scikit-learn-intelex 2024.5.0__py39-none-win_amd64.whl → 2024.7.0__py39-none-win_amd64.whl

scikit_learn_intelex-2024.5.0.data/data/Lib/site-packages/sklearnex/tests/test_memory_usage.py

@@ -1,231 +0,0 @@
-# ==============================================================================
-# Copyright 2021 Intel Corporation
-#
-# Licensed 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 gc
-import logging
-import tracemalloc
-import types
-
-import numpy as np
-import pandas as pd
-import pytest
-from scipy.stats import pearsonr
-from sklearn.base import BaseEstimator
-from sklearn.datasets import make_classification
-from sklearn.model_selection import KFold
-
-from sklearnex import get_patch_map
-from sklearnex.metrics import pairwise_distances, roc_auc_score
-from sklearnex.model_selection import train_test_split
-from sklearnex.utils import _assert_all_finite
-
-
-class TrainTestSplitEstimator:
-    def __init__(self):
-        pass
-
-    def fit(self, x, y):
-        train_test_split(x, y)
-
-
-class FiniteCheckEstimator:
-    def __init__(self):
-        pass
-
-    def fit(self, x, y):
-        _assert_all_finite(x)
-        _assert_all_finite(y)
-
-
-class PairwiseDistancesEstimator:
-    def fit(self, x, y):
-        pairwise_distances(x, metric=self.metric)
-
-
-class CosineDistancesEstimator(PairwiseDistancesEstimator):
-    def __init__(self):
-        self.metric = "cosine"
-
-
-class CorrelationDistancesEstimator(PairwiseDistancesEstimator):
-    def __init__(self):
-        self.metric = "correlation"
-
-
-class RocAucEstimator:
-    def __init__(self):
-        pass
-
-    def fit(self, x, y):
-        print(roc_auc_score(y, np.zeros(shape=y.shape, dtype=np.int32)))
-
-
-# add all daal4py estimators enabled in patching (except banned)
-
-
-def get_patched_estimators(ban_list, output_list):
-    patched_estimators = get_patch_map().values()
-    for listing in patched_estimators:
-        estimator, name = listing[0][0][2], listing[0][0][1]
-        if not isinstance(estimator, types.FunctionType):
-            if name not in ban_list:
-                if issubclass(estimator, BaseEstimator):
-                    if hasattr(estimator, "fit"):
-                        output_list.append(estimator)
-
-
-def remove_duplicated_estimators(estimators_list):
-    estimators_map = {}
-    for estimator in estimators_list:
-        full_name = f"{estimator.__module__}.{estimator.__name__}"
-        estimators_map[full_name] = estimator
-    return estimators_map.values()
-
-
-BANNED_ESTIMATORS = (
-    "IncrementalEmpiricalCovariance",  # dataframe_f issues
-    "IncrementalLinearRegression",  # TODO fix memory leak issue in private CI for data_shape = (1000, 100), data_transform_function = dataframe_f
-    "TSNE",  # too slow for using in testing on common data size
-)
-estimators = [
-    TrainTestSplitEstimator,
-    FiniteCheckEstimator,
-    CosineDistancesEstimator,
-    CorrelationDistancesEstimator,
-    RocAucEstimator,
-]
-get_patched_estimators(BANNED_ESTIMATORS, estimators)
-estimators = remove_duplicated_estimators(estimators)
-
-
-def ndarray_c(x, y):
-    return np.ascontiguousarray(x), y
-
-
-def ndarray_f(x, y):
-    return np.asfortranarray(x), y
-
-
-def dataframe_c(x, y):
-    return pd.DataFrame(np.ascontiguousarray(x)), pd.Series(y)
-
-
-def dataframe_f(x, y):
-    return pd.DataFrame(np.asfortranarray(x)), pd.Series(y)
-
-
-data_transforms = [ndarray_c, ndarray_f, dataframe_c, dataframe_f]
-
-data_shapes = [(1000, 100), (2000, 50)]
-
-EXTRA_MEMORY_THRESHOLD = 0.15
-N_SPLITS = 10
-
-
-def gen_clsf_data(n_samples, n_features):
-    data, label = make_classification(
-        n_classes=2, n_samples=n_samples, n_features=n_features, random_state=777
-    )
-    return (
-        data,
-        label,
-        data.size * data.dtype.itemsize + label.size * label.dtype.itemsize,
-    )
-
-
-def split_train_inference(kf, x, y, estimator):
-    mem_tracks = []
-    for train_index, test_index in kf.split(x):
-        if isinstance(x, np.ndarray):
