snowflake-ml-python 1.8.2__py3-none-any.whl → 1.8.4__py3-none-any.whl

snowflake/ml/modeling/linear_model/ridge.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class Ridge(BaseTransformer):
     r"""Linear least squares with l2 regularization
     For more details on this class, see [sklearn.linear_model.Ridge]

snowflake/ml/modeling/linear_model/ridge_classifier.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class RidgeClassifier(BaseTransformer):
     r"""Classifier using Ridge regression
     For more details on this class, see [sklearn.linear_model.RidgeClassifier]

snowflake/ml/modeling/linear_model/ridge_classifier_cv.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class RidgeClassifierCV(BaseTransformer):
     r"""Ridge classifier with built-in cross-validation
     For more details on this class, see [sklearn.linear_model.RidgeClassifierCV]

snowflake/ml/modeling/linear_model/ridge_cv.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class RidgeCV(BaseTransformer):
     r"""Ridge regression with built-in cross-validation
     For more details on this class, see [sklearn.linear_model.RidgeCV]

snowflake/ml/modeling/linear_model/sgd_classifier.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class SGDClassifier(BaseTransformer):
     r"""Linear classifiers (SVM, logistic regression, etc
     For more details on this class, see [sklearn.linear_model.SGDClassifier]

snowflake/ml/modeling/linear_model/sgd_one_class_svm.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class SGDOneClassSVM(BaseTransformer):
     r"""Solves linear One-Class SVM using Stochastic Gradient Descent
     For more details on this class, see [sklearn.linear_model.SGDOneClassSVM]

snowflake/ml/modeling/linear_model/sgd_regressor.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class SGDRegressor(BaseTransformer):
     r"""Linear model fitted by minimizing a regularized empirical loss with SGD
     For more details on this class, see [sklearn.linear_model.SGDRegressor]

snowflake/ml/modeling/linear_model/theil_sen_regressor.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class TheilSenRegressor(BaseTransformer):
     r"""Theil-Sen Estimator: robust multivariate regression model
     For more details on this class, see [sklearn.linear_model.TheilSenRegressor]

snowflake/ml/modeling/linear_model/tweedie_regressor.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class TweedieRegressor(BaseTransformer):
     r"""Generalized Linear Model with a Tweedie distribution
     For more details on this class, see [sklearn.linear_model.TweedieRegressor]

snowflake/ml/modeling/manifold/isomap.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class Isomap(BaseTransformer):
     r"""Isomap Embedding
     For more details on this class, see [sklearn.manifold.Isomap]

snowflake/ml/modeling/manifold/mds.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class MDS(BaseTransformer):
     r"""Multidimensional scaling
     For more details on this class, see [sklearn.manifold.MDS]

snowflake/ml/modeling/manifold/spectral_embedding.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class SpectralEmbedding(BaseTransformer):
     r"""Spectral embedding for non-linear dimensionality reduction
     For more details on this class, see [sklearn.manifold.SpectralEmbedding]

snowflake/ml/modeling/manifold/tsne.py

@@ -11,7 +11,7 @@ import cloudpickle as cp
 import numpy as np
 import pandas as pd
 from numpy import typing as npt
-
+from packaging import version
 
 import numpy
 import sklearn
@@ -60,6 +60,14 @@ DATAFRAME_TYPE = Union[DataFrame, pd.DataFrame]
 
 INFER_SIGNATURE_MAX_ROWS = 100
 
+SKLEARN_LOWER, SKLEARN_UPPER = ('1.4', '1.6')
+# Modeling library estimators require a smaller sklearn version range.
+if not version.Version(SKLEARN_LOWER) <= version.Version(sklearn.__version__) < version.Version(SKLEARN_UPPER):
+    raise Exception(
+        f"To use the modeling library, install scikit-learn version >= {SKLEARN_LOWER} and < {SKLEARN_UPPER}"
+    )
+
+
 class TSNE(BaseTransformer):
     r"""T-distributed Stochastic Neighbor Embedding
     For more details on this class, see [sklearn.manifold.TSNE]

snowflake/ml/modeling/metrics/__init__.py

@@ -5,7 +5,7 @@ import cloudpickle
 from snowflake.ml._internal import init_utils
 from snowflake.ml._internal.utils import result
 
