10 commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
zero323
|
31a16fbb40 |
[SPARK-32714][PYTHON] Initial pyspark-stubs port
### What changes were proposed in this pull request? This PR proposes migration of [`pyspark-stubs`](https://github.com/zero323/pyspark-stubs) into Spark codebase. ### Why are the changes needed? ### Does this PR introduce _any_ user-facing change? Yes. This PR adds type annotations directly to Spark source. This can impact interaction with development tools for users, which haven't used `pyspark-stubs`. ### How was this patch tested? - [x] MyPy tests of the PySpark source ``` mypy --no-incremental --config python/mypy.ini python/pyspark ``` - [x] MyPy tests of Spark examples ``` MYPYPATH=python/ mypy --no-incremental --config python/mypy.ini examples/src/main/python/ml examples/src/main/python/sql examples/src/main/python/sql/streaming ``` - [x] Existing Flake8 linter - [x] Existing unit tests Tested against: - `mypy==0.790+dev.e959952d9001e9713d329a2f9b196705b028f894` - `mypy==0.782` Closes #29591 from zero323/SPARK-32681. Authored-by: zero323 <mszymkiewicz@gmail.com> Signed-off-by: HyukjinKwon <gurwls223@apache.org> |
||
Louiszr
|
a0bd273bb0 |
[SPARK-32092][ML][PYSPARK][FOLLOWUP] Fixed CrossValidatorModel.copy() to copy models instead of list
### 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> |
||
|
Louiszr
|
d9eb06ea37 |
[SPARK-32092][ML][PYSPARK] Fix parameters not being copied in CrossValidatorModel.copy(), read() and write()
### 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> |
||
Fokko Driesprong
|
9fcf0ea718 |
[SPARK-32319][PYSPARK] Disallow the use of unused imports
Disallow the use of unused imports: - Unnecessary increases the memory footprint of the application - Removes the imports that are required for the examples in the docstring from the file-scope to the example itself. This keeps the files itself clean, and gives a more complete example as it also includes the imports :) ``` fokkodriesprongFan spark % flake8 python | grep -i "imported but unused" python/pyspark/cloudpickle.py:46:1: F401 'functools.partial' imported but unused python/pyspark/cloudpickle.py:55:1: F401 'traceback' imported but unused python/pyspark/heapq3.py:868:5: F401 '_heapq.*' imported but unused python/pyspark/__init__.py:61:1: F401 'pyspark.version.__version__' imported but unused python/pyspark/__init__.py:62:1: F401 'pyspark._globals._NoValue' imported but unused python/pyspark/__init__.py:115:1: F401 'pyspark.sql.SQLContext' imported but unused python/pyspark/__init__.py:115:1: F401 'pyspark.sql.HiveContext' imported but unused python/pyspark/__init__.py:115:1: F401 'pyspark.sql.Row' imported but unused python/pyspark/rdd.py:21:1: F401 're' imported but unused python/pyspark/rdd.py:29:1: F401 'tempfile.NamedTemporaryFile' imported but unused python/pyspark/mllib/regression.py:26:1: F401 'pyspark.mllib.linalg.SparseVector' imported but unused python/pyspark/mllib/clustering.py:28:1: F401 'pyspark.mllib.linalg.SparseVector' imported but unused python/pyspark/mllib/clustering.py:28:1: F401 'pyspark.mllib.linalg.DenseVector' imported but unused python/pyspark/mllib/classification.py:26:1: F401 'pyspark.mllib.linalg.SparseVector' imported but unused python/pyspark/mllib/feature.py:28:1: F401 'pyspark.mllib.linalg.DenseVector' imported but unused python/pyspark/mllib/feature.py:28:1: F401 'pyspark.mllib.linalg.SparseVector' imported but unused python/pyspark/mllib/feature.py:30:1: F401 'pyspark.mllib.regression.LabeledPoint' imported but unused python/pyspark/mllib/tests/test_linalg.py:18:1: F401 'sys' imported but unused python/pyspark/mllib/tests/test_linalg.py:642:5: F401 'pyspark.mllib.tests.test_linalg.