DataFrame

Aggregates on the entire DataFrame without groups.

// df.agg(...) is a shorthand for df.groupBy().agg(...)
df.agg(max($"age"), avg($"salary"))
df.groupBy().agg(max($"age"), avg($"salary"))

(Java-specific) Aggregates on the entire DataFrame without groups.

// df.agg(...) is a shorthand for df.groupBy().agg(...)
df.agg(Map("age" -> "max", "salary" -> "avg"))
df.groupBy().agg(Map("age" -> "max", "salary" -> "avg"))

(Scala-specific) Aggregates on the entire DataFrame without groups.

// df.agg(...) is a shorthand for df.groupBy().agg(...)
df.agg(Map("age" -> "max", "salary" -> "avg"))
df.groupBy().agg(Map("age" -> "max", "salary" -> "avg"))

(Scala-specific) Aggregates on the entire DataFrame without groups.

// df.agg(...) is a shorthand for df.groupBy().agg(...)
df.agg("age" -> "max", "salary" -> "avg")
df.groupBy().agg("age" -> "max", "salary" -> "avg")

(Scala-specific) Returns a new DataFrame with an alias set. Same as as.

Returns a new DataFrame with an alias set. Same as as.

Selects column based on the column name and return it as a Column. Note that the column name can also reference to a nested column like a.b.

(Scala-specific) Returns a new DataFrame with an alias set.

Returns a new DataFrame with an alias set.

:: Experimental :: Converts this DataFrame to a strongly-typed Dataset containing objects of the specified type, U.

Persist this DataFrame with the default storage level (MEMORY_AND_DISK).

Returns a new DataFrame that has exactly numPartitions partitions. Similar to coalesce defined on an RDD, this operation results in a narrow dependency, e.g. if you go from 1000 partitions to 100 partitions, there will not be a shuffle, instead each of the 100 new partitions will claim 10 of the current partitions.

Selects column based on the column name and return it as a Column. Note that the column name can also reference to a nested column like a.b.

Returns an array that contains all of Rows in this DataFrame.

Running collect requires moving all the data into the application's driver process, and doing so on a very large dataset can crash the driver process with OutOfMemoryError.

For Java API, use collectAsList.

Returns a Java list that contains all of Rows in this DataFrame.

Running collect requires moving all the data into the application's driver process, and doing so on a very large dataset can crash the driver process with OutOfMemoryError.

Returns all column names as an array.

Returns the number of rows in the DataFrame.

Create a multi-dimensional cube for the current DataFrame using the specified columns, so we can run aggregation on them. See GroupedData for all the available aggregate functions.

This is a variant of cube that can only group by existing columns using column names (i.e. cannot construct expressions).

// Compute the average for all numeric columns cubed by department and group.
df.cube("department", "group").avg()

// Compute the max age and average salary, cubed by department and gender.
df.cube($"department", $"gender").agg(Map(
  "salary" -> "avg",
  "age" -> "max"
))

Create a multi-dimensional cube for the current DataFrame using the specified columns, so we can run aggregation on them. See GroupedData for all the available aggregate functions.

// Compute the average for all numeric columns cubed by department and group.
df.cube($"department", $"group").avg()

// Compute the max age and average salary, cubed by department and gender.
df.cube($"department", $"gender").agg(Map(
  "salary" -> "avg",
  "age" -> "max"
))

Computes statistics for numeric columns, including count, mean, stddev, min, and max. If no columns are given, this function computes statistics for all numerical columns.

This function is meant for exploratory data analysis, as we make no guarantee about the backward compatibility of the schema of the resulting DataFrame. If you want to programmatically compute summary statistics, use the agg function instead.

df.describe("age", "height").show()

// output:
// summary age   height
// count   10.0  10.0
// mean    53.3  178.05
// stddev  11.6  15.7
// min     18.0  163.0
// max     92.0  192.0

Returns a new DataFrame that contains only the unique rows from this DataFrame. This is an alias for dropDuplicates.

Returns a new DataFrame with a column dropped. This version of drop accepts a Column rather than a name. This is a no-op if the DataFrame doesn't have a column with an equivalent expression.

Returns a new DataFrame with a column dropped. This is a no-op if schema doesn't contain column name.

Returns a new DataFrame with duplicate rows removed, considering only the subset of columns.

