snowpark-connect 0.24.0__py3-none-any.whl → 0.25.0__py3-none-any.whl

snowflake/snowpark_connect/includes/python/pyspark/ml/tests/connect/test_legacy_mode_tuning.py

@@ -1,292 +0,0 @@
-# -*- coding: utf-8 -*-
-#
-# Licensed to the Apache Software Foundation (ASF) under one or more
-# contributor license agreements.  See the NOTICE file distributed with
-# this work for additional information regarding copyright ownership.
-# The ASF licenses this file to You under the Apache License, Version 2.0
-# (the "License"); you may not use this file except in compliance with
-# the License.  You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-
-import tempfile
-import unittest
-import numpy as np
-import pandas as pd
-from pyspark.ml.param import Param, Params
-from pyspark.ml.connect import Model, Estimator
-from pyspark.ml.connect.feature import StandardScaler
-from pyspark.ml.connect.classification import LogisticRegression as LORV2
-from pyspark.ml.connect.pipeline import Pipeline
-from pyspark.ml.connect.tuning import CrossValidator, CrossValidatorModel
-from pyspark.ml.connect.evaluation import BinaryClassificationEvaluator, RegressionEvaluator
-from pyspark.ml.tuning import ParamGridBuilder
-from pyspark.sql import SparkSession
-from pyspark.sql.functions import rand
-
-from sklearn.datasets import load_breast_cancer
-
-
-have_torch = True
-try:
-    import torch  # noqa: F401
-except ImportError:
-    have_torch = False
-
-
-class HasInducedError(Params):
-    def __init__(self):
-        super(HasInducedError, self).__init__()
-        self.inducedError = Param(
-            self, "inducedError", "Uniformly-distributed error added to feature"
-        )
-
-    def getInducedError(self):
-        return self.getOrDefault(self.inducedError)
-
-
-class InducedErrorModel(Model, HasInducedError):
-    def __init__(self):
-        super(InducedErrorModel, self).__init__()
-
-    def _transform(self, dataset):
-        return dataset.withColumn(
-            "prediction", dataset.feature + (rand(0) * self.getInducedError())
-        )
-
-
-class InducedErrorEstimator(Estimator, HasInducedError):
-    def __init__(self, inducedError=1.0):
-        super(InducedErrorEstimator, self).__init__()
-        self._set(inducedError=inducedError)
-
-    def _fit(self, dataset):
-        model = InducedErrorModel()
-        self._copyValues(model)
-        return model
-
-
-class CrossValidatorTestsMixin:
-    def test_gen_avg_and_std_metrics(self):
-        metrics_all = [
-            [1.0, 3.0, 2.0, 4.0],
-            [3.0, 2.0, 2.0, 4.0],
-            [3.0, 2.5, 2.1, 8.0],
-        ]
-        avg_metrics, std_metrics = CrossValidator._gen_avg_and_std_metrics(metrics_all)
-        assert np.allclose(avg_metrics, [2.33333333, 2.5, 2.03333333, 5.33333333])
-        assert np.allclose(std_metrics, [0.94280904, 0.40824829, 0.04714045, 1.88561808])
-        assert isinstance(avg_metrics, list)
-        assert isinstance(std_metrics, list)
-
-    def test_copy(self):
-        dataset = self.spark.createDataFrame(
-            [(10, 10.0), (50, 50.0), (100, 100.0), (500, 500.0)] * 10, ["feature", "label"]
-        )
-
-        iee = InducedErrorEstimator()
-        evaluator = RegressionEvaluator(metricName="rmse")
-
-        grid = ParamGridBuilder().addGrid(iee.inducedError, [100.0, 0.0, 10000.0]).build()
-        cv = CrossValidator(
-            estimator=iee,
-            estimatorParamMaps=grid,
-            evaluator=evaluator,
-            numFolds=2,
-        )
-        cvCopied = cv.copy()
-        for param in [
-            lambda x: x.getEstimator().uid,
-            # SPARK-32092: CrossValidator.copy() needs to copy all existing params
-            lambda x: x.getNumFolds(),
-            lambda x: x.getFoldCol(),
-            lambda x: x.getParallelism(),
-            lambda x: x.getSeed(),
-        ]:
-            self.assertEqual(param(cv), param(cvCopied))
-
-        cvModel = cv.fit(dataset)
-        cvModelCopied = cvModel.copy()
-        for index in range(len(cvModel.avgMetrics)):
-            self.assertTrue(
-                abs(cvModel.avgMetrics[index] - cvModelCopied.avgMetrics[index]) < 0.0001
-            )
-        self.assertTrue(np.allclose(cvModel.stdMetrics, cvModelCopied.stdMetrics))
