### What changes were proposed in this pull request?
Revert 397b843890 and 5a48eb8d00
### Why are the changes needed?
As discussed in https://github.com/apache/spark/pull/33800#issuecomment-904140869, there is correctness issue in the current implementation. Let's revert the code changes from branch 3.2 and fix it on master branch later
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
Ci tests
Closes#33819 from gengliangwang/revert-SPARK-34415.
Authored-by: Gengliang Wang <gengliang@apache.org>
Signed-off-by: Dongjoon Hyun <dongjoon@apache.org>
(cherry picked from commit de932f51ce)
Signed-off-by: Dongjoon Hyun <dongjoon@apache.org>
### What changes were proposed in this pull request?
This PR is a followup of https://github.com/apache/spark/pull/32429 and https://github.com/apache/spark/pull/32644.
I was thinking about creating separate PRs but decided to include all in this PR because it shares the same context, and should be easier to review together.
This PR includes:
- Use `InheritableThread` and `inheritable_thread_target` in the current code base to prevent potential resource leak (since we enabled pinned thread mode by default now at https://github.com/apache/spark/pull/32429)
- Copy local properties when `start` at `InheritableThread` is called to mimic JVM behaviour. Previously it was copied when `InheritableThread` instance was created (related to #32644).
- https://github.com/apache/spark/pull/32429 missed one place at `inheritable_thread_target` (https://github.com/apache/spark/blob/master/python/pyspark/util.py#L308). More specifically, I missed one place that should enable pinned thread mode by default.
### Why are the changes needed?
To mimic the JVM behaviour about thread lifecycle
### Does this PR introduce _any_ user-facing change?
Ideally no. One possible case is that users use `InheritableThread` with pinned thread mode enabled.
In this case, the local properties will be copied when starting the thread instead of defining the `InheritableThread` object.
This is a small difference that wouldn't likely affect end users.
### How was this patch tested?
Existing tests should cover this.
Closes#32962 from HyukjinKwon/SPARK-35498-SPARK-35303.
Authored-by: Hyukjin Kwon <gurwls223@apache.org>
Signed-off-by: Hyukjin Kwon <gurwls223@apache.org>
### What changes were proposed in this pull request?
Add required imports to Pyspark ML examples in CrossValidator, TrainValidationSplit
### Why are the changes needed?
The examples pass doctests because of previous imports, but as they appear in Pyspark documentation, are incomplete. The additional imports are required to make the example work.
### Does this PR introduce _any_ user-facing change?
No, docs only change.
### How was this patch tested?
Existing tests.
Closes#32554 from srowen/TuningImports.
Authored-by: Sean Owen <srowen@gmail.com>
Signed-off-by: Sean Owen <srowen@gmail.com>
### What changes were proposed in this pull request?
Code in the PR generates random parameters for hyperparameter tuning. A discussion with Sean Owen can be found on the dev mailing list here:
http://apache-spark-developers-list.1001551.n3.nabble.com/Hyperparameter-Optimization-via-Randomization-td30629.html
All code is entirely my own work and I license the work to the project under the project’s open source license.
### Why are the changes needed?
Randomization can be a more effective techinique than a grid search since min/max points can fall between the grid and never be found. Randomisation is not so restricted although the probability of finding minima/maxima is dependent on the number of attempts.
Alice Zheng has an accessible description on how this technique works at https://www.oreilly.com/library/view/evaluating-machine-learning/9781492048756/ch04.html
Although there are Python libraries with more sophisticated techniques, not every Spark developer is using Python.
### Does this PR introduce _any_ user-facing change?
A new class (`ParamRandomBuilder.scala`) and its tests have been created but there is no change to existing code. This class offers an alternative to `ParamGridBuilder` and can be dropped into the code wherever `ParamGridBuilder` appears. Indeed, it extends `ParamGridBuilder` and is completely compatible with its interface. It merely adds one method that provides a range over which a hyperparameter will be randomly defined.
### How was this patch tested?
Tests `ParamRandomBuilderSuite.scala` and `RandomRangesSuite.scala` were added.
`ParamRandomBuilderSuite` is the analogue of the already existing `ParamGridBuilderSuite` which tests the user-facing interface.
`RandomRangesSuite` uses ScalaCheck to test the random ranges over which hyperparameters are distributed.
Closes#31535 from PhillHenry/ParamRandomBuilder.
