Project: Project Class
In five-dots/ml4e: Machine Learning for Everyone
Description
Usage
Arguments
Value
Super class
Active bindings
Methods
Description
A class defines a project specification.
Usage
1
2
3
4
5
6
7
8
9
10
Arguments
formulas
A Formulas class object.
datasets
A Datasets class object.
seeds
A Seeds class object.
models
A Models class object.
measure
A Measure class object. If NULL, a Measure class
object with default metrics are used.
show_progress
A logical scalar wheather to show a progress bar.
keep_data
A logical scalar wheather to keep original data.
If TRUE, the original dataset are kept in predicted values and
cross-validation folds.
parallel
A logical scalar wheather to calculate parallely.
Value
A new Project class object
Super class
ml4e::Options -> Project
Active bindings
cv_tableA cross-validation table.
grid_tableA grid search table.
random_tableA random search table.
bayes_tablesA list of bayesian search tables by metric_name.
Methods
Public methods
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Inherited methods
Method new()
Construct a new Project class object.
Usage
Project$new(
formulas = NULL,
datasets = NULL,
seeds = NULL,
models = NULL,
measure = NULL,
show_progress = TRUE,
keep_data = FALSE,
parallel = FALSE
)
Arguments
formulasA Formulas class object.
datasetsA Datasets class object.
seedsA Seeds class object.
modelsA Models class object.
measureA Measure class object. If NULL, a Measure class
object with default metrics are used.
show_progressA logical scalar wheather to show a progress bar.
keep_dataA logical scalar wheather to keep original data.
If TRUE, the original dataset are kept in predicted values and
cross-validation folds.
parallelA logical scalar wheather to calculate parallely.
Returns
A new Project class object
Method print()
Print object.
Usage
Project$print()
Returns
A Project object by invisible(self).
Method run_cv()
Run cross-validation by each key combinations.
Usage
Project$run_cv(...)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
Returns
A Project object by invisible(self).
Method run_grid()
Run grid search for parameters set by each key combinations.
Usage
Project$run_grid(..., num = NULL)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
numA integer scalar of how many parameters to be calculated.
If NULL, all parameters in param set will be calculated.
Returns
A Project object by invisible(self).
Method run_random()
Run random search for parameters set by each key combinations.
Usage
Project$run_random(..., num = NULL)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
numA integer scalar of how many parameters to be calculated.
Returns
A Project object by invisible(self).
Method run_bayes()
Run bayes search for parameters set by each key combinations.
Usage
Project$run_bayes(
...,
metric_name = NULL,
init_points = 4L,
n_iter = 5L,
acq = "ucb",
kappa = 2.576,
eps = 0,
kernel = list(type = "exponential", power = 2L)
)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
metric_nameA character scalar of metric name. If NULL, the
first metric_name in self$measure$metric_names is used.
init_pointsA integer scalar of init_points.
n_iterA integer scalar of n_iter.
acqA character scalar of aquisition function. Can be "ucb", "ei"
or "poi".
kappaA numeric scalar of kappa.
epsA numeric scalar of eps.
kernelA list of kernal parameters.
Returns
A Project object by invisible(self).
Method get_scores()
Get cross-validation scores.
Usage
Project$get_scores(..., simplify = TRUE)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
simplifyA logical scalar wheater to drop columns of a signle key.
Returns
A data.frame class object.
Method get_preds()
Get predictions.
Usage
Project$get_preds(...)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
Returns
A list of predictions by keys
Method get_stacking_data()
Get stacking data.
Usage
Project$get_stacking_data(..., prob = FALSE)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
probA logical scalar wheather to use probability data.
Returns
A data.frame of stacking data.
Method get_search_result()
Get merged search result.
Usage
Project$get_search_result(...)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
Returns
A data.frame of search result.
Method get_search_result_by_model()
Get merged search result by model key.
Usage
Project$get_search_result_by_model(model_key = NULL)
Arguments
model_keyA character scalar of model key.
Returns
A data.frame of search result.
Method get_ranks()
Get paramter ranking.
Usage
Project$get_ranks(..., metric_name = NULL, n = 5L)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
metric_nameA character scalar of metric name. If NULL, the
first metric_name in self$measure$metric_names is used.
nA integer scalar of a number of the ranking.
Returns
A list of ranking by keys
Method print_ranks()
Print parameters ranking.
Usage
Project$print_ranks(..., metric_name = NULL, n = 5L)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
metric_nameA character scalar of metric name. If NULL, the first
metric_name in self$measure$metric_names is used.
nA integer scalar of a number of the ranking.
Returns
A Project object by invisible(self).
Method get_best_params()
Get best parameters by grid/bayes search.
Usage
Project$get_best_params(..., metric_name = NULL)
Arguments
...Filtering expressions passed to dplyr::filter or
dplyr::slice.
metric_nameA character scalar of metric name. If NULL, the
first metric_name in self$measure$metric_names is used.
