# Sample rows Shape / rows random or stratified sample, train and test split Page: ## Status Runs as a SQL step. arcform carries it as a `sql:` step today. ## The DuckDB step a repeatable sample of four rows It reads the `orders` table the proofs share. Create it first: ```sql CREATE TABLE orders AS SELECT * FROM (VALUES (1, 'Alice Ng', 'north', DATE '2026-01-03', 120.00, 'widget', 'a|b', '{"channel":"web","items":[1,2]}'), (2, 'alice ng', 'north', DATE '2026-01-05', 80.50, 'gadget', 'b', '{"channel":"shop","items":[3]}'), (3, 'Bob Okafor', 'south', DATE '2026-01-05', 300.00, 'widget', 'a|c', '{"channel":"web","items":[4,5,6]}'), (4, 'Carla Díaz', 'east', DATE '2026-01-09', NULL, 'gizmo', NULL, '{"channel":"web","items":[]}'), (5, 'Bob Okafor', 'south', DATE '2026-01-12', 45.25, 'gadget', 'c', '{"channel":"phone","items":[7]}'), (6, 'Dev Patel', 'west', DATE '2026-01-12', 510.00, 'widget', 'a', '{"channel":"web","items":[8,9]}'), (7, 'Carla Diaz', 'east', DATE '2026-01-20', 99.99, 'gizmo', 'b|c', '{"channel":"shop","items":[10]}'), (8, 'Dev Patel', 'west', DATE '2026-02-02', 510.00, 'widget', 'a', '{"channel":"web","items":[8,9]}'), (9, 'Eve Larsen', NULL, DATE '2026-02-03', 15.00, 'gadget', NULL, '{"channel":"web","items":[11]}'), (10, 'Alice Ng', 'north', DATE '2026-02-10', 220.00, 'gizmo', 'a|b|c', '{"channel":"phone","items":[12,13]}') ) t(order_id, customer, region, order_date, amount, product, tags, meta); ``` Then run the statement: ```sql CREATE TEMP TABLE result AS SELECT (SELECT list(order_id ORDER BY order_id) FROM (SELECT * FROM orders USING SAMPLE reservoir(4 ROWS) REPEATABLE (42))) AS first_draw, (SELECT list(order_id ORDER BY order_id) FROM (SELECT * FROM orders USING SAMPLE reservoir(4 ROWS) REPEATABLE (42))) AS second_draw; SELECT * FROM result; SELECT CASE WHEN ((SELECT len(first_draw) = 4 AND first_draw = second_draw FROM result)) THEN 'as expected' ELSE error('the two draws with one seed differ, or a draw is not four rows') END AS verdict; ``` It needs nothing installed beyond DuckDB itself. ## Variants - Take every nth row: Keep rows at a fixed interval from an offset. - Other sampling methods: Sample by design or model, sample collections, or write samples aside for checking. - [Oversample or draw with replacement](https://meridian.online/operations/sample-rows/oversample-or-draw-with-replacement): Draw rows with replacement, or synthesise new minority-class rows, so rows can appear more than in the input. - Take a random sample: Keep a random number or share of rows, repeatable with a seed, by Bernoulli, system or reservoir sampling. - Shuffle values at random: Reorder rows, or permute one column's values, at random. - [Take a stratified sample](https://meridian.online/operations/sample-rows/take-a-stratified-sample): Sample within each class of a column so the sample keeps, or evens out, the class shares. - [Split rows into training and test sets](https://meridian.online/operations/sample-rows/split-rows-into-training-and-test-sets): Partition every row at random into two or more sets, such as training, validation and holdout. ## What other tools call it Meridian's classification; an independent blind rating agreed on 86.2% of 312 rows (exact class, version 2). - Alteryx Designer: [`Create Samples`](https://help.alteryx.com/current/en/designer/tools/preparation/create-samples-tool.html) (read 2026-09-27) - Alteryx Designer: [`Oversample Field`](https://help.alteryx.com/current/en/designer/tools/preparation/oversample-field-tool.html) (read 2026-09-27) - Alteryx Designer: [`Random % Sample`](https://help.alteryx.com/current/en/designer/tools/preparation/random---sample-tool.html) (read 2026-09-27) - DuckDB: [`SAMPLE (clause)`](https://duckdb.org/docs/current/sql/query_syntax/sample) (read 2026-09-27) - DuckDB: `duckdb_table_sample` (read 2026-09-27) - Duckle: [`Sample`](https://github.com/slothflowlabs/duckle/blob/f85f8960edfb6efbf4cb89d58a7b32e16df0e867/website/docs/components.html) (read 2026-09-25) - Duckle: [`Sample (Reproducible %)`](https://github.com/slothflowlabs/duckle/blob/f85f8960edfb6efbf4cb89d58a7b32e16df0e867/frontend/src/workflow-ui/palette-data.ts) (read 2026-09-25) - AWS Glue: [`Spigot`](https://docs.aws.amazon.com/glue/latest/dg/aws-glue-api-crawler-pyspark-transforms-spigot.html) (read 2026-09-27) - AWS