Authored-by: Phillip Henry <PhillHenry@gmail.com>
Signed-off-by: Sean Owen <srowen@gmail.com>
### What changes were proposed in this pull request?
make CrossValidator/TrainValidateSplit/OneVsRest Reader/Writer support Python backend estimator/model
### Why are the changes needed?
Currently, pyspark support third-party library to define python backend estimator/evaluator, i.e., estimator that inherit `Estimator` instead of `JavaEstimator`, and only can be used in pyspark.
CrossValidator and TrainValidateSplit support tuning these python backend estimator,
but cannot support saving/load, becase CrossValidator and TrainValidateSplit writer implementation is use JavaMLWriter, which require to convert nested estimator and evaluator into java instance.
OneVsRest saving/load now only support java backend classifier due to similar issue.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
Unit test.
Closes#30471 from WeichenXu123/support_pyio_tuning.
Authored-by: Weichen Xu <weichen.xu@databricks.com>
Signed-off-by: Weichen Xu <weichen.xu@databricks.com>
### What changes were proposed in this pull request?
Fix: Pyspark ML Validator params in estimatorParamMaps may be lost after saving and reloading
When saving validator estimatorParamMaps, will check all nested stages in tuned estimator to get correct param parent.
Two typical cases to manually test:
~~~python
tokenizer = Tokenizer(inputCol="text", outputCol="words")
hashingTF = HashingTF(inputCol=tokenizer.getOutputCol(), outputCol="features")
lr = LogisticRegression()
pipeline = Pipeline(stages=[tokenizer, hashingTF, lr])
paramGrid = ParamGridBuilder() \
.addGrid(hashingTF.numFeatures, [10, 100]) \
.addGrid(lr.maxIter, [100, 200]) \
.build()
tvs = TrainValidationSplit(estimator=pipeline,
estimatorParamMaps=paramGrid,
evaluator=MulticlassClassificationEvaluator())
tvs.save(tvsPath)
loadedTvs = TrainValidationSplit.load(tvsPath)
# check `loadedTvs.getEstimatorParamMaps()` restored correctly.
~~~
~~~python
lr = LogisticRegression()
ova = OneVsRest(classifier=lr)
grid = ParamGridBuilder().addGrid(lr.maxIter, [100, 200]).build()
evaluator = MulticlassClassificationEvaluator()
tvs = TrainValidationSplit(estimator=ova, estimatorParamMaps=grid, evaluator=evaluator)
tvs.save(tvsPath)
loadedTvs = TrainValidationSplit.load(tvsPath)
# check `loadedTvs.getEstimatorParamMaps()` restored correctly.
~~~
### Why are the changes needed?
Bug fix.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
Unit test.
Closes#30539 from WeichenXu123/fix_tuning_param_maps_io.
Authored-by: Weichen Xu <weichen.xu@databricks.com>
Signed-off-by: Ruifeng Zheng <ruifengz@foxmail.com>
### What changes were proposed in this pull request?
This PR proposes migration of `pyspark.ml` to NumPy documentation style.
### Why are the changes needed?
To improve documentation style.
### Does this PR introduce _any_ user-facing change?
Yes, this changes both rendered HTML docs and console representation (SPARK-33243).
### How was this patch tested?
`dev/lint-python` and manual inspection.
Closes#30285 from zero323/SPARK-33251.
Authored-by: zero323 <mszymkiewicz@gmail.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
### What changes were proposed in this pull request?
This PR adjusts signatures of methods decorated with `keyword_only` to indicate using [Python 3 keyword-only syntax](https://www.python.org/dev/peps/pep-3102/).
__Note__:
For the moment the goal is not to replace `keyword_only`. For justification see https://github.com/apache/spark/pull/29591#discussion_r489402579
### Why are the changes needed?
Right now it is not clear that `keyword_only` methods are indeed keyword only. This proposal addresses that.
In practice we could probably capture `locals` and drop `keyword_only` completel, i.e:
```python
keyword_only
def __init__(self, *, featuresCol="features"):
...
kwargs = self._input_kwargs
self.setParams(**kwargs)
```
could be replaced with
```python
def __init__(self, *, featuresCol="features"):
kwargs = locals()
del kwargs["self"]
...
self.setParams(**kwargs)
```
### Does this PR introduce _any_ user-facing change?
Docstrings and inspect tools will now indicate that `keyword_only` methods expect only keyword arguments.