*' imported but unused python/pyspark/mllib/tests/test_feature.py:21:1: F401 'numpy.random' imported but unused python/pyspark/mllib/tests/test_feature.py:21:1: F401 'numpy.exp' imported but unused python/pyspark/mllib/tests/test_feature.py:23:1: F401 'pyspark.mllib.linalg.Vector' imported but unused python/pyspark/mllib/tests/test_feature.py:23:1: F401 'pyspark.mllib.linalg.VectorUDT' imported but unused python/pyspark/mllib/tests/test_feature.py:185:5: F401 'pyspark.mllib.tests.test_feature.*' imported but unused python/pyspark/mllib/tests/test_util.py:97:5: F401 'pyspark.mllib.tests.test_util.*' imported but unused python/pyspark/mllib/tests/test_stat.py:23:1: F401 'pyspark.mllib.linalg.Vector' imported but unused python/pyspark/mllib/tests/test_stat.py:23:1: F401 'pyspark.mllib.linalg.SparseVector' imported but unused python/pyspark/mllib/tests/test_stat.py:23:1: F401 'pyspark.mllib.linalg.DenseVector' imported but unused python/pyspark/mllib/tests/test_stat.py:23:1: F401 'pyspark.mllib.linalg.VectorUDT' imported but unused python/pyspark/mllib/tests/test_stat.py:23:1: F401 'pyspark.mllib.linalg._convert_to_vector' imported but unused python/pyspark/mllib/tests/test_stat.py:23:1: F401 'pyspark.mllib.linalg.DenseMatrix' imported but unused python/pyspark/mllib/tests/test_stat.py:23:1: F401 'pyspark.mllib.linalg.SparseMatrix' imported but unused python/pyspark/mllib/tests/test_stat.py:23:1: F401 'pyspark.mllib.linalg.MatrixUDT' imported but unused python/pyspark/mllib/tests/test_stat.py:181:5: F401 'pyspark.mllib.tests.test_stat.*' imported but unused python/pyspark/mllib/tests/test_streaming_algorithms.py:18:1: F401 'time.time' imported but unused python/pyspark/mllib/tests/test_streaming_algorithms.py:18:1: F401 'time.sleep' imported but unused python/pyspark/mllib/tests/test_streaming_algorithms.py:470:5: F401 'pyspark.mllib.tests.test_streaming_algorithms.*' imported but unused python/pyspark/mllib/tests/test_algorithms.py:295:5: F401 'pyspark.mllib.tests.test_algorithms.*' imported but unused python/pyspark/tests/test_serializers.py:90:13: F401 'xmlrunner' imported but unused python/pyspark/tests/test_rdd.py:21:1: F401 'sys' imported but unused python/pyspark/tests/test_rdd.py:29:1: F401 'pyspark.resource.ResourceProfile' imported but unused python/pyspark/tests/test_rdd.py:885:5: F401 'pyspark.tests.test_rdd.*' imported but unused python/pyspark/tests/test_readwrite.py:19:1: F401 'sys' imported but unused python/pyspark/tests/test_readwrite.py:22:1: F401 'array.array' imported but unused python/pyspark/tests/test_readwrite.py:309:5: F401 'pyspark.tests.test_readwrite.*' imported but unused python/pyspark/tests/test_join.py:62:5: F401 'pyspark.tests.test_join.*' imported but unused python/pyspark/tests/test_taskcontext.py:19:1: F401 'shutil' imported but unused python/pyspark/tests/test_taskcontext.py:325:5: F401 'pyspark.tests.test_taskcontext.*' imported but unused python/pyspark/tests/test_conf.py:36:5: F401 'pyspark.tests.test_conf.*' imported but unused python/pyspark/tests/test_broadcast.py:148:5: F401 'pyspark.tests.test_broadcast.