Movatterモバイル変換

external CPP pointer that should be translated from Cpp to an Robject

A 'forestry' object.

Number of new trees to add

A data frame of all training predictors.

A vector of all training responses.

The size of total samples to draw for the training data.

Maximum iterations/epochs per configuration. Default is 1024.

Downsampling rate. Default value is 2.

if tuning process in verbose mode

random seed

Number of threads to train and predict theforest. The defaultnumber is 0 which represents using all cores.

parameters which are passed directly to 'autoforestry'

A 'forestry' object.

A data frame of testing predictors.

A string denoting the dimension for computing lp distances.

A positive real number determining the norm p-norm used.

a forestry object

A data frame of all training predictors.

A vector of all training responses.

The number of trees to grow in the forest. The default value is500.

An indicator of whether sampling of training data is withreplacement. The default value is TRUE.

The size of total samples to draw for the training data. Ifsampling with replacement, the default value is the length of the trainingdata. If samplying without replacement, the default value is two-third ofthe length of the training data.

if this is given, then sampsize is ignored and set tobe round(length(y) * sample.fraction). It must be a real number between 0and 1

The number of variables randomly selected at each split point.The default value is set to be one third of total number of features of thetraining data.

Minimum observations contained in terminal nodes. Thedefault value is 3.

Minimum size of terminal nodes for averaging dataset. Thedefault value is 3.

Minimum observations to follow strictly in terminalnodes. The default value is 1.

Minimum size of terminal nodes for averaging datasetto follow strictly. The default value is 1.

Minimum loss reduction to split a node further in a tree.

Maximum depth of a tree. The default value is 99.

All splits at or above interaction depth must be on variablesthat are not weighting variables (as provided by the interactionVariables argument)

Indices of weighting variables.

(optional) vector of sampling probablities/weights foreach feature used when subsampling mtry features at each node above or atinteractionDepth. The default is to use uniform probabilities.

used in place of featureWeights for splits below interactionDepth.

These denote the weights for each training observationwhich determines how likely the observation is to be selected in each bootstrapsample. This option is not allowed when sampling is done without replacement.

Proportion of the training data used as the splittingdataset. It is a ratio between 0 and 1. If the ratio is 1, then essentiallysplitting dataset becomes the total entire sampled set and the averagingdataset is empty. If the ratio is 0, then the splitting data set is emptyand all the data is used for the averaging data set (This is not a goodusage however since there will be no data available for splitting).

random seed

if training process in verbose mode

Number of threads to train and predict the forest. The defaultnumber is 0 which represents using all cores.

only variance is implemented at this point and it containsspecifies the loss function according to which the splits of random forestshould be made

if the split value is taking the average of two featurevalues. If false, it will take a point based on a uniform distributionbetween two feature values. (Default = FALSE)

The max number of observations to split on

Fit the model with a ridge regression or not

Specify which features to split linearly on when usinglinear (defaults to use all numerical features)

Specifies monotonic relationships between thecontinuous features and the outcome. Supplied as a vector of length p withentries in 1,0,-1 which 1 indicating an increasing monotonic relationship,-1 indicating a decreasing monotonic relationship, and 0 indicating norelationship. Constraints supplied for categorical will be ignored.

Value to determine how much to penalize magnitude ofcoefficients in ridge regression

if the number of tree is doubled as averaging and splittingdata can be exchanged to create decorrelated trees. (Default = FALSE)

pass in an 'forestry' object which will recycle thedataframe the old object created. It will save some space working on thesame dataset.

If TRUE, then RF is created in such a way that it can besaved and loaded using save(...) and load(...). Setting it to TRUE(default) will, however, take longer and it will use more memory. Whentraining many RF, it makes a lot of sense to set this to FALSE to savetime and memory.

deprecated. Do not use.

A 'forestry' object.

flag to not display warnings

A trained model object of class "forestry".

Flag to not display warnings.

A 'forestry' object.

flag to not display warnings

parameters which are passed directly to 'forestry'

an object of class 'forestry'

the feature data.frame we will impute

a random seed passed to the predict method of forestry

if TRUE, mean imputation (for numericvariables) and mode imputation (for factor variables) is used for missingfeatures for which all neighbors also had the corresponding featuremissing; if FALSE these missing features remain as NAs in the data framereturned by 'impute_features'.

a filename in which to store the 'forestry' object

an object of class 'forestry'

A data frame of all training predictors.

A vector of all training responses.

The number of trees to grow in the forest. The default value is500.

Number of iterations used for gradient boosting.

Step size shrinkage used in gradient boosting update.

An indicator of whether sampling of training data is withreplacement. The default value is TRUE.

The size of total samples to draw for the training data. Ifsampling with replacement, the default value is the length of the trainingdata. If samplying without replacement, the default value is two-third ofthe length of the training data.

if this is given, then sampsize is ignored and set tobe round(length(y) * sample.fraction). It must be a real number between 0and 1

The number of variables randomly selected at each split point.The default value is set to be one third of total number of features of thetraining data.

Minimum observations contained in terminal nodes. Thedefault value is 3.

Minimum size of terminal nodes for averaging dataset. Thedefault value is 3.

Minimum observations to follow strictly in terminalnodes. The default value is 1.

Minimum size of terminal nodes for averaging datasetto follow strictly. The default value is 1.

Minimum loss reduction to split a node further in a tree.

Maximum depth of a tree. The default value is 99.