For example with ` LinearSVC` will change from
```
>>> from pyspark.ml.classification import LinearSVC
>>> ?LinearSVC.__init__
Signature:
LinearSVC.__init__(
self,
featuresCol='features',
labelCol='label',
predictionCol='prediction',
maxIter=100,
regParam=0.0,
tol=1e-06,
rawPredictionCol='rawPrediction',
fitIntercept=True,
standardization=True,
threshold=0.0,
weightCol=None,
aggregationDepth=2,
)
Docstring: __init__(self, featuresCol="features", labelCol="label", predictionCol="prediction", maxIter=100, regParam=0.0, tol=1e-6, rawPredictionCol="rawPrediction", fitIntercept=True, standardization=True, threshold=0.0, weightCol=None, aggregationDepth=2):
File: /path/to/python/pyspark/ml/classification.py
Type: function
```
to
```
>>> from pyspark.ml.classification import LinearSVC
>>> ?LinearSVC.__init__
Signature:
LinearSVC.__init__ (
self,
*,
featuresCol='features',
labelCol='label',
predictionCol='prediction',
maxIter=100,
regParam=0.0,
tol=1e-06,
rawPredictionCol='rawPrediction',
fitIntercept=True,
standardization=True,
threshold=0.0,
weightCol=None,
aggregationDepth=2,
blockSize=1,
)
Docstring: __init__(self, \*, featuresCol="features", labelCol="label", predictionCol="prediction", maxIter=100, regParam=0.0, tol=1e-6, rawPredictionCol="rawPrediction", fitIntercept=True, standardization=True, threshold=0.0, weightCol=None, aggregationDepth=2, blockSize=1):
File: ~/Workspace/spark/python/pyspark/ml/classification.py
Type: function
```
### How was this patch tested?
Existing tests.
Closes#29799 from zero323/SPARK-32933.
Authored-by: zero323 <mszymkiewicz@gmail.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
https://issues.apache.org/jira/browse/SPARK-32719
### What changes were proposed in this pull request?
Add a check to detect missing imports. This makes sure that if we use a specific class, it should be explicitly imported (not using a wildcard).
### Why are the changes needed?
To make sure that the quality of the Python code is up to standard.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
Existing unit-tests and Flake8 static analysis
Closes#29563 from Fokko/fd-add-check-missing-imports.
Authored-by: Fokko Driesprong <fokko@apache.org>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
### What changes were proposed in this pull request?
Fixed `CrossValidatorModel.copy()` so that it correctly calls `.copy()` on the models instead of lists of models.
### Why are the changes needed?
`copy()` was first changed in #29445 . The issue was found in CI of #29524 and fixed. This PR introduces the exact same change so that `CrossValidatorModel.copy()` and its related tests are aligned in branch `master` and branch `branch-3.0`.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Updated `test_copy` to make sure `copy()` is called on models instead of lists of models.
Closes#29553 from Louiszr/fix-cv-copy.
Authored-by: Louiszr <zxhst14@gmail.com>
Signed-off-by: Huaxin Gao <huaxing@us.ibm.com>
### What changes were proposed in this pull request?
Changed the definitions of `CrossValidatorModel.copy()/_to_java()/_from_java()` so that exposed parameters (i.e. parameters with `get()` methods) are copied in these methods.
### Why are the changes needed?
Parameters are copied in the respective Scala interface for `CrossValidatorModel.copy()`.
It fits the semantics to persist parameters when calling `CrossValidatorModel.save()` and `CrossValidatorModel.load()` so that the user gets the same model by saving and loading it after. Not copying across `numFolds` also causes bugs like Array index out of bound and losing sub-models because this parameters will always default to 3 (as described in the JIRA ticket).
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Tests for `CrossValidatorModel.copy()` and `save()`/`load()` are updated so that they check parameters before and after function calls.
Closes#29445 from Louiszr/master.
Authored-by: Louiszr <zxhst14@gmail.com>
Signed-off-by: Sean Owen <srowen@gmail.com>
### What changes were proposed in this pull request?
set params default values in trait Params for feature and tuning in both Scala and Python.
### Why are the changes needed?
Make ML has the same default param values between estimator and its corresponding transformer, and also between Scala and Python.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
Existing and modified tests
Closes#29153 from huaxingao/default2.
Authored-by: Huaxin Gao <huaxing@us.ibm.com>
Signed-off-by: Huaxin Gao <huaxing@us.ibm.com>
### What changes were proposed in this pull request?
This PR aims to drop Python 2.7, 3.4 and 3.5.
