Bootstrap Sampling

Samples the data using bootstrapping. Bootstrapping is a sampling technique, which randomly draws rows from the input with replacement. The output table will therefore likely contain duplicate rows while other rows are not present in the output at all.

Options

Output size type

Defines how the size of the output is specified.

• Relative (%): The size of the output is a percentage of the total rows.
• Absolute: The size of the output is an absolute number of rows.

Relative size

The amount of samples relative to the original table.

Number of rows

The absolute amount of samples created.

Enable static seed

If selected, the sampling is driven by a static seed for reproducible results.

Static seed

You may enter a fixed seed here in order to get reproducible results upon re-execution. If you do not specify a seed, a new random seed is taken for each execution.

Draw seed

Generate a random seed and set it in the Random seed input above for reproducible runs.

Append count of occurrences

Will append a column containing the number of times this data is present in the bootstrap samples.

Append original RowID

Will append a column containing the original RowID in the bootstrap samples.

RowID separator

The bootstrap samples have a RowID that is composed of the original RowID, the separator and an incremented number for the copies of each row.

Views

This node has no views

Workflows

• [Latest] Consolidated 3 Mar 1210KNIME Hub
• 0205KNIME Hub
• 05_Cohen's_Kappa_for_Evaluating_Classification_ModelsKNIME Hub
• 05_Cohen's_Kappa_for_Evaluating_Classification_ModelsKNIME Hub
• 1208KNIME Hub
• Show all 60 workflows

Links

• No links available

Developers