*' imported but unused python/pyspark/tests/test_daemon.py:76:5: F401 'pyspark.tests.test_daemon.*' imported but unused python/pyspark/tests/test_util.py:77:5: F401 'pyspark.tests.test_util.*' imported but unused python/pyspark/tests/test_pin_thread.py:19:1: F401 'random' imported but unused python/pyspark/tests/test_pin_thread.py:149:5: F401 'pyspark.tests.test_pin_thread.*' imported but unused python/pyspark/tests/test_worker.py:19:1: F401 'sys' imported but unused python/pyspark/tests/test_worker.py:26:5: F401 'resource' imported but unused python/pyspark/tests/test_worker.py:203:5: F401 'pyspark.tests.test_worker.*' imported but unused python/pyspark/tests/test_profiler.py:101:5: F401 'pyspark.tests.test_profiler.*' imported but unused python/pyspark/tests/test_shuffle.py:18:1: F401 'sys' imported but unused python/pyspark/tests/test_shuffle.py:171:5: F401 'pyspark.tests.test_shuffle.*' imported but unused python/pyspark/tests/test_rddbarrier.py:43:5: F401 'pyspark.tests.test_rddbarrier.*' imported but unused python/pyspark/tests/test_context.py:129:13: F401 'userlibrary.UserClass' imported but unused python/pyspark/tests/test_context.py:140:13: F401 'userlib.UserClass' imported but unused python/pyspark/tests/test_context.py:310:5: F401 'pyspark.tests.test_context.*' imported but unused python/pyspark/tests/test_appsubmit.py:241:5: F401 'pyspark.tests.test_appsubmit.*' imported but unused python/pyspark/streaming/dstream.py:18:1: F401 'sys' imported but unused python/pyspark/streaming/tests/test_dstream.py:27:1: F401 'pyspark.RDD' imported but unused python/pyspark/streaming/tests/test_dstream.py:647:5: F401 'pyspark.streaming.tests.test_dstream.*' imported but unused python/pyspark/streaming/tests/test_kinesis.py:83:5: F401 'pyspark.streaming.tests.test_kinesis.*' imported but unused python/pyspark/streaming/tests/test_listener.py:152:5: F401 'pyspark.streaming.tests.test_listener.*' imported but unused python/pyspark/streaming/tests/test_context.py:178:5: F401 'pyspark.streaming.tests.test_context.*' imported but unused python/pyspark/testing/utils.py:30:5: F401 'scipy.sparse' imported but unused python/pyspark/testing/utils.py:36:5: F401 'numpy as np' imported but unused python/pyspark/ml/regression.py:25:1: F401 'pyspark.ml.tree._TreeEnsembleParams' imported but unused python/pyspark/ml/regression.py:25:1: F401 'pyspark.ml.tree._HasVarianceImpurity' imported but unused python/pyspark/ml/regression.py:29:1: F401 'pyspark.ml.wrapper.JavaParams' imported but unused python/pyspark/ml/util.py:19:1: F401 'sys' imported but unused python/pyspark/ml/__init__.py:25:1: F401 'pyspark.ml.pipeline' imported but unused python/pyspark/ml/pipeline.py:18:1: F401 'sys' imported but unused python/pyspark/ml/stat.py:22:1: F401 'pyspark.ml.linalg.DenseMatrix' imported but unused python/pyspark/ml/stat.py:22:1: F401 'pyspark.ml.linalg.Vectors' imported but unused python/pyspark/ml/tests/test_training_summary.py:18:1: F401 'sys' imported but unused python/pyspark/ml/tests/test_training_summary.py:364:5: F401 'pyspark.ml.tests.test_training_summary.*' imported but unused python/pyspark/ml/tests/test_linalg.py:381:5: F401 'pyspark.ml.tests.test_linalg.*' imported but unused python/pyspark/ml/tests/test_tuning.py:427:9: F401 'pyspark.sql.functions as F' imported but unused python/pyspark/ml/tests/test_tuning.py:757:5: F401 'pyspark.ml.tests.test_tuning.