Roughly speaking, it removes all the widely known Python 2 compatibility workarounds such as `sys.version` comparison, `__future__`. Also, it removes the Python 2 dedicated codes such as `ArrayConstructor` in Spark.
### Why are the changes needed?
1. Unsupport EOL Python versions
2. Reduce maintenance overhead and remove a bit of legacy codes and hacks for Python 2.
3. PyPy2 has a critical bug that causes a flaky test, SPARK-28358 given my testing and investigation.
4. Users can use Python type hints with Pandas UDFs without thinking about Python version
5. Users can leverage one latest cloudpickle, https://github.com/apache/spark/pull/28950. With Python 3.8+ it can also leverage C pickle.
### Does this PR introduce _any_ user-facing change?
Yes, users cannot use Python 2.7, 3.4 and 3.5 in the upcoming Spark version.
### How was this patch tested?
Manually tested and also tested in Jenkins.
Closes#28957 from HyukjinKwon/SPARK-32138.
Authored-by: HyukjinKwon <gurwls223@apache.org>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
### What changes were proposed in this pull request?
This patch adds user-specified fold column support to `CrossValidator`. User can assign fold numbers to dataset instead of letting Spark do random splits.
### Why are the changes needed?
This gives `CrossValidator` users more flexibility in splitting folds.
### Does this PR introduce _any_ user-facing change?
Yes, a new `foldCol` param is added to `CrossValidator`. User can use it to specify custom fold splitting.
### How was this patch tested?
Added unit tests.
Closes#28704 from viirya/SPARK-31777.
Authored-by: Liang-Chi Hsieh <viirya@gmail.com>
Signed-off-by: Liang-Chi Hsieh <liangchi@uber.com>
### What changes were proposed in this pull request?
There are some parity issues between python and scala
### Why are the changes needed?
keep parity between python and scala
### Does this PR introduce any user-facing change?
Yes
### How was this patch tested?
existing tests
Closes#27196 from huaxingao/spark-30498.
Authored-by: Huaxin Gao <huaxing@us.ibm.com>
Signed-off-by: zhengruifeng <ruifengz@foxmail.com>
modify Param._copyValues to check valid Param objects supplied as extra
### What changes were proposed in this pull request?
Estimator.fit() and Model.transform() accept a dictionary of extra parameters whose values are used to overwrite those supplied at initialization or by default. Additionally, the ParamGridBuilder.addGrid accepts a parameter and list of values. The keys are presumed to be valid Param objects. This change adds a check that only Param objects are supplied as keys.
### Why are the changes needed?
Param objects are created by and bound to an instance of Params (Estimator, Model, or Transformer). They may be obtained from their parent as attributes, or by name through getParam.
The documentation does not state that keys must be valid Param objects, nor describe how one may be obtained. The current behavior is to silently ignore keys which are not valid Param objects.
### Does this PR introduce any user-facing change?
If the user does not pass in a Param object as required for keys in `extra` for Estimator.fit() and Model.transform(), and `param` for ParamGridBuilder.addGrid, an error will be raised indicating it is an invalid object.
### How was this patch tested?
Added method test_copy_param_extras_check to test_param.py. Tested with Python 3.7
Closes#26527 from JohnHBauer/paramExtra.
Authored-by: John Bauer <john.h.bauer@gmail.com>
Signed-off-by: Bryan Cutler <cutlerb@gmail.com>
### What changes were proposed in this pull request?
Remove automatically generated param setters in _shared_params_code_gen.py
### Why are the changes needed?
To keep parity between scala and python
### Does this PR introduce any user-facing change?
Yes
Add some setters in Python ML XXXModels
### How was this patch tested?
unit tests
Closes#26232 from huaxingao/spark-29093.
Authored-by: Huaxin Gao <huaxing@us.ibm.com>
Signed-off-by: zhengruifeng <ruifengz@foxmail.com>
### What changes were proposed in this pull request?
Currently pyspark doesn't write/read `avgMetrics` in `CrossValidatorModel`, whereas scala supports it.