(Scala-specific) Returns a new DataFrame with duplicate rows removed, considering only the subset of columns.

Returns a new DataFrame that contains only the unique rows from this DataFrame. This is an alias for distinct.

Returns all column names and their data types as an array.

Returns a new DataFrame containing rows in this frame but not in another frame. This is equivalent to EXCEPT in SQL.

Prints the physical plan to the console for debugging purposes.

Prints the plans (logical and physical) to the console for debugging purposes.

(Scala-specific) Returns a new DataFrame where a single column has been expanded to zero or more rows by the provided function. This is similar to a LATERAL VIEW in HiveQL. All columns of the input row are implicitly joined with each value that is output by the function.

df.explode("words", "word"){words: String => words.split(" ")}

(Scala-specific) Returns a new DataFrame where each row has been expanded to zero or more rows by the provided function. This is similar to a LATERAL VIEW in HiveQL. The columns of the input row are implicitly joined with each row that is output by the function.

The following example uses this function to count the number of books which contain a given word:

case class Book(title: String, words: String)
val df: RDD[Book]

case class Word(word: String)
val allWords = df.explode('words) {
  case Row(words: String) => words.split(" ").map(Word(_))
}

val bookCountPerWord = allWords.groupBy("word").agg(countDistinct("title"))

Filters rows using the given SQL expression.

peopleDf.filter("age > 15")

Filters rows using the given condition.

// The following are equivalent:
peopleDf.filter($"age" > 15)
peopleDf.where($"age" > 15)

Returns the first row. Alias for head().

Returns a new RDD by first applying a function to all rows of this DataFrame, and then flattening the results.

Applies a function f to all rows.

Applies a function f to each partition of this DataFrame.

Groups the DataFrame using the specified columns, so we can run aggregation on them. See GroupedData for all the available aggregate functions.

This is a variant of groupBy that can only group by existing columns using column names (i.e. cannot construct expressions).

// Compute the average for all numeric columns grouped by department.
df.groupBy("department").avg()

// Compute the max age and average salary, grouped by department and gender.
df.groupBy($"department", $"gender").agg(Map(
  "salary" -> "avg",
  "age" -> "max"
))

Groups the DataFrame using the specified columns, so we can run aggregation on them. See GroupedData for all the available aggregate functions.

// Compute the average for all numeric columns grouped by department.
df.groupBy($"department").avg()

// Compute the max age and average salary, grouped by department and gender.
df.groupBy($"department", $"gender").agg(Map(
  "salary" -> "avg",
  "age" -> "max"
))

Returns the first row.

Returns the first n rows.

Returns a best-effort snapshot of the files that compose this DataFrame. This method simply asks each constituent BaseRelation for its respective files and takes the union of all results. Depending on the source relations, this may not find all input files. Duplicates are removed.

Returns a new DataFrame containing rows only in both this frame and another frame. This is equivalent to INTERSECT in SQL.

Returns true if the collect and take methods can be run locally (without any Spark executors).

Returns the content of the DataFrame as a JavaRDD of Rows.

Converts a JavaRDD to a PythonRDD.

Join with another DataFrame, using the given join expression. The following performs a full outer join between df1 and df2.

// Scala:
import org.apache.spark.sql.functions._
df1.join(df2, $"df1Key" === $"df2Key", "outer")

// Java:
import static org.apache.spark.sql.functions.*;
df1.join(df2, col("df1Key").equalTo(col("df2Key")), "outer");

Inner join with another DataFrame, using the given join expression.

// The following two are equivalent:
df1.join(df2, $"df1Key" === $"df2Key")
df1.join(df2).where($"df1Key" === $"df2Key")

Equi-join with another DataFrame using the given columns.

Different from other join functions, the join columns will only appear once in the output, i.e. similar to SQL's JOIN USING syntax.

Note that if you perform a self-join using this function without aliasing the input DataFrames, you will NOT be able to reference any columns after the join, since there is no way to disambiguate which side of the join you would like to reference.

Inner equi-join with another DataFrame using the given columns.

Different from other join functions, the join columns will only appear once in the output, i.e. similar to SQL's JOIN USING syntax.

// Joining df1 and df2 using the columns "user_id" and "user_name"
df1.join(df2, Seq("user_id", "user_name"))

Note that if you perform a self-join using this function without aliasing the input DataFrames, you will NOT be able to reference any columns after the join, since there is no way to disambiguate which side of the join you would like to reference.