-pkg_dir = os.path.dirname(os.path.abspath(__file__))
+pkg_dir = os.path.dirname(__file__)
 pkg_name = __name__
 exportable_functions = init_utils.fetch_functions_from_modules_in_pkg_dir(pkg_dir=pkg_dir, pkg_name=pkg_name)
 for k, v in exportable_functions.items():

snowflake/ml/modeling/metrics/classification.py

@@ -2,7 +2,7 @@ import inspect
 import json
 import math
 import warnings
-from typing import Any, Dict, Iterable, List, Optional, Set, Tuple, Union
+from typing import Any, Iterable, Optional, Union
 
 import cloudpickle
 import numpy as np
@@ -32,8 +32,8 @@ _SUBPROJECT = "Metrics"
 def accuracy_score(
     *,
     df: snowpark.DataFrame,
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
     normalize: bool = True,
     sample_weight_col_name: Optional[str] = None,
 ) -> float:
@@ -221,7 +221,7 @@ def confusion_matrix(
     return cm
 
 
-def _register_confusion_matrix_computer(*, session: snowpark.Session, statement_params: Dict[str, Any]) -> str:
+def _register_confusion_matrix_computer(*, session: snowpark.Session, statement_params: dict[str, Any]) -> str:
     """Registers confusion matrix computation UDTF in Snowflake and returns the name of the UDTF.
 
     Args:
@@ -247,7 +247,7 @@ def _register_confusion_matrix_computer(*, session: snowpark.Session, statement_
             # Number of labels.
             self._n_label = 0
 
-        def process(self, input_row: List[float], n_label: int) -> None:
+        def process(self, input_row: list[float], n_label: int) -> None:
             """Computes confusion matrix.
 
             Args:
@@ -270,7 +270,7 @@ def _register_confusion_matrix_computer(*, session: snowpark.Session, statement_
                 self.update_confusion_matrix()
                 self._cur_count = 0
 
-        def end_partition(self) -> Iterable[Tuple[bytes, str]]:
+        def end_partition(self) -> Iterable[tuple[bytes, str]]:
             # 3. Compute sum and dot_prod for the remaining rows in the batch.
             if self._cur_count > 0:
                 self.update_confusion_matrix()
@@ -313,8 +313,8 @@ def _register_confusion_matrix_computer(*, session: snowpark.Session, statement_
 def f1_score(
     *,
     df: snowpark.DataFrame,
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
     labels: Optional[npt.ArrayLike] = None,
     pos_label: Union[str, int] = 1,
     average: Optional[str] = "binary",
@@ -406,8 +406,8 @@ def f1_score(
 def fbeta_score(
     *,
     df: snowpark.DataFrame,
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
     beta: float,
     labels: Optional[npt.ArrayLike] = None,
     pos_label: Union[str, int] = 1,
@@ -501,8 +501,8 @@ def fbeta_score(
 def log_loss(
     *,
     df: snowpark.DataFrame,
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
     eps: Union[float, str] = "auto",
     normalize: bool = True,
     sample_weight_col_name: Optional[str] = None,
@@ -625,7 +625,7 @@ def log_loss(
 def _register_log_loss_computer(
     *,
     session: snowpark.Session,
-    statement_params: Dict[str, Any],
+    statement_params: dict[str, Any],
     labels: Optional[npt.ArrayLike] = None,
 ) -> str:
     """Registers log loss computation UDTF in Snowflake and returns the name of the UDTF.
@@ -644,16 +644,16 @@ def _register_log_loss_computer(
     class LogLossComputer:
         def __init__(self) -> None:
             self._labels = labels
-            self._y_true: List[List[int]] = []
-            self._y_pred: List[List[float]] = []
-            self._sample_weight: List[float] = []
+            self._y_true: list[list[int]] = []
+            self._y_pred: list[list[float]] = []
+            self._sample_weight: list[float] = []
 
-        def process(self, y_true: List[int], y_pred: List[float], sample_weight: float) -> None:
+        def process(self, y_true: list[int], y_pred: list[float], sample_weight: float) -> None:
             self._y_true.append(y_true)
             self._y_pred.append(y_pred)
             self._sample_weight.append(sample_weight)
 
