Apache Spark - A unified analytics engine for large-scale data processing
Go to file
HyukjinKwon ab0890bdb1 [SPARK-28264][PYTHON][SQL] Support type hints in pandas UDF and rename/move inconsistent pandas UDF types
### What changes were proposed in this pull request?

This PR proposes to redesign pandas UDFs as described in [the proposal](https://docs.google.com/document/d/1-kV0FS_LF2zvaRh_GhkV32Uqksm_Sq8SvnBBmRyxm30/edit?usp=sharing).

```python
from pyspark.sql.functions import pandas_udf
import pandas as pd

pandas_udf("long")
def plug_one(s: pd.Series) -> pd.Series:
    return s + 1

spark.range(10).select(plug_one("id")).show()
```

```
+------------+
|plug_one(id)|
+------------+
|           1|
|           2|
|           3|
|           4|
|           5|
|           6|
|           7|
|           8|
|           9|
|          10|
+------------+
```

Note that, this PR address one of the future improvements described [here](https://docs.google.com/document/d/1-kV0FS_LF2zvaRh_GhkV32Uqksm_Sq8SvnBBmRyxm30/edit#heading=h.h3ncjpk6ujqu), "A couple of less-intuitive pandas UDF types" (by zero323) together.

In short,

- Adds new way with type hints as an alternative and experimental way.
    ```python
    pandas_udf(schema='...')
    def func(c1: Series, c2: Series) -> DataFrame:
        pass
    ```

- Replace and/or add an alias for three types below from UDF, and make them as separate standalone APIs. So, `pandas_udf` is now consistent with regular `udf`s and other expressions.

    `df.mapInPandas(udf)`  -replace-> `df.mapInPandas(f, schema)`
    `df.groupby.apply(udf)`  -alias-> `df.groupby.applyInPandas(f, schema)`
    `df.groupby.cogroup.apply(udf)`  -replace-> `df.groupby.cogroup.applyInPandas(f, schema)`

    *`df.groupby.apply` was added from 2.3 while the other were added in the master only.

- No deprecation for the existing ways for now.
    ```python
    pandas_udf(schema='...', functionType=PandasUDFType.SCALAR)
    def func(c1, c2):
        pass
    ```
If users are happy with this, I plan to deprecate the existing way and declare using type hints is not experimental anymore.

One design goal in this PR was that, avoid touching the internal (since we didn't deprecate the old ways for now), but supports type hints with a minimised changes only at the interface.

- Once we deprecate or remove the old ways, I think it requires another refactoring for the internal in the future. At the very least, we should rename internal pandas evaluation types.
- If users find this experimental type hints isn't quite helpful, we should simply revert the changes at the interface level.

### Why are the changes needed?

In order to address old design issues. Please see [the proposal](https://docs.google.com/document/d/1-kV0FS_LF2zvaRh_GhkV32Uqksm_Sq8SvnBBmRyxm30/edit?usp=sharing).

### Does this PR introduce any user-facing change?

For behaviour changes, No.

It adds new ways to use pandas UDFs by using type hints. See below.

**SCALAR**:

```python
pandas_udf(schema='...')
def func(c1: Series, c2: DataFrame) -> Series:
    pass  # DataFrame represents a struct column
```

**SCALAR_ITER**:

```python
pandas_udf(schema='...')
def func(iter: Iterator[Tuple[Series, DataFrame, ...]]) -> Iterator[Series]:
    pass  # Same as SCALAR but wrapped by Iterator
```

**GROUPED_AGG**:

```python
pandas_udf(schema='...')
def func(c1: Series, c2: DataFrame) -> int:
    pass  # DataFrame represents a struct column
```

**GROUPED_MAP**:

This was added in Spark 2.3 as of SPARK-20396. As described above, it keeps the existing behaviour. Additionally, we now have a new alias `groupby.applyInPandas` for `groupby.apply`. See the example below:

```python
def func(pdf):
    return pdf

df.groupby("...").applyInPandas(func, schema=df.schema)
```

**MAP_ITER**: this is not a pandas UDF anymore

This was added in Spark 3.0 as of SPARK-28198; and this PR replaces the usages. See the example below:

```python
def func(iter):
    for df in iter:
        yield df

df.mapInPandas(func, df.schema)
```

**COGROUPED_MAP**: this is not a pandas UDF anymore

This was added in Spark 3.0 as of SPARK-27463; and this PR replaces the usages. See the example below:

```python
def asof_join(left, right):
    return pd.merge_asof(left, right, on="...", by="...")

 df1.groupby("...").cogroup(df2.groupby("...")).applyInPandas(asof_join, schema="...")
```

### How was this patch tested?

Unittests added and tested against Python 2.7, 3.6 and 3.7.