Inner equi-join with another DataFrame using the given column.

Different from other join functions, the join column will only appear once in the output, i.e. similar to SQL's JOIN USING syntax.

// Joining df1 and df2 using the column "user_id"
df1.join(df2, "user_id")

Note that if you perform a self-join using this function without aliasing the input DataFrames, you will NOT be able to reference any columns after the join, since there is no way to disambiguate which side of the join you would like to reference.

Cartesian join with another DataFrame.

Note that cartesian joins are very expensive without an extra filter that can be pushed down.

Returns a new DataFrame by taking the first n rows. The difference between this function and head is that head returns an array while limit returns a new DataFrame.

Returns a new RDD by applying a function to all rows of this DataFrame.

Returns a new RDD by applying a function to each partition of this DataFrame.

Returns a DataFrameNaFunctions for working with missing data.

// Dropping rows containing any null values.
df.na.drop()

Returns a new DataFrame sorted by the given expressions. This is an alias of the sort function.

Returns a new DataFrame sorted by the given expressions. This is an alias of the sort function.

Persist this DataFrame with the given storage level.

Persist this DataFrame with the default storage level (MEMORY_AND_DISK).

Prints the schema to the console in a nice tree format.

Randomly splits this DataFrame with the provided weights.

Randomly splits this DataFrame with the provided weights.

Represents the content of the DataFrame as an RDD of Rows. Note that the RDD is memoized. Once called, it won't change even if you change any query planning related Spark SQL configurations (e.g. spark.sql.shuffle.partitions).

Registers this DataFrame as a temporary table using the given name. The lifetime of this temporary table is tied to the SQLContext that was used to create this DataFrame.

Returns a new DataFrame partitioned by the given partitioning expressions preserving the existing number of partitions. The resulting DataFrame is hash partitioned.

This is the same operation as "DISTRIBUTE BY" in SQL (Hive QL).

Returns a new DataFrame partitioned by the given partitioning expressions into numPartitions. The resulting DataFrame is hash partitioned.

This is the same operation as "DISTRIBUTE BY" in SQL (Hive QL).

Returns a new DataFrame that has exactly numPartitions partitions.

Create a multi-dimensional rollup for the current DataFrame using the specified columns, so we can run aggregation on them. See GroupedData for all the available aggregate functions.

This is a variant of rollup that can only group by existing columns using column names (i.e. cannot construct expressions).

// Compute the average for all numeric columns rolluped by department and group.
df.rollup("department", "group").avg()

// Compute the max age and average salary, rolluped by department and gender.
df.rollup($"department", $"gender").agg(Map(
  "salary" -> "avg",
  "age" -> "max"
))

Create a multi-dimensional rollup for the current DataFrame using the specified columns, so we can run aggregation on them. See GroupedData for all the available aggregate functions.

// Compute the average for all numeric columns rolluped by department and group.
df.rollup($"department", $"group").avg()

// Compute the max age and average salary, rolluped by department and gender.
df.rollup($"department", $"gender").agg(Map(
  "salary" -> "avg",
  "age" -> "max"
))

Returns a new DataFrame by sampling a fraction of rows, using a random seed.

Returns a new DataFrame by sampling a fraction of rows.

Returns the schema of this DataFrame.

Selects a set of columns. This is a variant of select that can only select existing columns using column names (i.e. cannot construct expressions).

// The following two are equivalent:
df.select("colA", "colB")
df.select($"colA", $"colB")

Selects a set of column based expressions.

df.select($"colA", $"colB" + 1)

Selects a set of SQL expressions. This is a variant of select that accepts SQL expressions.

// The following are equivalent:
df.selectExpr("colA", "colB as newName", "abs(colC)")
df.select(expr("colA"), expr("colB as newName"), expr("abs(colC)"))

Displays the DataFrame in a tabular form. For example:

year  month AVG('Adj Close) MAX('Adj Close)
1980  12    0.503218        0.595103
1981  01    0.523289        0.570307
1982  02    0.436504        0.475256
1983  03    0.410516        0.442194
1984  04    0.450090        0.483521

Displays the top 20 rows of DataFrame in a tabular form.

Displays the top 20 rows of DataFrame in a tabular form. Strings more than 20 characters will be truncated, and all cells will be aligned right.