-            x_train, x_test = x[train_index], x[test_index]
-            y_train, y_test = y[train_index], y[test_index]
-        elif isinstance(x, pd.core.frame.DataFrame):
-            x_train, x_test = x.iloc[train_index], x.iloc[test_index]
-            y_train, y_test = y.iloc[train_index], y.iloc[test_index]
-        # TODO: add parameters for all estimators to prevent
-        # fallback to stock scikit-learn with default parameters
-
-        alg = estimator()
-        alg.fit(x_train, y_train)
-        if hasattr(alg, "predict"):
-            alg.predict(x_test)
-        elif hasattr(alg, "transform"):
-            alg.transform(x_test)
-        elif hasattr(alg, "kneighbors"):
-            alg.kneighbors(x_test)
-        del alg, x_train, x_test, y_train, y_test
-        mem_tracks.append(tracemalloc.get_traced_memory()[0])
-    return mem_tracks
-
-
-def _kfold_function_template(estimator, data_transform_function, data_shape):
-    tracemalloc.start()
-
-    n_samples, n_features = data_shape
-    x, y, data_memory_size = gen_clsf_data(n_samples, n_features)
-    kf = KFold(n_splits=N_SPLITS)
-    x, y = data_transform_function(x, y)
-
-    mem_before, _ = tracemalloc.get_traced_memory()
-    mem_tracks = split_train_inference(kf, x, y, estimator)
-    mem_iter_diffs = np.array(mem_tracks[1:]) - np.array(mem_tracks[:-1])
-    mem_incr_mean, mem_incr_std = mem_iter_diffs.mean(), mem_iter_diffs.std()
-    mem_incr_mean, mem_incr_std = round(mem_incr_mean), round(mem_incr_std)
-    mem_iter_corr, _ = pearsonr(mem_tracks, list(range(len(mem_tracks))))
-    if mem_iter_corr > 0.95:
-        logging.warning(
-            "Memory usage is steadily increasing with iterations "
-            "(Pearson correlation coefficient between "
-            f"memory tracks and iterations is {mem_iter_corr})\n"
-            "Memory usage increase per iteration: "
-            f"{mem_incr_mean}±{mem_incr_std} bytes"
-        )
-    mem_before_gc, _ = tracemalloc.get_traced_memory()
-    mem_diff = mem_before_gc - mem_before
-    message = (
-        "Size of extra allocated memory {} using garbage collector "
-        f"is greater than {EXTRA_MEMORY_THRESHOLD * 100}% of input data"
-        f"\n\tAlgorithm: {estimator.__name__}"
-        f"\n\tInput data size: {data_memory_size} bytes"
-        "\n\tExtra allocated memory size: {} bytes"
-        " / {} %"
-    )
-    if mem_diff >= EXTRA_MEMORY_THRESHOLD * data_memory_size:
-        logging.warning(
-            message.format(
-                "before", mem_diff, round((mem_diff) / data_memory_size * 100, 2)
-            )
-        )
-    gc.collect()
-    mem_after, _ = tracemalloc.get_traced_memory()
-    tracemalloc.stop()
-    mem_diff = mem_after - mem_before
-
-    assert mem_diff < EXTRA_MEMORY_THRESHOLD * data_memory_size, message.format(
-        "after", mem_diff, round((mem_diff) / data_memory_size * 100, 2)
-    )
-
-
-# disable fallback check as logging impacts memory use
-
-
-@pytest.mark.allow_sklearn_fallback
-@pytest.mark.parametrize("data_transform_function", data_transforms)
-@pytest.mark.parametrize("estimator", estimators)
-@pytest.mark.parametrize("data_shape", data_shapes)
-def test_memory_leaks(estimator, data_transform_function, data_shape):
-    _kfold_function_template(estimator, data_transform_function, data_shape)

scikit_learn_intelex-2024.5.0.data/data/Lib/site-packages/sklearnex/tests/test_run_to_run_stability_tests.py

@@ -1,428 +0,0 @@
-# ===============================================================================
-# Copyright 2020 Intel Corporation
-#
-# Licensed 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 random
-
-import numpy as np
-import pytest
-
-import daal4py as d4p
-from sklearnex import patch_sklearn
-
-patch_sklearn()
-
-from scipy import sparse
-from sklearn.cluster import DBSCAN, KMeans
-from sklearn.datasets import (
-    load_breast_cancer,
-    load_diabetes,
-    load_iris,
-    make_classification,
-    make_regression,
-)
-from sklearn.decomposition import PCA