Closes #27165 from HyukjinKwon/revisit-pandas.

Authored-by: HyukjinKwon <gurwls223@apache.org>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
2020-01-22 15:32:58 +09:00
.github [SPARK-30572][BUILD] Add a fallback Maven repository 2020-01-19 17:42:34 -08:00
assembly [SPARK-30489][BUILD] Make build delete pyspark.zip file properly 2020-01-10 16:59:51 -08:00
bin [SPARK-28525][DEPLOY] Allow Launcher to be applied Java options 2019-07-30 12:45:32 -07:00
build [SPARK-30121][BUILD] Fix memory usage in sbt build script 2019-12-05 11:50:55 -06:00
common [SPARK-30593][SQL] Revert interval ISO/ANSI SQL Standard output since we decide not to follow ANSI and no round trip 2020-01-21 20:51:10 +08:00
conf [SPARK-29032][CORE] Add PrometheusServlet to monitor Master/Worker/Driver 2019-09-13 21:31:21 +00:00
core [SPARK-30599][CORE][TESTS] Increase the maximum number of log events in LogAppender 2020-01-21 14:27:55 -08:00
data [SPARK-22666][ML][SQL] Spark datasource for image format 2018-09-05 11:59:00 -07:00
dev [SPARK-28264][PYTHON][SQL] Support type hints in pandas UDF and rename/move inconsistent pandas UDF types 2020-01-22 15:32:58 +09:00
docs [SPARK-30553][DOCS] fix structured-streaming java example error 2020-01-21 21:37:21 -08:00
examples [SPARK-30423][SQL] Deprecate UserDefinedAggregateFunction 2020-01-14 22:07:13 +08:00
external [SPARK-30475][SQL] File source V2: Push data filters for file listing 2020-01-20 20:20:37 -08:00
graphx [INFRA] Reverts commit 56dcd79 and c216ef1 2019-12-16 19:57:44 -07:00
hadoop-cloud [INFRA] Reverts commit 56dcd79 and c216ef1 2019-12-16 19:57:44 -07:00
launcher [INFRA] Reverts commit 56dcd79 and c216ef1 2019-12-16 19:57:44 -07:00
licenses [SPARK-27557][DOC] Add copy button to Python API docs for easier copying of code-blocks 2019-05-01 11:26:18 -05:00
licenses-binary [SPARK-29308][BUILD] Update deps in dev/deps/spark-deps-hadoop-3.2 for hadoop-3.2 2019-10-13 12:53:12 -05:00
mllib [MINOR][ML] Change DecisionTreeClassifier to FMClassifier in OneVsRest setWeightCol test 2020-01-17 10:04:41 +08:00
mllib-local [SPARK-30329][ML] add iterator/foreach methods for Vectors 2019-12-31 15:52:17 +08:00
project [SPARK-30544][BUILD] Upgrade the version of Genjavadoc to 0.15 2020-01-18 00:15:49 -08:00
python [SPARK-28264][PYTHON][SQL] Support type hints in pandas UDF and rename/move inconsistent pandas UDF types 2020-01-22 15:32:58 +09:00
R [SPARK-30188][SQL] Resolve the failed unit tests when enable AQE 2020-01-13 22:55:19 +08:00
repl [INFRA] Reverts commit 56dcd79 and c216ef1 2019-12-16 19:57:44 -07:00
resource-managers [SPARK-30371][K8S] Add spark.kubernetes.driver.master conf 2020-01-19 14:14:45 -08:00
sbin [SPARK-28164] Fix usage description of start-slave.sh 2019-06-26 12:42:33 -05:00
sql [SPARK-15616][SQL] Add optimizer rule PruneHiveTablePartitions 2020-01-21 21:26:30 +08:00
streaming [INFRA] Reverts commit 56dcd79 and c216ef1 2019-12-16 19:57:44 -07:00
tools [INFRA] Reverts commit 56dcd79 and c216ef1 2019-12-16 19:57:44 -07:00
.gitattributes [SPARK-3870] EOL character enforcement 2014-10-31 12:39:52 -07:00
.gitignore [SPARK-30084][DOCS] Document how to trigger Jekyll build on Python API doc changes 2019-12-04 17:31:23 -06:00
appveyor.yml [SPARK-29991][INFRA] Support Hive 1.2 and Hive 2.3 (default) in PR builder 2019-11-30 12:48:15 +09:00
CONTRIBUTING.md [MINOR][DOCS] Tighten up some key links to the project and download pages to use HTTPS 2019-05-21 10:56:42 -07:00
LICENSE [SPARK-29674][CORE] Update dropwizard metrics to 4.1.x for JDK 9+ 2019-11-03 15:13:06 -08:00
LICENSE-binary Revert [SPARK-27300][GRAPH] Add Spark Graph modules and dependencies 2019-12-17 09:06:23 -08:00
NOTICE [SPARK-29674][CORE] Update dropwizard metrics to 4.1.x for JDK 9+ 2019-11-03 15:13:06 -08:00
NOTICE-binary [SPARK-29674][CORE] Update dropwizard metrics to 4.1.x for JDK 9+ 2019-11-03 15:13:06 -08:00
pom.xml Revert "[SPARK-30534][INFRA] Use mvn in dev/scalastyle" 2020-01-21 18:23:03 +09:00
README.md [MINOR][DOCS] Fix Jenkins build image and link in README.md 2020-01-20 23:08:24 -08:00
scalastyle-config.xml [SPARK-30030][INFRA] Use RegexChecker instead of TokenChecker to check org.apache.commons.lang. 2019-11-25 12:03:15 -08:00