Displays the DataFrame in a tabular form. Strings more than 20 characters will be truncated, and all cells will be aligned right. For example:

year  month AVG('Adj Close) MAX('Adj Close)
1980  12    0.503218        0.595103
1981  01    0.523289        0.570307
1982  02    0.436504        0.475256
1983  03    0.410516        0.442194
1984  04    0.450090        0.483521

Returns a new DataFrame sorted by the given expressions. For example:

df.sort($"col1", $"col2".desc)

Returns a new DataFrame sorted by the specified column, all in ascending order.

// The following 3 are equivalent
df.sort("sortcol")
df.sort($"sortcol")
df.sort($"sortcol".asc)

Returns a new DataFrame with each partition sorted by the given expressions.

This is the same operation as "SORT BY" in SQL (Hive QL).

Returns a new DataFrame with each partition sorted by the given expressions.

This is the same operation as "SORT BY" in SQL (Hive QL).

Returns a DataFrameStatFunctions for working statistic functions support.

// Finding frequent items in column with name 'a'.
df.stat.freqItems(Seq("a"))

Returns the first n rows in the DataFrame.

Running take requires moving data into the application's driver process, and doing so with a very large n can crash the driver process with OutOfMemoryError.

Returns the first n rows in the DataFrame as a list.

Running take requires moving data into the application's driver process, and doing so with a very large n can crash the driver process with OutOfMemoryError.

Returns a new DataFrame with columns renamed. This can be quite convenient in conversion from a RDD of tuples into a DataFrame with meaningful names. For example:

val rdd: RDD[(Int, String)] = ...
rdd.toDF()  // this implicit conversion creates a DataFrame with column name _1 and _2
rdd.toDF("id", "name")  // this creates a DataFrame with column name "id" and "name"

Returns the object itself.

Returns the content of the DataFrame as a RDD of JSON strings.

Returns the content of the DataFrame as a JavaRDD of Rows.

Concise syntax for chaining custom transformations.

def featurize(ds: DataFrame) = ...

df
  .transform(featurize)
  .transform(...)

Returns a new DataFrame containing union of rows in this frame and another frame. This is equivalent to UNION ALL in SQL.

Mark the DataFrame as non-persistent, and remove all blocks for it from memory and disk.

Mark the DataFrame as non-persistent, and remove all blocks for it from memory and disk.

Filters rows using the given SQL expression.

peopleDf.where("age > 15")

Filters rows using the given condition. This is an alias for filter.

// The following are equivalent:
peopleDf.filter($"age" > 15)
peopleDf.where($"age" > 15)

Returns a new DataFrame by adding a column or replacing the existing column that has the same name.

Returns a new DataFrame with a column renamed. This is a no-op if schema doesn't contain existingName.

:: Experimental :: Interface for saving the content of the DataFrame out into external storage.

Save this DataFrame to a JDBC database at url under the table name table. This will run a CREATE TABLE and a bunch of INSERT INTO statements. If you pass true for allowExisting, it will drop any table with the given name; if you pass false, it will throw if the table already exists.

Adds the rows from this RDD to the specified table. Throws an exception if the table already exists.

Adds the rows from this RDD to the specified table, optionally overwriting the existing data.

Save this DataFrame to a JDBC database at url under the table name table. Assumes the table already exists and has a compatible schema. If you pass true for overwrite, it will TRUNCATE the table before performing the INSERTs.

The table must already exist on the database. It must have a schema that is compatible with the schema of this RDD; inserting the rows of the RDD in order via the simple statement INSERT INTO table VALUES (?, ?, ..., ?) should not fail.

(Scala-specific) Saves the contents of this DataFrame based on the given data source, SaveMode specified by mode, and a set of options

Saves the contents of this DataFrame based on the given data source, SaveMode specified by mode, and a set of options.

Saves the contents of this DataFrame to the given path based on the given data source and SaveMode specified by mode.

Saves the contents of this DataFrame to the given path based on the given data source, using SaveMode.ErrorIfExists as the save mode.

Saves the contents of this DataFrame to the given path and SaveMode specified by mode, using the default data source configured by spark.sql.sources.default.

Saves the contents of this DataFrame to the given path, using the default data source configured by spark.sql.sources.default and SaveMode.ErrorIfExists as the save mode.