-from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor
-from sklearn.linear_model import (
-    ElasticNet,
-    Lasso,
-    LinearRegression,
-    LogisticRegression,
-    LogisticRegressionCV,
-    Ridge,
-)
-from sklearn.manifold import TSNE
-from sklearn.metrics import pairwise_distances, roc_auc_score
-from sklearn.model_selection import train_test_split
-from sklearn.neighbors import (
-    KNeighborsClassifier,
-    KNeighborsRegressor,
-    LocalOutlierFactor,
-    NearestNeighbors,
-)
-from sklearn.svm import SVC, SVR, NuSVC, NuSVR
-
-from daal4py.sklearn._utils import daal_check_version
-
-# to reproduce errors even in CI
-d4p.daalinit(nthreads=100)
-
-
-def get_class_name(x):
-    return x.__class__.__name__
-
-
-def method_processing(X, clf, methods):
-    res = []
-    name = []
-    for i in methods:
-        if i == "predict":
-            res.append(clf.predict(X))
-            name.append(get_class_name(clf) + ".predict(X)")
-        elif i == "predict_proba":
-            res.append(clf.predict_proba(X))
-            name.append(get_class_name(clf) + ".predict_proba(X)")
-        elif i == "decision_function":
-            res.append(clf.decision_function(X))
-            name.append(get_class_name(clf) + ".decision_function(X)")
-        elif i == "kneighbors":
-            dist, idx = clf.kneighbors(X)
-            res.append(dist)
-            name.append("dist")
-            res.append(idx)
-            name.append("idx")
-        elif i == "fit_predict":
-            predict = clf.fit_predict(X)
-            res.append(predict)
-            name.append(get_class_name(clf) + ".fit_predict")
-        elif i == "fit_transform":
-            res.append(clf.fit_transform(X))
-            name.append(get_class_name(clf) + ".fit_transform")
-        elif i == "transform":
-            res.append(clf.transform(X))
-            name.append(get_class_name(clf) + ".transform(X)")
-        elif i == "get_covariance":
-            res.append(clf.get_covariance())
-            name.append(get_class_name(clf) + ".get_covariance()")
-        elif i == "get_precision":
-            res.append(clf.get_precision())
-            name.append(get_class_name(clf) + ".get_precision()")
-        elif i == "score_samples":
-            res.append(clf.score_samples(X))
-            name.append(get_class_name(clf) + ".score_samples(X)")
-    return res, name
-
-
-def func(X, Y, clf, methods):
-    clf.fit(X, Y)
-    res, name = method_processing(X, clf, methods)
-
-    for i in clf.__dict__.keys():
-        ans = getattr(clf, i)
-        if isinstance(ans, (bool, float, int, np.ndarray, np.float64)):
-            if isinstance(ans, np.ndarray) and None in ans:
-                continue
-            res.append(ans)
-            name.append(get_class_name(clf) + "." + i)
-    return res, name
-
-
-def _run_test(model, methods, dataset):
-    datasets = []
-    if dataset in ["blobs", "classifier", "sparse"]:
-        X1, y1 = load_iris(return_X_y=True)
-        if dataset == "sparse":
-            X1 = sparse.csr_matrix(X1)
-        datasets.append((X1, y1))
-        X2, y2 = load_breast_cancer(return_X_y=True)
-        if dataset == "sparse":
-            X2 = sparse.csr_matrix(X2)
-        datasets.append((X2, y2))
-    elif dataset == "regression":
-        X1, y1 = make_regression(
-            n_samples=500, n_features=10, noise=64.0, random_state=42
-        )
-        datasets.append((X1, y1))
-        X2, y2 = load_diabetes(return_X_y=True)
-        datasets.append((X2, y2))
-    else:
-        raise ValueError("Unknown dataset type")
-
-    for X, y in datasets:
-        baseline, name = func(X, y, model, methods)
-        for i in range(10):
-            res, _ = func(X, y, model, methods)
-
-            for a, b, n in zip(res, baseline, name):
-                np.testing.assert_allclose(
-                    a, b, rtol=0.0, atol=0.0, err_msg=str(n + " is incorrect")
-                )
-
-
-MODELS_INFO = [
-    {
-        "model": KNeighborsClassifier(