Apache Spark

Spark is a unified analytics engine for large-scale data processing. It provides high-level APIs in Scala, Java, Python, and R, and an optimized engine that supports general computation graphs for data analysis. It also supports a rich set of higher-level tools including Spark SQL for SQL and DataFrames, MLlib for machine learning, GraphX for graph processing, and Structured Streaming for stream processing.

https://spark.apache.org/

Jenkins Build AppVeyor Build PySpark Coverage

Online Documentation

You can find the latest Spark documentation, including a programming guide, on the project web page. This README file only contains basic setup instructions.

Building Spark

Spark is built using Apache Maven. To build Spark and its example programs, run:

./build/mvn -DskipTests clean package

(You do not need to do this if you downloaded a pre-built package.)

More detailed documentation is available from the project site, at "Building Spark".

For general development tips, including info on developing Spark using an IDE, see "Useful Developer Tools".

Interactive Scala Shell

The easiest way to start using Spark is through the Scala shell:

./bin/spark-shell

Try the following command, which should return 1,000,000,000:

scala> spark.range(1000 * 1000 * 1000).count()

Interactive Python Shell

Alternatively, if you prefer Python, you can use the Python shell:

./bin/pyspark

And run the following command, which should also return 1,000,000,000:

>>> spark.range(1000 * 1000 * 1000).count()

Example Programs

Spark also comes with several sample programs in the examples directory. To run one of them, use ./bin/run-example <class> [params]. For example:

./bin/run-example SparkPi

will run the Pi example locally.

You can set the MASTER environment variable when running examples to submit examples to a cluster. This can be a mesos:// or spark:// URL, "yarn" to run on YARN, and "local" to run locally with one thread, or "local[N]" to run locally with N threads. You can also use an abbreviated class name if the class is in the examples package. For instance:

MASTER=spark://host:7077 ./bin/run-example SparkPi

Many of the example programs print usage help if no params are given.

Running Tests

Testing first requires building Spark. Once Spark is built, tests can be run using:

./dev/run-tests

Please see the guidance on how to run tests for a module, or individual tests.

There is also a Kubernetes integration test, see resource-managers/kubernetes/integration-tests/README.md

A Note About Hadoop Versions

Spark uses the Hadoop core library to talk to HDFS and other Hadoop-supported storage systems. Because the protocols have changed in different versions of Hadoop, you must build Spark against the same version that your cluster runs.

Please refer to the build documentation at "Specifying the Hadoop Version and Enabling YARN" for detailed guidance on building for a particular distribution of Hadoop, including building for particular Hive and Hive Thriftserver distributions.

Configuration

Please refer to the Configuration Guide in the online documentation for an overview on how to configure Spark.

Contributing

Please review the Contribution to Spark guide for information on how to get started contributing to the project.