Returns
A list of best parameters by keys
Method clone()
The objects of this class are cloneable with this method.
Usage
Project$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
five-dots/ml4e documentation built on June 19, 2020, 4:26 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Description Usage Arguments Value Super class Active bindings Methods
Description
A class defines a project specification.
Usage
1 2 3 4 5 6 7 8 9 10 |
Arguments
formulas |
A Formulas class object. |
datasets |
A Datasets class object. |
seeds |
A Seeds class object. |
models |
A Models class object. |
measure |
A Measure class object. If |
show_progress |
A logical scalar wheather to show a progress bar. |
keep_data |
A logical scalar wheather to keep original data.
If |
parallel |
A logical scalar wheather to calculate parallely. |
Value
A new Project class object
Super class
ml4e::Options -> Project
Active bindings
cv_tableA cross-validation table.
grid_tableA grid search table.
random_tableA random search table.
bayes_tablesA list of bayesian search tables by
metric_name.
Methods
Public methods
Inherited methods
Method new()
Construct a new Project class object.
Usage
Project$new( formulas = NULL, datasets = NULL, seeds = NULL, models = NULL, measure = NULL, show_progress = TRUE, keep_data = FALSE, parallel = FALSE )
Arguments
formulasA Formulas class object.
datasetsA Datasets class object.
seedsA Seeds class object.
modelsA Models class object.
measureA Measure class object. If
NULL, a Measure class object with default metrics are used.show_progressA logical scalar wheather to show a progress bar.
keep_dataA logical scalar wheather to keep original data. If
TRUE, the original dataset are kept in predicted values and cross-validation folds.parallelA logical scalar wheather to calculate parallely.
Returns
A new Project class object
Method print()
Print object.
Usage
Project$print()
Returns
A Project object by invisible(self).
Method run_cv()
Run cross-validation by each key combinations.
Usage
Project$run_cv(...)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
Returns
A Project object by invisible(self).
Method run_grid()
Run grid search for parameters set by each key combinations.
Usage
Project$run_grid(..., num = NULL)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
numA integer scalar of how many parameters to be calculated. If
NULL, all parameters in param set will be calculated.
Returns
A Project object by invisible(self).
Method run_random()
Run random search for parameters set by each key combinations.
Usage
Project$run_random(..., num = NULL)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
numA integer scalar of how many parameters to be calculated.
Returns
A Project object by invisible(self).
Method run_bayes()
Run bayes search for parameters set by each key combinations.
Usage
Project$run_bayes( ..., metric_name = NULL, init_points = 4L, n_iter = 5L, acq = "ucb", kappa = 2.576, eps = 0, kernel = list(type = "exponential", power = 2L) )
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
metric_nameA character scalar of metric name. If
NULL, the first metric_name inself$measure$metric_namesis used.init_pointsA integer scalar of
init_points.n_iterA integer scalar of
n_iter.acqA character scalar of aquisition function. Can be "ucb", "ei" or "poi".
kappaA numeric scalar of
kappa.epsA numeric scalar of
eps.kernelA list of kernal parameters.
Returns
A Project object by invisible(self).
Method get_scores()
Get cross-validation scores.
Usage
Project$get_scores(..., simplify = TRUE)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
simplifyA logical scalar wheater to drop columns of a signle key.
Returns
A data.frame class object.
Method get_preds()
Get predictions.
Usage
Project$get_preds(...)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
Returns
A list of predictions by keys
Method get_stacking_data()
Get stacking data.
Usage
Project$get_stacking_data(..., prob = FALSE)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
probA logical scalar wheather to use probability data.
Returns
A data.frame of stacking data.
Method get_search_result()
Get merged search result.
Usage
Project$get_search_result(...)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
Returns
A data.frame of search result.
Method get_search_result_by_model()
Get merged search result by model key.
Usage
Project$get_search_result_by_model(model_key = NULL)
Arguments
model_keyA character scalar of model key.
Returns
A data.frame of search result.
Method get_ranks()
Get paramter ranking.
Usage
Project$get_ranks(..., metric_name = NULL, n = 5L)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
metric_nameA character scalar of metric name. If
NULL, the first metric_name inself$measure$metric_namesis used.nA integer scalar of a number of the ranking.
Returns
A list of ranking by keys
Method print_ranks()
Print parameters ranking.
Usage
Project$print_ranks(..., metric_name = NULL, n = 5L)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
metric_nameA character scalar of metric name. If NULL, the first metric_name in
self$measure$metric_namesis used.nA integer scalar of a number of the ranking.
Returns
A Project object by invisible(self).
Method get_best_params()
Get best parameters by grid/bayes search.
Usage
Project$get_best_params(..., metric_name = NULL)
Arguments
...Filtering expressions passed to dplyr::filter or dplyr::slice.
metric_nameA character scalar of metric name. If
NULL, the first metric_name inself$measure$metric_namesis used.
Returns
A list of best parameters by keys
Method clone()
The objects of this class are cloneable with this method.
Usage
Project$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.