### Why are the changes needed?
Test step to reproduce it:
```
dataset = spark.createDataFrame([(Vectors.dense([0.0]), 0.0),
(Vectors.dense([0.4]), 1.0),
(Vectors.dense([0.5]), 0.0),
(Vectors.dense([0.6]), 1.0),
(Vectors.dense([1.0]), 1.0)] * 10,
["features", "label"])
lr = LogisticRegression()
grid = ParamGridBuilder().addGrid(lr.maxIter, [0, 1]).build()
evaluator = BinaryClassificationEvaluator()
cv = CrossValidator(estimator=lr, estimatorParamMaps=grid, evaluator=evaluator,parallelism=2)
cvModel = cv.fit(dataset)
cvModel.write().save("/tmp/model")
cvModel2 = CrossValidatorModel.read().load("/tmp/model")
print(cvModel.avgMetrics) # prints non empty result as expected
print(cvModel2.avgMetrics) # Bug: prints an empty result.
```
### Does this PR introduce any user-facing change?
No
### How was this patch tested?
Manually tested
Before patch:
```
>>> cvModel.write().save("/tmp/model_0")
>>> cvModel2 = CrossValidatorModel.read().load("/tmp/model_0")
>>> print(cvModel2.avgMetrics)
[]
```
After patch:
```
>>> cvModel2 = CrossValidatorModel.read().load("/tmp/model_2")
>>> print(cvModel2.avgMetrics[0])
0.5
```
Closes#26038 from shahidki31/avgMetrics.
Authored-by: shahid <shahidki31@gmail.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
### What changes were proposed in this pull request?
Follow Scala ml tuning implementation
- put leading underscore before python ```ValidatorParams``` to indicate private
- add ```_CrossValidatorParams``` and ```_TrainValidationSplitParams```
- separate the getters and setters. Put getters in _XXXParams and setters in the Classes.
### Why are the changes needed?
Keep parity between scala and python
### Does this PR introduce any user-facing change?
add ```CrossValidatorModel.getNumFolds``` and ```TrainValidationSplitModel.getTrainRatio()```
### How was this patch tested?
Add doctest
Closes#26057 from huaxingao/spark-tuning.
Authored-by: Huaxin Gao <huaxing@us.ibm.com>
Signed-off-by: zhengruifeng <ruifengz@foxmail.com>
### What changes were proposed in this pull request?
The Experimental and Evolving annotations are both (like Unstable) used to express that a an API may change. However there are many things in the code that have been marked that way since even Spark 1.x. Per the dev thread, anything introduced at or before Spark 2.3.0 is pretty much 'stable' in that it would not change without a deprecation cycle. Therefore I'd like to remove most of these annotations. And, remove the `:: Experimental ::` scaladoc tag too. And likewise for Python, R.
The changes below can be summarized as:
- Generally, anything introduced at or before Spark 2.3.0 has been unmarked as neither Evolving nor Experimental
- Obviously experimental items like DSv2, Barrier mode, ExperimentalMethods are untouched
- I _did_ unmark a few MLlib classes introduced in 2.4, as I am quite confident they're not going to change (e.g. KolmogorovSmirnovTest, PowerIterationClustering)
It's a big change to review, so I'd suggest scanning the list of _files_ changed to see if any area seems like it should remain partly experimental and examine those.
### Why are the changes needed?
Many of these annotations are incorrect; the APIs are de facto stable. Leaving them also makes legitimate usages of the annotations less meaningful.
### Does this PR introduce any user-facing change?
No.
### How was this patch tested?
Existing tests.
Closes#25558 from srowen/SPARK-28855.
Authored-by: Sean Owen <sean.owen@databricks.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
The hashSeed method allocates 64 bytes instead of 8. Other bytes are always zeros (thanks to default behavior of ByteBuffer). And they could be excluded from hash calculation because they don't differentiate inputs.
## How was this patch tested?
By running the existing tests - XORShiftRandomSuite
Closes#20793 from MaxGekk/hash-buff-size.
Lead-authored-by: Maxim Gekk <maxim.gekk@databricks.com>
Co-authored-by: Maxim Gekk <max.gekk@gmail.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
When the grid of the parameters is created in `ParamGridBuilder`, the implicit type coercion is not enforced. So using an integer in the list of parameters to set for a parameter accepting a double can cause a class cast exception.
The PR proposes to enforce the type coercion when building the parameters.
## How was this patch tested?
added UT
Closes#22076 from mgaido91/SPARK-25090.
Authored-by: Marco Gaido <marcogaido91@gmail.com>
Signed-off-by: hyukjinkwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
Add python API for collecting sub-models during CrossValidator/TrainValidationSplit fitting.
## How was this patch tested?
UT added.