*' imported but unused python/pyspark/ml/tests/test_wrapper.py:120:5: F401 'pyspark.ml.tests.test_wrapper.*' imported but unused python/pyspark/ml/tests/test_feature.py:19:1: F401 'sys' imported but unused python/pyspark/ml/tests/test_feature.py:304:5: F401 'pyspark.ml.tests.test_feature.*' imported but unused python/pyspark/ml/tests/test_image.py:19:1: F401 'py4j' imported but unused python/pyspark/ml/tests/test_image.py:22:1: F401 'pyspark.testing.mlutils.PySparkTestCase' imported but unused python/pyspark/ml/tests/test_image.py:71:5: F401 'pyspark.ml.tests.test_image.*' imported but unused python/pyspark/ml/tests/test_persistence.py:456:5: F401 'pyspark.ml.tests.test_persistence.*' imported but unused python/pyspark/ml/tests/test_evaluation.py:56:5: F401 'pyspark.ml.tests.test_evaluation.*' imported but unused python/pyspark/ml/tests/test_stat.py:43:5: F401 'pyspark.ml.tests.test_stat.*' imported but unused python/pyspark/ml/tests/test_base.py:70:5: F401 'pyspark.ml.tests.test_base.*' imported but unused python/pyspark/ml/tests/test_param.py:20:1: F401 'sys' imported but unused python/pyspark/ml/tests/test_param.py:375:5: F401 'pyspark.ml.tests.test_param.*' imported but unused python/pyspark/ml/tests/test_pipeline.py:62:5: F401 'pyspark.ml.tests.test_pipeline.*' imported but unused python/pyspark/ml/tests/test_algorithms.py:333:5: F401 'pyspark.ml.tests.test_algorithms.*' imported but unused python/pyspark/ml/param/__init__.py:18:1: F401 'sys' imported but unused python/pyspark/resource/tests/test_resources.py:17:1: F401 'random' imported but unused python/pyspark/resource/tests/test_resources.py:20:1: F401 'pyspark.resource.ResourceProfile' imported but unused python/pyspark/resource/tests/test_resources.py:75:5: F401 'pyspark.resource.tests.test_resources.*' imported but unused python/pyspark/sql/functions.py:32:1: F401 'pyspark.sql.udf.UserDefinedFunction' imported but unused python/pyspark/sql/functions.py:34:1: F401 'pyspark.sql.pandas.functions.pandas_udf' imported but unused python/pyspark/sql/session.py:30:1: F401 'pyspark.sql.types.Row' imported but unused python/pyspark/sql/session.py:30:1: F401 'pyspark.sql.types.StringType' imported but unused python/pyspark/sql/readwriter.py:1084:5: F401 'pyspark.sql.Row' imported but unused python/pyspark/sql/context.py:26:1: F401 'pyspark.sql.types.IntegerType' imported but unused python/pyspark/sql/context.py:26:1: F401 'pyspark.sql.types.Row' imported but unused python/pyspark/sql/context.py:26:1: F401 'pyspark.sql.types.StringType' imported but unused python/pyspark/sql/context.py:27:1: F401 'pyspark.sql.udf.UDFRegistration' imported but unused python/pyspark/sql/streaming.py:1212:5: F401 'pyspark.sql.Row' imported but unused python/pyspark/sql/tests/test_utils.py:55:5: F401 'pyspark.sql.tests.test_utils.*' imported but unused python/pyspark/sql/tests/test_pandas_map.py:18:1: F401 'sys' imported but unused python/pyspark/sql/tests/test_pandas_map.py:22:1: F401 'pyspark.sql.functions.pandas_udf' imported but unused python/pyspark/sql/tests/test_pandas_map.py:22:1: F401 'pyspark.sql.functions.PandasUDFType' imported but unused python/pyspark/sql/tests/test_pandas_map.py:119:5: F401 'pyspark.sql.tests.test_pandas_map.*' imported but unused python/pyspark/sql/tests/test_catalog.py:193:5: F401 'pyspark.sql.tests.test_catalog.*' imported but unused python/pyspark/sql/tests/test_group.py:39:5: F401 'pyspark.sql.tests.test_group.