-        # SPARK-32092: CrossValidatorModel.copy() needs to copy all existing params
-        for param in [lambda x: x.getNumFolds(), lambda x: x.getFoldCol(), lambda x: x.getSeed()]:
-            self.assertEqual(param(cvModel), param(cvModelCopied))
-
-        cvModel.avgMetrics[0] = "foo"
-        self.assertNotEqual(
-            cvModelCopied.avgMetrics[0],
-            "foo",
-            "Changing the original avgMetrics should not affect the copied model",
-        )
-        cvModel.stdMetrics[0] = "foo"
-        self.assertNotEqual(
-            cvModelCopied.stdMetrics[0],
-            "foo",
-            "Changing the original stdMetrics should not affect the copied model",
-        )
-
-    def test_fit_minimize_metric(self):
-        dataset = self.spark.createDataFrame(
-            [(10, 10.0), (50, 50.0), (100, 100.0), (500, 500.0)] * 10, ["feature", "label"]
-        )
-
-        iee = InducedErrorEstimator()
-        evaluator = RegressionEvaluator(metricName="rmse")
-
-        grid = ParamGridBuilder().addGrid(iee.inducedError, [100.0, 0.0, 10000.0]).build()
-        cv = CrossValidator(estimator=iee, estimatorParamMaps=grid, evaluator=evaluator)
-        cvModel = cv.fit(dataset)
-        bestModel = cvModel.bestModel
-        bestModelMetric = evaluator.evaluate(bestModel.transform(dataset))
-
-        self.assertEqual(
-            0.0, bestModel.getOrDefault("inducedError"), "Best model should have zero induced error"
-        )
-        self.assertEqual(0.0, bestModelMetric, "Best model has RMSE of 0")
-
-    def test_fit_maximize_metric(self):
-        dataset = self.spark.createDataFrame(
-            [(10, 10.0), (50, 50.0), (100, 100.0), (500, 500.0)] * 10, ["feature", "label"]
-        )
-
-        iee = InducedErrorEstimator()
-        evaluator = RegressionEvaluator(metricName="r2")
-
-        grid = ParamGridBuilder().addGrid(iee.inducedError, [100.0, 0.0, 10000.0]).build()
-        cv = CrossValidator(estimator=iee, estimatorParamMaps=grid, evaluator=evaluator)
-        cvModel = cv.fit(dataset)
-        bestModel = cvModel.bestModel
-        bestModelMetric = evaluator.evaluate(bestModel.transform(dataset))
-
-        self.assertEqual(
-            0.0, bestModel.getOrDefault("inducedError"), "Best model should have zero induced error"
-        )
-        self.assertEqual(1.0, bestModelMetric, "Best model has R-squared of 1")
-
-    @staticmethod
-    def _check_result(result_dataframe, expected_predictions, expected_probabilities=None):
-        np.testing.assert_array_equal(list(result_dataframe.prediction), expected_predictions)
-        if "probability" in result_dataframe.columns:
-            np.testing.assert_allclose(
-                list(result_dataframe.probability),
-                expected_probabilities,
-                rtol=1e-1,
-            )
-
-    def test_crossvalidator_on_pipeline(self):
-        sk_dataset = load_breast_cancer()
-
-        train_dataset = self.spark.createDataFrame(
-            zip(sk_dataset.data.tolist(), [int(t) for t in sk_dataset.target]),
-            schema="features: array<double>, label: long",
-        )
-
-        scaler = StandardScaler(inputCol="features", outputCol="scaled_features")
-        lorv2 = LORV2(numTrainWorkers=2, featuresCol="scaled_features")
-        pipeline = Pipeline(stages=[scaler, lorv2])
-
-        grid2 = ParamGridBuilder().addGrid(lorv2.maxIter, [2, 200]).build()
-        cv = CrossValidator(
-            estimator=pipeline,
-            estimatorParamMaps=grid2,
-            parallelism=2,
-            evaluator=BinaryClassificationEvaluator(),
-        )
-        cv_model = cv.fit(train_dataset)
-        transformed_result = (
-            cv_model.transform(train_dataset).select("prediction", "probability").toPandas()
-        )
-        expected_transformed_result = (
-            cv_model.bestModel.transform(train_dataset)
-            .select("prediction", "probability")
-            .toPandas()
-        )
-        pd.testing.assert_frame_equal(transformed_result, expected_transformed_result)
-
-        assert cv_model.bestModel.stages[1].getMaxIter() == 200
-
-        # trial of index 2 should have better metric value
-        # because it sets higher `maxIter` param.