-        def end_partition(self) -> Iterable[Tuple[float]]:
+        def end_partition(self) -> Iterable[tuple[float]]:
             res = metrics.log_loss(
                 self._y_true,
                 self._y_pred,
@@ -685,18 +685,18 @@ def _register_log_loss_computer(
 def precision_recall_fscore_support(
     *,
     df: snowpark.DataFrame,
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
     beta: float = 1.0,
     labels: Optional[npt.ArrayLike] = None,
     pos_label: Union[str, int] = 1,
     average: Optional[str] = None,
-    warn_for: Union[Tuple[str, ...], Set[str]] = ("precision", "recall", "f-score"),
+    warn_for: Union[tuple[str, ...], set[str]] = ("precision", "recall", "f-score"),
     sample_weight_col_name: Optional[str] = None,
     zero_division: Union[str, int] = "warn",
 ) -> Union[
-    Tuple[float, float, float, None],
-    Tuple[npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_]],
+    tuple[float, float, float, None],
+    tuple[npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_]],
 ]:
     """
     Compute precision, recall, F-measure and support for each class.
@@ -854,8 +854,8 @@ def precision_recall_fscore_support(
         result_object = result.deserialize(session, precision_recall_fscore_support_anon_sproc(session, **kwargs))
 
         res: Union[
-            Tuple[float, float, float, None],
-            Tuple[npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_]],
+            tuple[float, float, float, None],
+            tuple[npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_]],
         ] = result_object[:4]
         warning = result_object[-1]
         if warning:
@@ -1039,18 +1039,18 @@ def _register_multilabel_confusion_matrix_computer(
         def __init__(self) -> None:
             self._labels = labels
             self._samplewise = samplewise
-            self._y_true: List[List[int]] = []
-            self._y_pred: List[List[int]] = []
-            self._sample_weight: List[float] = []
+            self._y_true: list[list[int]] = []
+            self._y_pred: list[list[int]] = []
+            self._sample_weight: list[float] = []
 
-        def process(self, y_true: List[int], y_pred: List[int], sample_weight: float) -> None:
+        def process(self, y_true: list[int], y_pred: list[int], sample_weight: float) -> None:
             self._y_true.append(y_true)
             self._y_pred.append(y_pred)
             self._sample_weight.append(sample_weight)
 
         def end_partition(
             self,
-        ) -> Iterable[Tuple[npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_]]]:
+        ) -> Iterable[tuple[npt.NDArray[np.float_], npt.NDArray[np.float_], npt.NDArray[np.float_]]]:
             MCM = metrics.multilabel_confusion_matrix(
                 self._y_true,
                 self._y_pred,
@@ -1093,8 +1093,8 @@ def _register_multilabel_confusion_matrix_computer(
 def _binary_precision_score(
     *,
     df: snowpark.DataFrame,
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
     pos_label: Union[str, int] = 1,
     sample_weight_col_name: Optional[str] = None,
     zero_division: Union[str, int] = "warn",
@@ -1166,8 +1166,8 @@ def _binary_precision_score(
 def precision_score(
     *,
     df: snowpark.DataFrame,
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
     labels: Optional[npt.ArrayLike] = None,
     pos_label: Union[str, int] = 1,
     average: Optional[str] = "binary",
@@ -1264,8 +1264,8 @@ def precision_score(
 def recall_score(
     *,
     df: snowpark.DataFrame,
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
     labels: Optional[npt.ArrayLike] = None,
     pos_label: Union[str, int] = 1,
     average: Optional[str] = "binary",
@@ -1376,9 +1376,9 @@ def _sum_array_col(df: snowpark.DataFrame, col_name: str) -> snowpark.DataFrame:
 
 
 def _check_binary_labels(
-    labels: List[Any],
+    labels: list[Any],
     pos_label: Union[str, int] = 1,
-) -> List[Any]:
+) -> list[Any]:
     """Validation associated with binary average labels.
 
     Args:
@@ -1411,7 +1411,7 @@ def _prf_divide(
     metric: str,
     modifier: str,
     average: Optional[str] = None,
-    warn_for: Union[Tuple[str, ...], Set[str]] = ("precision", "recall", "f-score"),
+    warn_for: Union[tuple[str, ...], set[str]] = ("precision", "recall", "f-score"),
     zero_division: Union[str, int] = "warn",
 ) -> npt.NDArray[np.float_]:
     """Performs division and handles divide-by-zero.

snowflake/ml/modeling/metrics/metrics_utils.py

@@ -1,6 +1,6 @@
 import math
 import warnings
-from typing import Any, Collection, Dict, Iterable, List, Optional, Tuple, Union
+from typing import Any, Collection, Iterable, Optional, Union
 
 import cloudpickle
 import numpy as np
@@ -18,7 +18,7 @@ INDEX = "INDEX"
 BATCH_SIZE = 1000
 
 
-def register_accumulator_udtf(*, session: Session, statement_params: Dict[str, Any]) -> str:
+def register_accumulator_udtf(*, session: Session, statement_params: dict[str, Any]) -> str:
     """Registers accumulator UDTF in Snowflake and returns the name of the UDTF.
 