*' imported but unused python/pyspark/sql/tests/test_session.py:361:5: F401 'pyspark.sql.tests.test_session.*' imported but unused python/pyspark/sql/tests/test_conf.py:49:5: F401 'pyspark.sql.tests.test_conf.*' imported but unused python/pyspark/sql/tests/test_pandas_cogrouped_map.py:19:1: F401 'sys' imported but unused python/pyspark/sql/tests/test_pandas_cogrouped_map.py:21:1: F401 'pyspark.sql.functions.sum' imported but unused python/pyspark/sql/tests/test_pandas_cogrouped_map.py:21:1: F401 'pyspark.sql.functions.PandasUDFType' imported but unused python/pyspark/sql/tests/test_pandas_cogrouped_map.py:29:5: F401 'pandas.util.testing.assert_series_equal' imported but unused python/pyspark/sql/tests/test_pandas_cogrouped_map.py:32:5: F401 'pyarrow as pa' imported but unused python/pyspark/sql/tests/test_pandas_cogrouped_map.py:248:5: F401 'pyspark.sql.tests.test_pandas_cogrouped_map.*' imported but unused python/pyspark/sql/tests/test_udf.py:24:1: F401 'py4j' imported but unused python/pyspark/sql/tests/test_pandas_udf_typehints.py:246:5: F401 'pyspark.sql.tests.test_pandas_udf_typehints.*' imported but unused python/pyspark/sql/tests/test_functions.py:19:1: F401 'sys' imported but unused python/pyspark/sql/tests/test_functions.py:362:9: F401 'pyspark.sql.functions.exists' imported but unused python/pyspark/sql/tests/test_functions.py:387:5: F401 'pyspark.sql.tests.test_functions.*' imported but unused python/pyspark/sql/tests/test_pandas_udf_scalar.py:21:1: F401 'sys' imported but unused python/pyspark/sql/tests/test_pandas_udf_scalar.py:45:5: F401 'pyarrow as pa' imported but unused python/pyspark/sql/tests/test_pandas_udf_window.py:355:5: F401 'pyspark.sql.tests.test_pandas_udf_window.*' imported but unused python/pyspark/sql/tests/test_arrow.py:38:5: F401 'pyarrow as pa' imported but unused python/pyspark/sql/tests/test_pandas_grouped_map.py:20:1: F401 'sys' imported but unused python/pyspark/sql/tests/test_pandas_grouped_map.py:38:5: F401 'pyarrow as pa' imported but unused python/pyspark/sql/tests/test_dataframe.py:382:9: F401 'pyspark.sql.DataFrame' imported but unused python/pyspark/sql/avro/functions.py:125:5: F401 'pyspark.sql.Row' imported but unused python/pyspark/sql/pandas/functions.py:19:1: F401 'sys' imported but unused ``` After: ``` fokkodriesprongFan spark % flake8 python | grep -i "imported but unused" fokkodriesprongFan spark % ``` ### What changes were proposed in this pull request? Removing unused imports from the Python files to keep everything nice and tidy. ### Why are the changes needed? Cleaning up of the imports that aren't used, and suppressing the imports that are used as references to other modules, preserving backward compatibility. ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? Adding the rule to the existing Flake8 checks. Closes #29121 from Fokko/SPARK-32319. Authored-by: Fokko Driesprong <fokko@apache.org> Signed-off-by: Dongjoon Hyun <dongjoon@apache.org> |
||
Liang-Chi Hsieh
|
7f6a8ab166 |
[SPARK-31777][ML][PYSPARK] Add user-specified fold column to CrossValidator
### 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> |
||
Weichen Xu
|
4a21c4cc92 |
[SPARK-31497][ML][PYSPARK] Fix Pyspark CrossValidator/TrainValidationSplit with pipeline estimator cannot save and load model
### What changes were proposed in this pull request?
Fix Pyspark CrossValidator/TrainValidationSplit with pipeline estimator cannot save and load model.