-        assert cv_model.avgMetrics[1] > cv_model.avgMetrics[0]
-
-        def _verify_cv_saved_params(instance, loaded_instance):
-            assert instance.getEstimator().uid == loaded_instance.getEstimator().uid
-            assert instance.getEvaluator().uid == loaded_instance.getEvaluator().uid
-            assert instance.getEstimatorParamMaps() == loaded_instance.getEstimatorParamMaps()
-
-        # Test save / load
-        with tempfile.TemporaryDirectory() as tmp_dir:
-            cv.saveToLocal(f"{tmp_dir}/cv")
-            loaded_cv = CrossValidator.loadFromLocal(f"{tmp_dir}/cv")
-
-            _verify_cv_saved_params(cv, loaded_cv)
-
-            cv_model.saveToLocal(f"{tmp_dir}/cv_model")
-            loaded_cv_model = CrossValidatorModel.loadFromLocal(f"{tmp_dir}/cv_model")
-
-            _verify_cv_saved_params(cv_model, loaded_cv_model)
-
-            assert cv_model.uid == loaded_cv_model.uid
-            assert cv_model.bestModel.uid == loaded_cv_model.bestModel.uid
-            assert cv_model.bestModel.stages[0].uid == loaded_cv_model.bestModel.stages[0].uid
-            assert cv_model.bestModel.stages[1].uid == loaded_cv_model.bestModel.stages[1].uid
-            assert loaded_cv_model.bestModel.stages[1].getMaxIter() == 200
-
-            np.testing.assert_allclose(cv_model.avgMetrics, loaded_cv_model.avgMetrics)
-            np.testing.assert_allclose(cv_model.stdMetrics, loaded_cv_model.stdMetrics)
-
-    def test_crossvalidator_with_fold_col(self):
-        sk_dataset = load_breast_cancer()
-
-        train_dataset = self.spark.createDataFrame(
-            zip(
-                sk_dataset.data.tolist(),
-                [int(t) for t in sk_dataset.target],
-                [int(i % 3) for i in range(len(sk_dataset.target))],
-            ),
-            schema="features: array<double>, label: long, fold: long",
-        )
-
-        lorv2 = LORV2(numTrainWorkers=2)
-
-        grid2 = ParamGridBuilder().addGrid(lorv2.maxIter, [2, 200]).build()
-        cv = CrossValidator(
-            estimator=lorv2,
-            estimatorParamMaps=grid2,
-            parallelism=2,
-            evaluator=BinaryClassificationEvaluator(),
-            foldCol="fold",
-            numFolds=3,
-        )
-        cv.fit(train_dataset)
-
-
-class CrossValidatorTests(CrossValidatorTestsMixin, unittest.TestCase):
-    def setUp(self) -> None:
-        self.spark = SparkSession.builder.master("local[2]").getOrCreate()
-
-    def tearDown(self) -> None:
-        self.spark.stop()
-
-
-if __name__ == "__main__":
-    from pyspark.ml.tests.connect.test_legacy_mode_tuning import *  # noqa: F401,F403
-
-    try:
-        import xmlrunner  # type: ignore[import]
-
-        testRunner = xmlrunner.XMLTestRunner(output="target/test-reports", verbosity=2)
-    except ImportError:
-        testRunner = None
-    unittest.main(testRunner=testRunner, verbosity=2)

snowflake/snowpark_connect/includes/python/pyspark/ml/tests/connect/test_parity_torch_data_loader.py

@@ -1,50 +0,0 @@
-#
-# Licensed to the Apache Software Foundation (ASF) under one or more
-# contributor license agreements.  See the NOTICE file distributed with
-# this work for additional information regarding copyright ownership.