Glue: [`Spigot`](https://docs.aws.amazon.com/glue/latest/dg/transforms-configure-spigot.html) (read 2026-09-27) - Apache Hop: [`Reservoir Sampling`](https://hop.apache.org/manual/latest/pipeline/transforms/reservoirsampling.html) (read 2026-09-27) - Ibis: [`Sample`](https://github.com/ibis-project/ibis/blob/05d2b293344e86cb41c1a426071734f7e270335a/ibis/expr/operations/relations.py) (read 2026-09-27) - KNIME Analytics Platform: [`Bootstrap Sampling`](https://api.hub.knime.com/nodes/*i1MQOr0EbxLu4c1B?details=full) (read 2026-09-27) - KNIME Analytics Platform: [`DB Row Sampler`](https://api.hub.knime.com/nodes/*4hFtwIsFE99gW1Q6?details=full) (read 2026-09-27) - KNIME Analytics Platform: [`DB Table Partitioner`](https://api.hub.knime.com/nodes/*NKuW-RponmBdKGHF?details=full) (read 2026-09-27) - KNIME Analytics Platform: [`Equal Size Sampling`](https://api.hub.knime.com/nodes/*nZoZFKRELGnk6y5i?details=full) (read 2026-09-27) - KNIME Analytics Platform: [`Row Sampler`](https://api.hub.knime.com/nodes/*3o-UY2qMENf5piCd?details=full) (read 2026-09-27) - KNIME Analytics Platform: [`SMOTE`](https://api.hub.knime.com/nodes/*YznaX_v45d3OzPEV?details=full) (read 2026-09-27) - KNIME Analytics Platform: [`Table Partitioner`](https://api.hub.knime.com/nodes/*T0AhxSe0Yi42rQK8?details=full) (read 2026-09-27) - KNIME Analytics Platform: [`Target Shuffling`](https://api.hub.knime.com/nodes/*GfRYHsSKd2cu-1Rl?details=full) (read 2026-09-27) - mlpack: [`Dataset splitting`](https://www.mlpack.org/doc/user/preprocessing.html) (read 2026-09-27) - Apache NiFi: [`SampleRecord`](https://nifi.apache.org/components/org.apache.nifi.processors.standard.SampleRecord/) (read 2026-09-27) - Pentaho Data Integration: [`Reservoir Sampling`](https://docs.pentaho.com/pdia-data-integration/9.3-data-integration/pdi-transformation-steps-reference-overview) (read 2026-09-27) - Polars: [`DataFrame.gather_every`](https://docs.pola.rs/api/python/stable/reference/dataframe/modify_select.html) (read 2026-09-27) - Polars: [`DataFrame.sample`](https://docs.pola.rs/api/python/stable/reference/dataframe/modify_select.html) (read 2026-09-27) - Polars: [`Expr manipulation/selection: sampling (group)`](https://docs.pola.rs/api/python/stable/reference/expressions/modify_select.html) (read 2026-09-27) - RapidMiner Studio: [`SMOTE Upsampling`](https://docs.rapidminer.com/latest/studio/operators/extensions/Operator%20Toolbox/blending/smote.html) (read 2026-09-27) - RapidMiner Studio: [`Sample`](https://docs.rapidminer.com/latest/studio/operators/blending/examples/sampling/sample.html) (read 2026-09-27) - RapidMiner Studio: [`Sample (Bootstrapping)`](https://docs.rapidminer.com/latest/studio/operators/blending/examples/sampling/sample_bootstrapping.html) (read 2026-09-27) - RapidMiner Studio: [`Sample (Collection)`](https://docs.rapidminer.com/latest/studio/operators/extensions/Operator%20Toolbox/blending/table/sample_collection.html) (read 2026-09-27) - RapidMiner Studio: [`Sample (In Database)`](https://docs.rapidminer.com/latest/studio/operators/extensions/In-Database%20Processing/indb_blending/indb_sampling.html) (read 2026-09-27) - RapidMiner Studio: [`Sample (Kennard-Stone)`](https://docs.rapidminer.com/latest/studio/operators/blending/examples/sampling/sample_kennard_stone.html) (read 2026-09-27) - RapidMiner Studio: [`Sample (Model-Based)`](https://docs.rapidminer.com/latest/studio/operators/blending/examples/sampling/sample_model_based.html) (read 2026-09-27) - RapidMiner Studio: [`Sample (Stratified)`](https://docs.rapidminer.com/latest/studio/operators/blending/examples/sampling/sample_stratified.html) (read 2026-09-27) - RapidMiner Studio: [`Split Data`](https://docs.rapidminer.com/latest/studio/operators/blending/examples/sampling/split_data.html) (read 2026-09-27) - Tableau Prep: [`Data sample`](https://help.tableau.com/current/prep/en-us/prep_configure_dastaset.htm) (read 2026-09-27) - Talend: [`tReservoirSampling`](https://help.qlik.com/talend/en-US/components/8.0/sampling/sampling-component) (read 2026-09-27) - Talend: [`tSample`](https://help.qlik.com/talend/en-US/components/8.0/processing/processing-component) (read 2026-09-27) - VisiData: [`random-rows`](https://github.com/saulpw/visidata/blob/7609212ec2febafa9caba7183cbebfaf49627909/visidata/features/random_sample.py#L3) (read 2026-09-23)