Tree Ensemble Learner

Options

Target column

Select the column containing the value to be learned. Rows with missing values in this column are ignored during the learning process.

Training attributes

Choose whether to derive attributes from a fingerprint vector column or from ordinary table columns.

• Use column: Select ordinary columns and configure include/exclude rules for training attributes.
• Use fingerprint: Expand a fingerprint (bit, byte, or double vector) column into individual attributes during training.

Fingerprint attribute

Use a fingerprint (bit, byte, or double vector) column to learn the model. Each entry of the vector is treated as a separate attribute. All vectors must share the same length.

Attribute selection

Ignore columns without domain information

If selected, nominal columns with no domain information are ignored (as they likely have too many possible values anyway).

Split criterion

Choose the split criterion here. Gini is usually a good choice and is used in "Classification and Regression Trees" (Breiman et al, 1984) and the original random forest algorithm (as described by Breiman et al, 2001); information gain is used in C4.5; the information gain ratio normalizes the standard information gain by the split entropy to overcome any unfair preference for nominal splits with many child nodes.

• Information Gain: Maximizes the information gain when creating splits. Equivalent to entropy reduction.
• Information Gain Ratio: Uses the information gain ratio to compensate for the bias towards attributes with many distinct values.
• Gini: Measures the impurity of a split based on the Gini index. Common default for classification tasks.

Use mid-point splits

For numerical splits, use the mid-point between two class boundaries. Otherwise the split value corresponds to the lower boundary with a ≤ comparison.

Use binary splits for nominal columns

If selected, nominal columns also produce binary splits instead of multiway splits in which each nominal value corresponds to one child node.

Limit number of levels (tree depth)

Limit the maximal number of tree levels. When disabled the tree depth is unbounded. For instance, a value of 1 would only split the (single) root node resulting in a decision stump

Minimum split node size

Minimum number of records in a decision tree node so that another split is attempted. Note, this option does not make any implications on the minimum number of records in a terminal node. If enabled, this number needs to be at least twice as large as the minimum child node size (as otherwise for binary splits one of the two children would have less records than specified).

Minimum child node size

Minimum number of records allowed in the child nodes after a split. Must not exceed half the minimum split node size. Note, this parameter is currently ignored for nominal columns if binary nominal splits are disabled.

Use fixed root attribute

Force the selected column to be used as the root split attribute in all trees -- even if the column is not in the attribute sample.

Enable row sampling

Enable sampling of training rows for each individual tree. If disabled, each tree learner gets the full dataset, otherwise each tree is learned with a different data sample. A data fraction of 1 (=100%) chosen "with replacement" is called bootstrapping. For sufficiently large datasets this bootstrap sample contains about 2/3 different data rows from the input, some of which are replicated multiple times. Rows that are not used in the training of a tree are called out-of-bag.

Fraction of data to learn single model

Fraction of data rows to sample for learning each individual tree. A value of 1 means 100% of the data rows are sampled.

Sample with replacement

Draw sampled rows with replacement (bootstrap sampling). When disabled, rows are sampled without replacement.

Data sampling mode

The sampling mode decides how the rows are sampled. In the random mode, the rows are sampled from the whole dataset, i.e. each row has exactly the same probability as in the sample. In case of equal size sampling, first a sample from the minority class is drawn and then the same number of rows as in the minority sample are drawn from all other classes so that each class is represented with the same number of rows in the sample. If stratified sampling is selected, the same fraction of rows is drawn from each class so the class distribution in the sample is approximately the same as in the full dataset.

• Random: Sample rows uniformly at random.
• Stratified: Sample rows stratified by the target class distribution.
• Equal size: Sample the same number of rows for each class. Available for classification tasks.

Number of models

Number of decision trees to learn. Larger ensembles generally provide more stable results but increase runtime. For most datasets, a value between 100 and 500 yields good results; however, the optimal number is data dependent and should thus be subject to hyperparameter tuning.

Attribute sampling (columns)

Defines the sampling of attributes to learn an individual tree. This can either be a function based on the number of attributes (linear fraction or square root) or some absolute value. The latter can be used in conjunction with flow variables to inject some other value derived from the number of attributes (e.g. Breiman suggests starting with the square root of the number of attributes but also to try to double or half that number).

• All columns: Disable column sampling and use all available attributes for every tree.
• Square root: Sample the square root of the number of available attributes for each tree (default random forest behaviour).
• Linear fraction: Sample a fraction of the available attributes for each tree.
• Absolute number: Sample a fixed number of attributes for each tree.

Linear fraction

Fraction of attributes to sample for each tree.

Absolute number

Number of attributes to sample for each tree.

Attribute selection

Use the same set of attributes for each tree means that the attributes are sampled once for each tree and this sample is then used to construct the tree.

Use a 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. This is the option used in random forests.

Use static random seed

Provide a seed to obtain deterministic results. Leave disabled to use a time-dependent seed.

Generate new seed

Generate a random seed and apply it to the field above for reproducible runs.

Enable highlighting

If selected, the node stores the configured number of rows and allows highlighting them in the node view.

Save target distribution in tree nodes

If selected, the model stores the distribution of the target category values in each tree node. Storing the class distribution may increase memory consumption considerably and we therefore recommend disabling it if your use-case doesn't require it. Class distribution is only needed if

• You want to see the class distribution for each tree node in the node view.
• You want to export individual decision trees to PMML.
• You want to use soft-voting (i.e. aggregation of probability distributions instead of votes) in the predictor node.

Views

Tree Views

A decision tree viewer for all the trained models. Use the spinner to iterate through the different models.

Workflows

• _Learner_FlowKNIME Hub
• _Learner_FlowKNIME Hub
• 01_Guided_Analytics_for_ML_AutomationKNIME Hub
• 01_Guided_Analytics_for_ML_AutomationKNIME Hub
• 01_Guided_Analytics_for_ML_AutomationKNIME Hub
• Show all 113 workflows

Links

• No links available

Developers