Author: WeichenXu <weichen.xu@databricks.com>
Closes#19627 from WeichenXu123/expose-model-list-py.
The exit() builtin is only for interactive use. applications should use sys.exit().
## What changes were proposed in this pull request?
All usage of the builtin `exit()` function is replaced by `sys.exit()`.
## How was this patch tested?
I ran `python/run-tests`.
Please review http://spark.apache.org/contributing.html before opening a pull request.
Author: Benjamin Peterson <benjamin@python.org>
Closes#20682 from benjaminp/sys-exit.
## What changes were proposed in this pull request?
Adding fitMultiple API to `Estimator` with default implementation. Also update have ml.tuning meta-estimators use this API.
## How was this patch tested?
Unit tests.
Author: Bago Amirbekian <bago@databricks.com>
Closes#20058 from MrBago/python-fitMultiple.
## What changes were proposed in this pull request?
Add parallelism support for ML tuning in pyspark.
## How was this patch tested?
Test updated.
Author: WeichenXu <weichen.xu@databricks.com>
Closes#19122 from WeichenXu123/par-ml-tuning-py.
## What changes were proposed in this pull request?
Added functionality for CrossValidator and TrainValidationSplit to persist nested estimators such as OneVsRest. Also added CrossValidator and TrainValidation split persistence to pyspark.
## How was this patch tested?
Performed both cross validation and train validation split with a one vs. rest estimator and tested read/write functionality of the estimator parameter maps required by these meta-algorithms.
Author: Ajay Saini <ajays725@gmail.com>
Closes#18428 from ajaysaini725/MetaAlgorithmPersistNestedEstimators.
Changes:
pyspark.ml Estimators can take either a list of param maps or a dict of params. This change allows the CrossValidator and TrainValidationSplit Estimators to pass through lists of param maps to the underlying estimators so that those estimators can handle parallelization when appropriate (eg distributed hyper parameter tuning).
Testing:
Existing unit tests.
Author: Bago Amirbekian <bago@databricks.com>
Closes#18077 from MrBago/delegate_params.
## What changes were proposed in this pull request?
The `keyword_only` decorator in PySpark is not thread-safe. It writes kwargs to a static class variable in the decorator, which is then retrieved later in the class method as `_input_kwargs`. If multiple threads are constructing the same class with different kwargs, it becomes a race condition to read from the static class variable before it's overwritten. See [SPARK-19348](https://issues.apache.org/jira/browse/SPARK-19348) for reproduction code.
This change will write the kwargs to a member variable so that multiple threads can operate on separate instances without the race condition. It does not protect against multiple threads operating on a single instance, but that is better left to the user to synchronize.
## How was this patch tested?
Added new unit tests for using the keyword_only decorator and a regression test that verifies `_input_kwargs` can be overwritten from different class instances.
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#16782 from BryanCutler/pyspark-keyword_only-threadsafe-SPARK-19348.
## What changes were proposed in this pull request?
avgMetrics was summed, not averaged, across folds
Author: =^_^= <maxmoroz@gmail.com>
Closes#14456 from pkch/pkch-patch-1.
## What changes were proposed in this pull request?
Updated ML pipeline Cross Validation Scaladoc & PyDoc.
## How was this patch tested?
Documentation update
(If this patch involves UI changes, please attach a screenshot; otherwise, remove this)
Author: krishnakalyan3 <krishnakalyan3@gmail.com>
Closes#13894 from krishnakalyan3/kfold-cv.
## What changes were proposed in this pull request?
General decisions to follow, except where noted:
* spark.mllib, pyspark.mllib: Remove all Experimental annotations. Leave DeveloperApi annotations alone.
* spark.ml, pyspark.ml
** Annotate Estimator-Model pairs of classes and companion objects the same way.
** For all algorithms marked Experimental with Since tag <= 1.6, remove Experimental annotation.
** For all algorithms marked Experimental with Since tag = 2.0, leave Experimental annotation.
* DeveloperApi annotations are left alone, except where noted.
* No changes to which types are sealed.
Exceptions where I am leaving items Experimental in spark.ml, pyspark.ml, mainly because the items are new:
* Model Summary classes
* MLWriter, MLReader, MLWritable, MLReadable
* Evaluator and subclasses: There is discussion of changes around evaluating multiple metrics at once for efficiency.
* RFormula: Its behavior may need to change slightly to match R in edge cases.