-# The ASF licenses this file to You under the Apache License, Version 2.0
-# (the "License"); you may not use this file except in compliance with
-# the License.  You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-
-import os
-import unittest
-from pyspark.sql import SparkSession
-
-have_torch = "SPARK_SKIP_CONNECT_COMPAT_TESTS" not in os.environ
-try:
-    import torch  # noqa: F401
-except ImportError:
-    have_torch = False
-
-from pyspark.ml.torch.tests.test_data_loader import TorchDistributorDataLoaderUnitTests
-
-
-@unittest.skipIf(not have_torch, "torch is required")
-class TorchDistributorBaselineUnitTestsOnConnect(TorchDistributorDataLoaderUnitTests):
-    def setUp(self) -> None:
-        self.spark = (
-            SparkSession.builder.remote("local[1]")
-            .config("spark.default.parallelism", "1")
-            .getOrCreate()
-        )
-
-
-if __name__ == "__main__":
-    from pyspark.ml.tests.connect.test_parity_torch_data_loader import *  # noqa: F401,F403
-
-    try:
-        import xmlrunner
-
-        testRunner = xmlrunner.XMLTestRunner(output="target/test-reports", verbosity=2)
-    except ImportError:
-        testRunner = None
-    unittest.main(testRunner=testRunner, verbosity=2)

snowflake/snowpark_connect/includes/python/pyspark/ml/tests/connect/test_parity_torch_distributor.py

@@ -1,152 +0,0 @@
-#
-# Licensed to the Apache Software Foundation (ASF) under one or more
-# contributor license agreements.  See the NOTICE file distributed with
-# this work for additional information regarding copyright ownership.
-# The ASF licenses this file to You under the Apache License, Version 2.0
-# (the "License"); you may not use this file except in compliance with
-# the License.  You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-
-import os
-import shutil
-import unittest
-
-have_torch = "SPARK_SKIP_CONNECT_COMPAT_TESTS" not in os.environ
-try:
-    import torch  # noqa: F401
-except ImportError:
-    have_torch = False
-
-from pyspark.sql import SparkSession
-
-from pyspark.ml.torch.tests.test_distributor import (
-    TorchDistributorBaselineUnitTestsMixin,
-    TorchDistributorLocalUnitTestsMixin,
-    TorchDistributorDistributedUnitTestsMixin,
-    TorchWrapperUnitTestsMixin,
-    set_up_test_dirs,
-    get_local_mode_conf,
-    get_distributed_mode_conf,
-)
-
-
-@unittest.skipIf(not have_torch, "torch is required")
-class TorchDistributorBaselineUnitTestsOnConnect(
-    TorchDistributorBaselineUnitTestsMixin, unittest.TestCase
-):
-    @classmethod
-    def setUpClass(cls):
-        cls.spark = SparkSession.builder.remote("local[4]").getOrCreate()
-
-    @classmethod
-    def tearDownClass(cls):
-        cls.spark.stop()
-
-
-@unittest.skipIf(not have_torch, "torch is required")
-class TorchDistributorLocalUnitTestsOnConnect(
-    TorchDistributorLocalUnitTestsMixin, unittest.TestCase
-):
-    @classmethod
-    def setUpClass(cls):
-        (cls.gpu_discovery_script_file_name, cls.mnist_dir_path) = set_up_test_dirs()
-        builder = SparkSession.builder.appName(cls.__name__)
-        for k, v in get_local_mode_conf().items():
-            builder = builder.config(k, v)