Proportion of the training data used as the splittingdataset. It is a ratio between 0 and 1. If the ratio is 1, then essentiallysplitting dataset becomes the total entire sampled set and the averagingdataset is empty. If the ratio is 0, then the splitting data set is emptyand all the data is used for the averaging data set (This is not a goodusage however since there will be no data available for splitting).

random seed

if training process in verbose mode

Number of threads to train and predict the forest. The defaultnumber is 0 which represents using all cores.

only variance is implemented at this point and it containsspecifies the loss function according to which the splits of random forestshould be made

if the split value is taking the average of two featurevalues. If false, it will take a point based on a uniform distributionbetween two feature values. (Default = FALSE)

The max number of observations to split on

Fit the model with a ridge regression or not

Specify which features to split linearly on when usinglinear (defaults to use all numerical features)

Specifies monotonic relationships between thecontinuous features and the outcome. Supplied as a vector of length p withentries in 1,0,-1 which 1 indicating an increasing monotonic relationship,-1 indicating a decreasing monotonic relationship, and 0 indicating norelationship. Constraints supplied for categorical will be ignored.

weights used when subsampling features for nodes above or at interactionDepth.

weights used when subsampling features for nodes below interactionDepth.

These denote the weights for each training observationwhich determines how likely the observation is to be selected in each bootstrapsample. This option is not allowed when sampling is done without replacement.

Value to determine how much to penalize magnitude ofcoefficients in ridge regression

if the number of tree is doubled as averaging and splittingdata can be exchanged to create decorrelated trees. (Default = FALSE)

pass in an 'forestry' object which will recycle thedataframe the old object created. It will save some space working on thesame dataset.

If TRUE, then RF is created in such a way that it can besaved and loaded using save(...) and load(...). Setting it to TRUE(default) will, however, take longer and it will use more memory. Whentraining many RF, it makes a lot of sense to set this to FALSE to savetime and memory.

deprecated. Do not use.

A forestry x.

Specifies the tree number that should be visulaized.

Should the data for the plot be printed?

The length of the beta values in leaf noderepresentation.

additional arguments that are not used.

A 'forestry' object.

A data frame of testing predictors.

How the individual tree predictions are aggregated:'average' returns the mean of all trees in the forest; 'weightMatrix'returns a list consisting of "weightMatrix", the adaptive nearest neighborweights used to construct the predictions; "terminalNodes", a matrix wherethe ith entry of the jth column is the index of the leaf node to which theith observation is assigned in the jth tree; and "sparse", a matrixwhere the ith entry in the jth column is 1 if the ith observation infeature.new is assigned to the jth leaf and 0 otherwise. In each tree theleaves are indexed using a depth first ordering, and, in the "sparse"representation, the first leaf in the second tree has column index one more thanthe number of leaves in the first tree and so on. So, for example, if thefirst tree has 5 leaves, the sixth column of the "sparse" matrix correspondsto the first leaf in the second tree.

random seed

additional arguments.

A 'multilayerForestry' object.

A data frame of testing predictors.

How shall the leaf be aggregated. The default is to returnthe mean of the leave 'average'. Other options are 'weightMatrix'.

random seed

additional arguments.

A data frame of all training predictors.

A list of index for all categorical data. Usedfor trees to detect categorical columns.

A list of encoding details for eachcategorical column, including all unique factor values and theircorresponding numeric representation.

A data frame of all training predictors.

A vector of all training responses.

an object of class 'forestry' or class 'multilayerForestry'

an object of class 'forestry'

a filename in which to store the 'forestry' object

additional arguments useful for specifying compression type and level

A forestry object.

A data frame of testing predictors.

TRUE if the there were NAs in the training data FALSE otherwise.

A data frame of all training predictors.

A vector of all training responses.

The number of trees to grow in the forest. The default value is500.

An indicator of whether sampling of training data is withreplacement. The default value is TRUE.

The size of total samples to draw for the training data. Ifsampling with replacement, the default value is the length of the trainingdata. If samplying without replacement, the default value is two-third ofthe length of the training data.

The number of variables randomly selected at each split point.The default value is set to be one third of total number of features of thetraining data.

Minimum observations contained in terminal nodes. Thedefault value is 3.

Minimum size of terminal nodes for averaging dataset. Thedefault value is 3.

Minimum observations to follow strictly in terminalnodes. The default value is 1.

Minimum size of terminal nodes for averaging datasetto follow strictly. The default value is 1.

Minimum loss reduction to split a node further in a tree.

Maximum depth of a tree. The default value is 99.

All splits at or above interaction depth must be on variablesthat are not weighting variables (as provided by the interactionVariables argument)

Proportion of the training data used as the splittingdataset. It is a ratio between 0 and 1. If the ratio is 1, then essentiallysplitting dataset becomes the total entire sampled set and the averagingdataset is empty. If the ratio is 0, then the splitting data set is emptyand all the data is used for the averaging data set (This is not a goodusage however since there will be no data available for splitting).

Number of threads to train and predict the forest. The defaultnumber is 0 which represents using all cores.

if the split value is taking the average of two featurevalues. If false, it will take a point based on a uniform distributionbetween two feature values. (Default = FALSE)

if the number of tree is doubled as averaging and splittingdata can be exchanged to create decorrelated trees. (Default = FALSE)

Specify which features to split linearly on when usinglinear (defaults to use all numerical features)

Specifies monotonic relationships between thecontinuous features and the outcome. Supplied as a vector of length p withentries in 1,0,-1 which 1 indicating an increasing monotonic relationship,-1 indicating a decreasing monotonic relationship, and 0 indicating norelationship. Constraints supplied for categorical will be ignored.

weights used when subsampling features for nodes above or at interactionDepth.

weights used when subsampling features for nodes below interactionDepth.

These denote the weights for each training observationwhich determines how likely the observation is to be selected in each bootstrapsample. This option is not allowed when sampling is done without replacement.

Fit the model with a ridge regression or not

indicates if there is any missingness in x.