* AFTSurvivalRegression
* MultilayerPerceptronClassifier
DeveloperApi changes:
* ml.tree.Node, ml.tree.Split, and subclasses should no longer be DeveloperApi
## How was this patch tested?
N/A
Note to reviewers:
* spark.ml.clustering.LDA underwent significant changes (additional methods), so let me know if you want me to leave it Experimental.
* Be careful to check for cases where a class should no longer be Experimental but has an Experimental method, val, or other feature. I did not find such cases, but please verify.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#14147 from jkbradley/experimental-audit.
## What changes were proposed in this pull request?
Now we have PySpark picklers for new and old vector/matrix, individually. However, they are all implemented under `PythonMLlibAPI`. To separate spark.mllib from spark.ml, we should implement the picklers of new vector/matrix under `spark.ml.python` instead.
## How was this patch tested?
Existing tests.
Author: Liang-Chi Hsieh <simonh@tw.ibm.com>
Closes#13219 from viirya/pyspark-pickler-ml.
## What changes were proposed in this pull request?
Replace SQLContext and SparkContext with SparkSession using builder pattern in python test code.
## How was this patch tested?
Existing test.
Author: WeichenXu <WeichenXu123@outlook.com>
Closes#13242 from WeichenXu123/python_doctest_update_sparksession.
## What changes were proposed in this pull request?
This pull request includes supporting validationMetrics for TrainValidationSplitModel with Python and test for it.
## How was this patch tested?
test in `python/pyspark/ml/tests.py`
Author: Takuya Kuwahara <taakuu19@gmail.com>
Closes#12767 from taku-k/spark-14978.
## What changes were proposed in this pull request?
Once SPARK-14487 and SPARK-14549 are merged, we will migrate to use the new vector and matrix type in the new ml pipeline based apis.
## How was this patch tested?
Unit tests
Author: DB Tsai <dbt@netflix.com>
Author: Liang-Chi Hsieh <simonh@tw.ibm.com>
Author: Xiangrui Meng <meng@databricks.com>
Closes#12627 from dbtsai/SPARK-14615-NewML.
## What changes were proposed in this pull request?
Tag classes in ml.tuning as experimental, add docs for kfolds avg metric, and copy TrainValidationSplit scaladoc for more detailed explanation.
## How was this patch tested?
built docs locally
Author: Holden Karau <holden@us.ibm.com>
Closes#12967 from holdenk/SPARK-15195-pydoc-ml-tuning.
## What changes were proposed in this pull request?
PySpark ML Params setter code clean up.
For examples,
```setInputCol``` can be simplified from
```
self._set(inputCol=value)
return self
```
to:
```
return self._set(inputCol=value)
```
This is a pretty big sweeps, and we cleaned wherever possible.
## How was this patch tested?
Exist unit tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12749 from yanboliang/spark-14971.
## What changes were proposed in this pull request?
Per discussion on [https://github.com/apache/spark/pull/12604], this removes ML persistence for Python tuning (TrainValidationSplit, CrossValidator, and their Models) since they do not handle nesting easily. This support should be re-designed and added in the next release.
## How was this patch tested?
Removed unit test elements saving and loading the tuning algorithms, but kept tests to save and load their bestModel fields.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12782 from jkbradley/remove-python-tuning-saveload.
## What changes were proposed in this pull request?
support avgMetrics in CrossValidatorModel with Python
## How was this patch tested?
Doctest and `test_save_load` in `pyspark/ml/test.py`
[JIRA](https://issues.apache.org/jira/browse/SPARK-12810)
Author: Kai Jiang <jiangkai@gmail.com>
Closes#12464 from vectorijk/spark-12810.
## What changes were proposed in this pull request?
#11663 adds type conversion functionality for parameters in Pyspark. This PR find out the omissive ```Param``` that did not pass corresponding ```TypeConverter``` argument and fix them. After this PR, all params in pyspark/ml/ used ```TypeConverter```.
## How was this patch tested?
Existing tests.
cc jkbradley sethah
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12529 from yanboliang/typeConverter.
## What changes were proposed in this pull request?
This patch provides a first cut of python APIs for structured streaming. This PR provides the new classes:
- ContinuousQuery
- Trigger
- ProcessingTime
in pyspark under `pyspark.sql.streaming`.