Most pyspark estimators/transformers inherit `JavaParams`, but some estimators are special (in order to support pure python implemented nested estimators/transformers):
* Pipeline
* OneVsRest
* CrossValidator
* TrainValidationSplit
But note that, currently, in pyspark, estimators listed above, their model reader/writer do NOT support pure python implemented nested estimators/transformers. Because they use java reader/writer wrapper as python side reader/writer.
Pyspark CrossValidator/TrainValidationSplit model reader/writer require all estimators define the `_transfer_param_map_to_java` and `_transfer_param_map_from_java` (used in model read/write).
OneVsRest class already defines the two methods, but Pipeline do not, so it lead to this bug.
In this PR I add `_transfer_param_map_to_java` and `_transfer_param_map_from_java` into Pipeline class.
### Why are the changes needed?
Bug fix.
### Does this PR introduce any user-facing change?
No
### How was this patch tested?
Unit test.
Manually test in pyspark shell:
1) CrossValidator with Simple Pipeline estimator
```
from pyspark.ml import Pipeline
from pyspark.ml.classification import LogisticRegression
from pyspark.ml.evaluation import BinaryClassificationEvaluator
from pyspark.ml.feature import HashingTF, Tokenizer
from pyspark.ml.tuning import CrossValidator, CrossValidatorModel, ParamGridBuilder
training = spark.createDataFrame([
(0, "a b c d e spark", 1.0),
(1, "b d", 0.0),
(2, "spark f g h", 1.0),
(3, "hadoop mapreduce", 0.0),
(4, "b spark who", 1.0),
(5, "g d a y", 0.0),
(6, "spark fly", 1.0),
(7, "was mapreduce", 0.0),
], ["id", "text", "label"])
# Configure an ML pipeline, which consists of tree stages: tokenizer, hashingTF, and lr.
tokenizer = Tokenizer(inputCol="text", outputCol="words")
hashingTF = HashingTF(inputCol=tokenizer.getOutputCol(), outputCol="features")
lr = LogisticRegression(maxIter=10)
pipeline = Pipeline(stages=[tokenizer, hashingTF, lr])
paramGrid = ParamGridBuilder() \
.addGrid(hashingTF.numFeatures, [10, 100, 1000]) \
.addGrid(lr.regParam, [0.1, 0.01]) \
.build()
crossval = CrossValidator(estimator=pipeline,
estimatorParamMaps=paramGrid,
evaluator=BinaryClassificationEvaluator(),
numFolds=2) # use 3+ folds in practice
# Run cross-validation, and choose the best set of parameters.
cvModel = crossval.fit(training)
cvModel.save('/tmp/cv_model001')
CrossValidatorModel.load('/tmp/cv_model001')
```
2) CrossValidator with Pipeline estimator which include a OneVsRest estimator stage, and OneVsRest estimator nest a LogisticRegression estimator.
