pyspark.sql.functions.lag#

pyspark.sql.functions.lag(col, offset=1, default=None)[source]#

Window function: returns the value that is offset rows before the current row, and default if there is less than offset rows before the current row. For example, an offset of one will return the previous row at any given point in the window partition.

This is equivalent to the LAG function in SQL.

New in version 1.4.0.

Changed in version 3.4.0: Supports Spark Connect.

Parameters

colColumn or column name: name of column or expression
offsetint, optional default 1: number of row to extend
defaultoptional: default value

Returns

Column: value before current row based on offset.

See also

pyspark.sql.functions.lead()

Examples

>>> from pyspark.sql import functions as sf
>>> from pyspark.sql import Window
>>> df = spark.createDataFrame(
...     [("a", 1), ("a", 2), ("a", 3), ("b", 8), ("b", 2)], ["c1", "c2"])
>>> df.show()
+---+---+
| c1| c2|
+---+---+
|  a|  1|
|  a|  2|
|  a|  3|
|  b|  8|
|  b|  2|
+---+---+

>>> w = Window.partitionBy("c1").orderBy("c2")
>>> df.withColumn("previous_value", sf.lag("c2").over(w)).show()
+---+---+--------------+
| c1| c2|previous_value|
+---+---+--------------+
|  a|  1|          NULL|
|  a|  2|             1|
|  a|  3|             2|
|  b|  2|          NULL|
|  b|  8|             2|
+---+---+--------------+

>>> df.withColumn("previous_value", sf.lag("c2", 1, 0).over(w)).show()
+---+---+--------------+
| c1| c2|previous_value|
+---+---+--------------+
|  a|  1|             0|
|  a|  2|             1|
|  a|  3|             2|
|  b|  2|             0|
|  b|  8|             2|
+---+---+--------------+

>>> df.withColumn("previous_value", sf.lag("c2", 2, -1).over(w)).show()
+---+---+--------------+
| c1| c2|previous_value|
+---+---+--------------+
|  a|  1|            -1|
|  a|  2|            -1|
|  a|  3|             1|
|  b|  2|            -1|
|  b|  8|            -1|
+---+---+--------------+