The following node is available in the Open Source KNIME predictive analytics and data mining platform version 2.7.1. Discover over 1000 other nodes, as well as enterprise functionality at http://knime.com.
Learns an ensemble of regression trees (such as random forest variants). Each of the regression tree models is learned on a different set of rows (records) and/or a different set of columns (describing attributes), whereby the latter can also be a bit-vector descriptor (e.g. molecular fingerprint). The output model describes an ensemble of regression tree models and is applied in the corresponding predictor node using a simply mean of the individual predictions.
For a more general description and suggested default parameters see the node description of the classification Tree Ensemble Learner.
Note: The node is currently under development; models used with the current version are likely not backward compatible.
Select the attributes to use learn the model. Two variants are possible.
Fingerprint attribute uses the different bit positions in the selected bit vector as learning attributes (for instance a bit vector of length 1024 is expanded to 1024 binary attributes). All bit vectors in the selected column must have the same length.
Column attributes are nominal and numeric columns used as descriptors. Numeric columns are split in a <= fashion; nominal columns are currently split by creating child nodes for each of the values.
Use same set of attributes for each tree describes that the attributes are sampled once for each tree and this sample is then used to construct the tree.
Use different set of attributes for each tree node samples a different set of candidate attributes in each of the tree nodes from which the optimal one is chosen to perform the split.
0 | The data to learn from. It must contain at least one numeric target column and either a fingerprint (bitvector) column or another numeric or nominal column. |
0 | The input data with the out-of-bag response estimates, i.e. for each input row the mean output of all models that did not use the row in the training. If the entire data was used to train the individual models then this output will contain the input data with missing response and response variance values. The appended columns are equivalent to the columns appended by the corresponding predictor node. There is one additional column model count, which contains the number of models used for the voting (number of models not using the row throughout the learning.) |
1 | A statistics table on the attributes used in the different tree learners. Each row represents one training attribute with these statistics: #splits (level x) as the number of models, which use the attribute as split on level x (with level 0 as root split); #candidates (level x) is the number of times an attribute was in the attribute sample for level x (in a random forest setup these samples differ from node to node). If no attribute sampling is used #candidates is the number of models. Note, these numbers are uncorrected, i.e. if an attribute is selected on level 0 but is also in the candidate set of level 1 (but will not be split on level 1 because it has been split one level up), the #candidate number will still count the attribute as candidate. |
2 | The trained model. |