     Args:
@@ -47,7 +47,7 @@ def register_accumulator_udtf(*, session: Session, statement_params: Dict[str, A
             else:
                 self._accumulated_row = self._accumulated_row + row
 
-        def end_partition(self) -> Iterable[Tuple[bytes]]:
+        def end_partition(self) -> Iterable[tuple[bytes]]:
             yield (cloudpickle.dumps(self._accumulated_row),)
 
     accumulator = snowpark_utils.random_name_for_temp_object(snowpark_utils.TempObjectType.TABLE_FUNCTION)
@@ -68,7 +68,7 @@ def register_accumulator_udtf(*, session: Session, statement_params: Dict[str, A
     return accumulator
 
 
-def register_sharded_dot_sum_computer(*, session: Session, statement_params: Dict[str, Any]) -> str:
+def register_sharded_dot_sum_computer(*, session: Session, statement_params: dict[str, Any]) -> str:
     """Registers dot and sum computation UDTF in Snowflake and returns the name of the UDTF.
 
     Args:
@@ -110,7 +110,7 @@ def register_sharded_dot_sum_computer(*, session: Session, statement_params: Dic
             # Square root of count - ddof
             self._sqrt_count_d = -1.0
 
-        def process(self, input_row: List[float], count: int, ddof: int) -> None:
+        def process(self, input_row: list[float], count: int, ddof: int) -> None:
             """Computes sum and dot product.
 
             Args:
@@ -138,7 +138,7 @@ def register_sharded_dot_sum_computer(*, session: Session, statement_params: Dic
                 self.accumulate_batch_sum_and_dot_prod()
                 self._cur_count = 0
 
-        def end_partition(self) -> Iterable[Tuple[bytes, str]]:
+        def end_partition(self) -> Iterable[tuple[bytes, str]]:
             # 3. Compute sum and dot_prod for the remaining rows in the batch.
             if self._cur_count > 0:
                 self.accumulate_batch_sum_and_dot_prod()
@@ -185,7 +185,7 @@ def register_sharded_dot_sum_computer(*, session: Session, statement_params: Dic
 
 def validate_and_return_dataframe_and_columns(
     *, df: snowpark.DataFrame, columns: Optional[Collection[str]] = None
-) -> Tuple[snowpark.DataFrame, Collection[str]]:
+) -> tuple[snowpark.DataFrame, Collection[str]]:
     """Validates that the columns are all numeric and returns a dataframe with those columns.
 
     Args:
@@ -212,8 +212,8 @@ def validate_and_return_dataframe_and_columns(
 
 
 def check_label_columns(
-    y_true_col_names: Union[str, List[str]],
-    y_pred_col_names: Union[str, List[str]],
+    y_true_col_names: Union[str, list[str]],
+    y_pred_col_names: Union[str, list[str]],
 ) -> None:
     """Check y true and y pred columns.
 
@@ -238,7 +238,7 @@ def check_label_columns(
         )
 
 
-def flatten_cols(cols: List[Optional[Union[str, List[str]]]]) -> List[str]:
+def flatten_cols(cols: list[Optional[Union[str, list[str]]]]) -> list[str]:
     res = []
     for col in cols:
         if isinstance(col, str):
@@ -251,7 +251,7 @@ def flatten_cols(cols: List[Optional[Union[str, List[str]]]]) -> List[str]:
 def unique_labels(
     *,
     df: snowpark.DataFrame,
-    columns: List[snowpark.Column],
+    columns: list[snowpark.Column],
 ) -> snowpark.DataFrame:
     """Extract indexed ordered unique labels as a dataframe.
 
@@ -311,7 +311,7 @@ def weighted_sum(
     sample_score_column: snowpark.Column,
     sample_weight_column: Optional[snowpark.Column] = None,
     normalize: bool = False,
-    statement_params: Dict[str, str],
+    statement_params: dict[str, str],
 ) -> float:
     """Weighted sum of the sample score column.