In addition, it contains the new methods added under:
- `DataFrameWriter`
a) `startStream`
b) `trigger`
c) `queryName`
- `DataFrameReader`
a) `stream`
- `DataFrame`
a) `isStreaming`
This PR doesn't contain all methods exposed for `ContinuousQuery`, for example:
- `exception`
- `sourceStatuses`
- `sinkStatus`
They may be added in a follow up.
This PR also contains some very minor doc fixes in the Scala side.
## How was this patch tested?
Python doc tests
TODO:
- [ ] verify Python docs look good
Author: Burak Yavuz <brkyvz@gmail.com>
Author: Burak Yavuz <burak@databricks.com>
Closes#12320 from brkyvz/stream-python.
## What changes were proposed in this pull request?
Param setters in python previously accessed the _paramMap directly to update values. The `_set` method now implements type checking, so it should be used to update all parameters. This PR eliminates all direct accesses to `_paramMap` besides the one in the `_set` method to ensure type checking happens.
Additional changes:
* [SPARK-13068](https://github.com/apache/spark/pull/11663) missed adding type converters in evaluation.py so those are done here
* An incorrect `toBoolean` type converter was used for StringIndexer `handleInvalid` param in previous PR. This is fixed here.
## How was this patch tested?
Existing unit tests verify that parameters are still set properly. No new functionality is actually added in this PR.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#11939 from sethah/SPARK-14104.
Currently, JavaWrapper is only a wrapper class for pipeline classes that have Params and JavaCallable is a separate mixin that provides methods to make Java calls. This change simplifies the class structure and to define the Java wrapper in a plain base class along with methods to make Java calls. Also, renames Java wrapper classes to better reflect their purpose.
Ran existing Python ml tests and generated documentation to test this change.
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#12304 from BryanCutler/pyspark-cleanup-JavaWrapper-SPARK-14472.
## What changes were proposed in this pull request?
Cleanups to documentation. No changes to code.
* GBT docs: Move Scala doc for private object GradientBoostedTrees to public docs for GBTClassifier,Regressor
* GLM regParam: needs doc saying it is for L2 only
* TrainValidationSplitModel: add .. versionadded:: 2.0.0
* Rename “_transformer_params_from_java” to “_transfer_params_from_java”
* LogReg Summary classes: “probability” col should not say “calibrated”
* LR summaries: coefficientStandardErrors —> document that intercept stderr comes last. Same for t,p-values
* approxCountDistinct: Document meaning of “rsd" argument.
* LDA: note which params are for online LDA only
## How was this patch tested?
Doc build
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12266 from jkbradley/ml-doc-cleanups.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-13786
Add save/load for Python CrossValidator/Model and TrainValidationSplit/Model.
## How was this patch tested?
Test with Python doctest.
Author: Xusen Yin <yinxusen@gmail.com>
Closes#12020 from yinxusen/SPARK-13786.
## What changes were proposed in this pull request?
This patch adds type conversion functionality for parameters in Pyspark. A `typeConverter` field is added to the constructor of `Param` class. This argument is a function which converts values passed to this param to the appropriate type if possible. This is beneficial so that the params can fail at set time if they are given inappropriate values, but even more so because coherent error messages are now provided when Py4J cannot cast the python type to the appropriate Java type.
This patch also adds a `TypeConverters` class with factory methods for common type conversions. Most of the changes involve adding these factory type converters to existing params. The previous solution to this issue, `expectedType`, is deprecated and can be removed in 2.1.0 as discussed on the Jira.
## How was this patch tested?
Unit tests were added in python/pyspark/ml/tests.py to test parameter type conversion. These tests check that values that should be convertible are converted correctly, and that the appropriate errors are thrown when invalid values are provided.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#11663 from sethah/SPARK-13068-tc.
## What changes were proposed in this pull request?
The changes proposed were to add train-validation-split to pyspark.ml.tuning.
## How was the this patch tested?
This patch was tested through unit tests located in pyspark/ml/test.py.
This is my original work and I license it to Spark.
Author: JeremyNixon <jnixon2@gmail.com>
Closes#11335 from JeremyNixon/tvs_pyspark.
The current python ml params require cut-and-pasting the param setup and description between the class & ```__init__``` methods. Remove this possible case of errors & simplify use of custom params by adding a ```_copy_new_parent``` method to param so as to avoid cut and pasting (and cut and pasting at different indentation levels urgh).
Author: Holden Karau <holden@us.ibm.com>
Closes#10216 from holdenk/SPARK-10509-excessive-param-boiler-plate-code.
