal_model()`.
unmarshal()Unmarshals all stored learner models. This will do nothing to models which are not marshaled.
ResultData$unmarshal(...)
...: (any)
Additional arguments passed to `unmarshal_model()`.
discard()Shrinks the object by discarding parts of the stored data.
ResultData$discard(backends = FALSE, models = FALSE)
backends: (logical(1))
If `TRUE`, the DataBackend is removed from all stored Task s.
models: (logical(1))
If `TRUE`, the stored model is removed from all Learner s.
Modified self (invisibly).
as_data_table()Combines internal tables into a single flat data.table().
ResultData$as_data_table(
view = NULL,
reassemble_learners = TRUE,
convert_predictions = TRUE,
predict_sets = "test"
)
view: character(1)
Single `uhash` to restrict the results to.
reassemble_learners: (logical(1))
Reassemble the tasks?
convert_predictions: (logical(1))
Convert PredictionData to Prediction ?
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
predict_sets: (character())
Prediction sets to operate on, used in `aggregate()` to extract the matching `predict_sets` from the ResampleResult . Multiple predict sets are calculated by the respective Learner during `resample()`/`benchmark()`. Must be a non-empty subset of `{"train", "test", "internal_valid"}`. If multiple sets are provided, these are first combined to a single prediction object. Default is `"test"`.
logs()Get a table of recorded learner logs.
ResultData$logs(view = NULL, condition)
view: character(1)
Single `uhash` to restrict the results to.
condition: (character(1)) The condition to extract. One of "message", "warning" or "error".
data.table::data.table().
clone()The objects of this class are cloneable with this method.
ResultData$clone(deep = FALSE)
deep: Whether to make a deep clone.
Useful links