```
from pyspark.ml.linalg import Vectors
from pyspark.ml import Estimator, Model
from pyspark.ml.classification import LogisticRegression, LogisticRegressionModel, OneVsRest
from pyspark.ml.evaluation import BinaryClassificationEvaluator, \
MulticlassClassificationEvaluator, RegressionEvaluator
from pyspark.ml.linalg import Vectors
from pyspark.ml.param import Param, Params
from pyspark.ml.tuning import CrossValidator, CrossValidatorModel, ParamGridBuilder, \
TrainValidationSplit, TrainValidationSplitModel
from pyspark.sql.functions import rand
from pyspark.testing.mlutils import SparkSessionTestCase
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"])
ova = OneVsRest(classifier=LogisticRegression())
lr1 = LogisticRegression().setMaxIter(100)
lr2 = LogisticRegression().setMaxIter(150)
grid = ParamGridBuilder().addGrid(ova.classifier, [lr1, lr2]).build()
evaluator = MulticlassClassificationEvaluator()
pipeline = Pipeline(stages=[ova])
cv = CrossValidator(estimator=pipeline, estimatorParamMaps=grid, evaluator=evaluator)
cvModel = cv.fit(dataset)
cvModel.save('/tmp/model002')
cvModel2 = CrossValidatorModel.load('/tmp/model002')
```
TrainValidationSplit testing code are similar so I do not paste them.
Closes #28279 from WeichenXu123/fix_pipeline_tuning.
Authored-by: Weichen Xu <weichen.xu@databricks.com>
Signed-off-by: Xiangrui Meng <meng@databricks.com>
|
||
John Bauer
|
e804ed5e33 |
[SPARK-29691][ML][PYTHON] ensure Param objects are valid in fit, transform
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> |
||
HyukjinKwon
|
7c05f61514 |
[SPARK-28130][PYTHON] Print pretty messages for skipped tests when xmlrunner is available in PySpark
## What changes were proposed in this pull request?
Currently, pretty skipped message added by
|
||
|
hyukjinkwon
|
bbbdaa82a4 |
[SPARK-26105][PYTHON] Clean unittest2 imports up that were added for Python 2.6 before
## What changes were proposed in this pull request?
Currently, some of PySpark tests sill assume the tests could be ran in Python 2.6 by importing `unittest2`. For instance:
```python
if sys.version_info[:2] <= (2, 6):
try:
import unittest2 as unittest
except ImportError:
sys.stderr.write('Please install unittest2 to test with Python 2.6 or earlier')
sys.exit(1)
else:
import unittest
```
While I am here, I removed some of unused imports and reordered imports per PEP 8.
We officially dropped Python 2.6 support a while ago and started to discuss about Python 2 drop. It's better to remove them out.
## How was this patch tested?
Manually tests, and existing tests via Jenkins.
Closes #23077 from HyukjinKwon/SPARK-26105.
Lead-authored-by: hyukjinkwon <gurwls223@apache.org>
Co-authored-by: Bryan Cutler <cutlerb@gmail.com>
Signed-off-by: hyukjinkwon <gurwls223@apache.org>
|
||
Bryan Cutler
|
034ae305c3 |
[SPARK-26033][PYTHON][TESTS] Break large ml/tests.py file into smaller files
## What changes were proposed in this pull request? This PR breaks down the large ml/tests.py file that contains all Python ML unit tests into several smaller test files to be easier to read and maintain. The tests are broken down as follows: ``` pyspark ├── __init__.py ... ├── ml │ ├── __init__.py ... │ ├── tests │ │ ├── __init__.py │ │ ├── test_algorithms.py │ │ ├── test_base.py │ │ ├── test_evaluation.py │ │ ├── test_feature.py │ │ ├── test_image.py │ │ ├── test_linalg.py │ │ ├── test_param.py │ │ ├── test_persistence.py │ │ ├── test_pipeline.py │ │ ├── test_stat.py │ │ ├── test_training_summary.py │ │ ├── test_tuning.py │ │ └── test_wrapper.py ... ├── testing ... │ ├── mlutils.py ... ``` ## How was this patch tested? Ran tests manually by module to ensure test count was the same, and ran `python/run-tests --modules=pyspark-ml` to verify all passing with Python 2.7 and Python 3.6. Closes #23063 from BryanCutler/python-test-breakup-ml-SPARK-26033. Authored-by: Bryan Cutler <cutlerb@gmail.com> Signed-off-by: hyukjinkwon <gurwls223@apache.org> |