-            n_neighbors=10, algorithm="brute", weights="uniform"
-        ),
-        "methods": ["predict", "predict_proba", "kneighbors"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KNeighborsClassifier(
-            n_neighbors=10, algorithm="brute", weights="distance"
-        ),
-        "methods": ["predict", "predict_proba", "kneighbors"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KNeighborsClassifier(
-            n_neighbors=10, algorithm="kd_tree", weights="uniform"
-        ),
-        "methods": ["predict", "predict_proba", "kneighbors"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KNeighborsClassifier(
-            n_neighbors=10, algorithm="kd_tree", weights="distance"
-        ),
-        "methods": ["predict", "predict_proba", "kneighbors"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KNeighborsRegressor(
-            n_neighbors=10, algorithm="kd_tree", weights="distance"
-        ),
-        "methods": ["predict", "kneighbors"],
-        "dataset": "regression",
-    },
-    {
-        "model": KNeighborsRegressor(
-            n_neighbors=10, algorithm="kd_tree", weights="uniform"
-        ),
-        "methods": ["predict", "kneighbors"],
-        "dataset": "regression",
-    },
-    {
-        "model": KNeighborsRegressor(
-            n_neighbors=10, algorithm="brute", weights="distance"
-        ),
-        "methods": ["predict", "kneighbors"],
-        "dataset": "regression",
-    },
-    {
-        "model": KNeighborsRegressor(
-            n_neighbors=10, algorithm="brute", weights="uniform"
-        ),
-        "methods": ["predict", "kneighbors"],
-        "dataset": "regression",
-    },
-    {
-        "model": NearestNeighbors(n_neighbors=10, algorithm="brute"),
-        "methods": ["kneighbors"],
-        "dataset": "blobs",
-    },
-    {
-        "model": NearestNeighbors(n_neighbors=10, algorithm="kd_tree"),
-        "methods": ["kneighbors"],
-        "dataset": "blobs",
-    },
-    {
-        "model": LocalOutlierFactor(n_neighbors=10, novelty=False),
-        "methods": ["fit_predict"],
-        "dataset": "blobs",
-    },
-    {
-        "model": LocalOutlierFactor(n_neighbors=10, novelty=True),
-        "methods": ["predict"],
-        "dataset": "blobs",
-    },
-    {
-        "model": DBSCAN(algorithm="brute", n_jobs=-1),
-        "methods": [],
-        "dataset": "blobs",
-    },
-    {
-        "model": SVC(kernel="rbf"),
-        "methods": ["predict", "decision_function"],
-        "dataset": "classifier",
-    },
-    {
-        "model": SVC(kernel="rbf"),
-        "methods": ["predict", "decision_function"],
-        "dataset": "sparse",
-    },
-    {
-        "model": NuSVC(kernel="rbf"),
-        "methods": ["predict", "decision_function"],
-        "dataset": "classifier",
-    },
-    {
-        "model": SVR(kernel="rbf"),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": NuSVR(kernel="rbf"),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": TSNE(random_state=0),
-        "methods": ["fit_transform"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KMeans(random_state=0, init="k-means++"),
-        "methods": ["predict"],
-        "dataset": "blobs",
-    },
-    {
-        "model": KMeans(random_state=0, init="random"),
-        "methods": ["predict"],
-        "dataset": "blobs",
-    },
-    {
-        "model": KMeans(random_state=0, init="k-means++"),
-        "methods": ["predict"],
-        "dataset": "sparse",
-    },
-    {
-        "model": KMeans(random_state=0, init="random"),
-        "methods": ["predict"],
-        "dataset": "sparse",
-    },
-    {
-        "model": ElasticNet(random_state=0),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": Lasso(random_state=0),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": PCA(n_components=0.5, svd_solver="covariance_eigh", random_state=0),
-        "methods": ["transform", "get_covariance", "get_precision", "score_samples"],
-        "dataset": "classifier",
-    },
-    {
-        "model": RandomForestClassifier(