Saves the contents of this DataFrame as a parquet file, preserving the schema. Files that are written out using this method can be read back in as a DataFrame using the parquetFile function in SQLContext.

(Scala-specific) Creates a table from the the contents of this DataFrame based on a given data source, SaveMode specified by mode, and a set of options.

Note that this currently only works with DataFrames that are created from a HiveContext as there is no notion of a persisted catalog in a standard SQL context. Instead you can write an RDD out to a parquet file, and then register that file as a table. This "table" can then be the target of an insertInto.

When the DataFrame is created from a non-partitioned HadoopFsRelation with a single input path, and the data source provider can be mapped to an existing Hive builtin SerDe (i.e. ORC and Parquet), the table is persisted in a Hive compatible format, which means other systems like Hive will be able to read this table. Otherwise, the table is persisted in a Spark SQL specific format.

Creates a table at the given path from the the contents of this DataFrame based on a given data source, SaveMode specified by mode, and a set of options.

Note that this currently only works with DataFrames that are created from a HiveContext as there is no notion of a persisted catalog in a standard SQL context. Instead you can write an RDD out to a parquet file, and then register that file as a table. This "table" can then be the target of an insertInto.

When the DataFrame is created from a non-partitioned HadoopFsRelation with a single input path, and the data source provider can be mapped to an existing Hive builtin SerDe (i.e. ORC and Parquet), the table is persisted in a Hive compatible format, which means other systems like Hive will be able to read this table. Otherwise, the table is persisted in a Spark SQL specific format.

:: Experimental :: Creates a table at the given path from the the contents of this DataFrame based on a given data source, SaveMode specified by mode, and a set of options.

Note that this currently only works with DataFrames that are created from a HiveContext as there is no notion of a persisted catalog in a standard SQL context. Instead you can write an RDD out to a parquet file, and then register that file as a table. This "table" can then be the target of an insertInto.

When the DataFrame is created from a non-partitioned HadoopFsRelation with a single input path, and the data source provider can be mapped to an existing Hive builtin SerDe (i.e. ORC and Parquet), the table is persisted in a Hive compatible format, which means other systems like Hive will be able to read this table. Otherwise, the table is persisted in a Spark SQL specific format.

Creates a table at the given path from the the contents of this DataFrame based on a given data source and a set of options, using SaveMode.ErrorIfExists as the save mode.

Note that this currently only works with DataFrames that are created from a HiveContext as there is no notion of a persisted catalog in a standard SQL context. Instead you can write an RDD out to a parquet file, and then register that file as a table. This "table" can then be the target of an insertInto.

When the DataFrame is created from a non-partitioned HadoopFsRelation with a single input path, and the data source provider can be mapped to an existing Hive builtin SerDe (i.e. ORC and Parquet), the table is persisted in a Hive compatible format, which means other systems like Hive will be able to read this table. Otherwise, the table is persisted in a Spark SQL specific format.

Creates a table from the the contents of this DataFrame, using the default data source configured by spark.sql.sources.default and SaveMode.ErrorIfExists as the save mode.

Note that this currently only works with DataFrames that are created from a HiveContext as there is no notion of a persisted catalog in a standard SQL context. Instead you can write an RDD out to a parquet file, and then register that file as a table. This "table" can then be the target of an insertInto.

When the DataFrame is created from a non-partitioned HadoopFsRelation with a single input path, and the data source provider can be mapped to an existing Hive builtin SerDe (i.e. ORC and Parquet), the table is persisted in a Hive compatible format, which means other systems like Hive will be able to read this table. Otherwise, the table is persisted in a Spark SQL specific format.

Creates a table from the the contents of this DataFrame. It will use the default data source configured by spark.sql.sources.default. This will fail if the table already exists.

Note that this currently only works with DataFrames that are created from a HiveContext as there is no notion of a persisted catalog in a standard SQL context. Instead you can write an RDD out to a parquet file, and then register that file as a table. This "table" can then be the target of an insertInto.

When the DataFrame is created from a non-partitioned HadoopFsRelation with a single input path, and the data source provider can be mapped to an existing Hive builtin SerDe (i.e. ORC and Parquet), the table is persisted in a Hive compatible format, which means other systems like Hive will be able to read this table. Otherwise, the table is persisted in a Spark SQL specific format.