Spark Partitioning

The input data is split into two partitions (i.e. row-wise), e.g. train and test data. The two partitions are available at the two output ports.

Options

Absolute

Specify the absolute number of rows in the sample. If there are less rows than specified here, all rows are used.

Relative

The percentage of the number of rows in the sample. Must be between 0 and 100, inclusively.

Take from top

This mode selects the top most rows of the input data.

Draw randomly

Random sampling of all rows, you may optionally specify a fixed seed and adapt the sample with replacement setting (see below).

Stratified sampling

Check this button if you want stratified sampling, i.e. the distribution of values in the selected column is (approximately) retained in the output table. You may optionally specify a fixed seed and adapt the exact sampling and sample with replacement setting (see below).

Exact sampling

Exact sampling requires significant more resources than the per-stratum simple random sampling used in by default, but will provide the exact sampling size with 99.99% confidence.

Use random seed

If either random or stratified sampling is selected, 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.

Sample with replacement

If selected a row from the input data can be chosen more than once.

Views

This node has no views

Workflows

• 02 Taxi Demand Prediction on Spark DeploymentKNIME Hub
• 03 Manipulating Big DataKNIME Hub
• 03 Manipulating Big Data - SolutionKNIME Hub
• 03 Manipulating Big Data - SolutionKNIME Hub
• 03 Manipulating Big Data - SolutionKNIME Hub
• Show all 40 workflows

Links

• No links available

Developers