-            random_state=0, oob_score=True, max_samples=0.5, max_features="sqrt"
-        ),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": LogisticRegression(random_state=0, solver="newton-cg", max_iter=1000),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": LogisticRegression(random_state=0, solver="lbfgs", max_iter=1000),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": LogisticRegressionCV(
-            random_state=0, solver="newton-cg", n_jobs=-1, max_iter=1000
-        ),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": LogisticRegressionCV(
-            random_state=0, solver="lbfgs", n_jobs=-1, max_iter=1000
-        ),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": RandomForestRegressor(
-            random_state=0, oob_score=True, max_samples=0.5, max_features="sqrt"
-        ),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": LinearRegression(),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": Ridge(random_state=0),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-]
-
-TO_SKIP = [
-    "TSNE",  # Absolute diff is 1e-10, potential problem in KNN,
-    # will be fixed for next release. (UPD. KNN is fixed but there is a problem
-    # with stability of stock sklearn. It is already stable in master, so, we
-    # need to wait for the next sklearn release)
-    "LogisticRegression",  # Absolute diff is 1e-8, will be fixed for next release
-    "LogisticRegressionCV",  # Absolute diff is 1e-10, will be fixed for next release
-    "RandomForestRegressor",  # Absolute diff is 1e-14 in OOB score,
-    # will be fixed for next release
-]
-
-
-@pytest.mark.parametrize("model_head", MODELS_INFO)
-def test_models(model_head):
-    stable_algos = []
-    if get_class_name(model_head["model"]) in stable_algos and daal_check_version(
-        (2021, "P", 300)
-    ):
-        try:
-            TO_SKIP.remove(get_class_name(model_head["model"]))
-        except ValueError:
-            pass
-    if get_class_name(model_head["model"]) in TO_SKIP:
-        pytest.skip("Unstable", allow_module_level=False)
-    _run_test(model_head["model"], model_head["methods"], model_head["dataset"])
-
-
-@pytest.mark.parametrize("features", range(5, 10))
-def test_train_test_split(features):
-    X, y = make_classification(
-        n_samples=4000,
-        n_features=features,
-        n_informative=features,
-        n_redundant=0,
-        n_clusters_per_class=8,
-        random_state=0,
-    )
-    (
-        baseline_X_train,
-        baseline_X_test,
-        baseline_y_train,
-        baseline_y_test,
-    ) = train_test_split(X, y, test_size=0.33, random_state=0)
-    baseline = [baseline_X_train, baseline_X_test, baseline_y_train, baseline_y_test]
-    for _ in range(10):
-        X_train, X_test, y_train, y_test = train_test_split(
-            X, y, test_size=0.33, random_state=0
-        )
-        res = [X_train, X_test, y_train, y_test]
-        for a, b in zip(res, baseline):
-            np.testing.assert_allclose(
-                a, b, rtol=0.0, atol=0.0, err_msg=str("train_test_split is incorrect")
-            )
-
-
-@pytest.mark.parametrize("metric", ["cosine", "correlation"])
-def test_pairwise_distances(metric):
-    X = np.random.rand(1000)
-    X = np.array(X, dtype=np.float64)
-    baseline = pairwise_distances(X.reshape(1, -1), metric=metric)
-    for _ in range(5):
-        res = pairwise_distances(X.reshape(1, -1), metric=metric)
-        for a, b in zip(res, baseline):
-            np.testing.assert_allclose(
-                a, b, rtol=0.0, atol=0.0, err_msg=str("pairwise_distances is incorrect")
-            )
-
-
-@pytest.mark.parametrize("array_size", [100, 1000, 10000])
-def test_roc_auc(array_size):
-    a = [random.randint(0, 1) for i in range(array_size)]
-    b = [random.randint(0, 1) for i in range(array_size)]
-    baseline = roc_auc_score(a, b)
-    for _ in range(5):
-        res = roc_auc_score(a, b)
-        np.testing.assert_allclose(
-            baseline, res, rtol=0.0, atol=0.0, err_msg=str("roc_auc is incorrect")
-        )