-        builder = builder.config(
-            "spark.driver.resource.gpu.discoveryScript", cls.gpu_discovery_script_file_name
-        )
-        cls.spark = builder.remote("local-cluster[2,2,512]").getOrCreate()
-
-    @classmethod
-    def tearDownClass(cls):
-        shutil.rmtree(cls.mnist_dir_path)
-        os.unlink(cls.gpu_discovery_script_file_name)
-        cls.spark.stop()
-
-    def _get_inputs_for_test_local_training_succeeds(self):
-        return [
-            ("0,1,2", 1, True, "0,1,2"),
-            ("0,1,2", 3, True, "0,1,2"),
-            ("0,1,2", 2, False, "0,1,2"),
-            (None, 3, False, "NONE"),
-        ]
-
-
-@unittest.skipIf("SPARK_SKIP_CONNECT_COMPAT_TESTS" in os.environ, "Requires JVM access")
-class TorchDistributorLocalUnitTestsIIOnConnect(
-    TorchDistributorLocalUnitTestsMixin, unittest.TestCase
-):
-    @classmethod
-    def setUpClass(cls):
-        (cls.gpu_discovery_script_file_name, cls.mnist_dir_path) = set_up_test_dirs()
-        builder = SparkSession.builder.appName(cls.__name__)
-        for k, v in get_local_mode_conf().items():
-            builder = builder.config(k, v)
-
-        builder = builder.config(
-            "spark.driver.resource.gpu.discoveryScript", cls.gpu_discovery_script_file_name
-        )
-        cls.spark = builder.remote("local[4]").getOrCreate()
-
-    @classmethod
-    def tearDownClass(cls):
-        shutil.rmtree(cls.mnist_dir_path)
-        os.unlink(cls.gpu_discovery_script_file_name)
-        cls.spark.stop()
-
-    def _get_inputs_for_test_local_training_succeeds(self):
-        return [
-            ("0,1,2", 1, True, "0,1,2"),
-            ("0,1,2", 3, True, "0,1,2"),
-            ("0,1,2", 2, False, "0,1,2"),
-            (None, 3, False, "NONE"),
-        ]
-
-
-@unittest.skipIf(not have_torch, "torch is required")
-class TorchDistributorDistributedUnitTestsOnConnect(
-    TorchDistributorDistributedUnitTestsMixin, unittest.TestCase
-):
-    @classmethod
-    def setUpClass(cls):
-        (cls.gpu_discovery_script_file_name, cls.mnist_dir_path) = set_up_test_dirs()
-        builder = SparkSession.builder.appName(cls.__name__)
-        for k, v in get_distributed_mode_conf().items():
-            builder = builder.config(k, v)
-
-        builder = builder.config(
-            "spark.worker.resource.gpu.discoveryScript", cls.gpu_discovery_script_file_name
-        )
-        cls.spark = builder.remote("local-cluster[2,2,512]").getOrCreate()
-
-    @classmethod
-    def tearDownClass(cls):
-        shutil.rmtree(cls.mnist_dir_path)
-        os.unlink(cls.gpu_discovery_script_file_name)
-        cls.spark.stop()
-
-
-@unittest.skipIf(not have_torch, "torch is required")
-class TorchWrapperUnitTestsOnConnect(TorchWrapperUnitTestsMixin, unittest.TestCase):
-    pass
-
-
-if __name__ == "__main__":
-    from pyspark.ml.tests.connect.test_parity_torch_distributor import *  # noqa: F401,F403
-
-    try:
-        import xmlrunner
-
-        testRunner = xmlrunner.XMLTestRunner(output="target/test-reports", verbosity=2)
-    except ImportError:
-        testRunner = None
-    unittest.main(testRunner=testRunner, verbosity=2)