snowflake-ml-python 1.1.2__py3-none-any.whl → 1.2.1__py3-none-any.whl

snowflake/ml/modeling/xgboost/xgb_classifier.py

@@ -53,6 +53,18 @@ _PROJECT = "ModelDevelopment"
 _SUBPROJECT = "".join([s.capitalize() for s in "xgboost".replace("sklearn.", "").split("_")])
 
 
+def _is_fit_predict_method_enabled() -> Callable[[Any], bool]:
+    def check(self: BaseTransformer) -> TypeGuard[Callable[..., object]]:
+        return False and callable(getattr(self._sklearn_object, "fit_predict", None))
+    return check
+
+
+def _is_fit_transform_method_enabled() -> Callable[[Any], bool]:
+    def check(self: BaseTransformer) -> TypeGuard[Callable[..., object]]:
+        return False and callable(getattr(self._sklearn_object, "fit_transform", None))
+    return check
+
+
 class XGBClassifier(BaseTransformer):
     r"""Implementation of the scikit-learn API for XGBoost classification
     For more details on this class, see [xgboost.XGBClassifier]
@@ -105,6 +117,22 @@ class XGBClassifier(BaseTransformer):
     drop_input_cols: Optional[bool], default=False
         If set, the response of predict(), transform() methods will not contain input columns.
 
+    use_external_memory_version: bool, default=False
+        If set, external memory version of XGBoost trainer is used. External memory training
+        is done in a two-step process. First,in the preprocessing step, input data is read and
+        parsed into an internal format, which can be CSR, CSC, or sorted CSC, and stored in
+        in-memory buffers. The in-memory buffers are continuously flushed out to disk when
+        predefined memory limit is reached. Second, in the tree construction step, the data
+        pages are streamed from disk via a multi-threaded pre-fetcher as needed for tree construction.
+        Note:'tree_method's 'approx', and 'hist' are supported in the external memory version.
+        Note:'grow_policy=depthwise' is used for optimal performance in the external memory version.
+
+    batch_size: int, default=10000
+        Number of rows in each batch of input data while using external memory training.
+        It is not recommended to set small batch sizes, like 32 samples per batch, as this
+        can seriously hurt performance in gradient boosting. Set the batch_size as large as possible
+        based on the available memory.
+
     n_estimators: int
             Number of boosting rounds.
 
@@ -325,6 +353,8 @@ class XGBClassifier(BaseTransformer):
         passthrough_cols: Optional[Union[str, Iterable[str]]] = None,
         drop_input_cols: Optional[bool] = False,
         sample_weight_col: Optional[str] = None,
+        use_external_memory_version: bool = False,
+        batch_size: int = 10000,
         **kwargs,
     ) -> None:
         super().__init__()
@@ -334,7 +364,9 @@ class XGBClassifier(BaseTransformer):
         self.set_label_cols(label_cols)
         self.set_passthrough_cols(passthrough_cols)
         self.set_drop_input_cols(drop_input_cols)
-        self.set_sample_weight_col(sample_weight_col)        
+        self.set_sample_weight_col(sample_weight_col)
+        self._use_external_memory_version = use_external_memory_version
+        self._batch_size = batch_size        
         deps: Set[str] = set([f'numpy=={np.__version__}', f'xgboost=={xgboost.__version__}', f'cloudpickle=={cp.__version__}'])
         
         self._deps = list(deps)
@@ -412,11 +444,6 @@ class XGBClassifier(BaseTransformer):
         if isinstance(dataset, DataFrame):
             session = dataset._session
             assert session is not None  # keep mypy happy
-            # Validate that key package version in user workspace are supported in snowflake conda channel
-            # If customer doesn't have package in conda channel, replace the ones have the closest versions
-            self._deps = pkg_version_utils.get_valid_pkg_versions_supported_in_snowflake_conda_channel(
-                pkg_versions=self._get_dependencies(), session=session, subproject=_SUBPROJECT)
-            
             # Specify input columns so column pruning will be enforced
             selected_cols = self._get_active_columns()
             if len(selected_cols) > 0:
@@ -444,7 +471,9 @@ class XGBClassifier(BaseTransformer):
             label_cols=self.label_cols,
             sample_weight_col=self.sample_weight_col,
             autogenerated=self._autogenerated,
-            subproject=_SUBPROJECT
+            subproject=_SUBPROJECT,
+            use_external_memory_version=self._use_external_memory_version,
+            batch_size=self._batch_size,
         )
         self._sklearn_object = model_trainer.train()
         self._is_fitted = True
@@ -715,6 +744,22 @@ class XGBClassifier(BaseTransformer):
                 # each row containing a list of values.
                 expected_dtype = "ARRAY"
 
+            # If we were unable to assign a type to this transform in the factory, infer the type here.
+            if expected_dtype == "":
+                # If this is a clustering transformer, if the number of output columns does not equal the number of clusters the response will be an "ARRAY"
+                if hasattr(self._sklearn_object, "n_clusters") and getattr(self._sklearn_object, "n_clusters") != len(self.output_cols):
+                    expected_dtype = "ARRAY"
+                # If this is a decomposition transformer, if the number of output columns does not equal the number of components the response will be an "ARRAY"
+                elif hasattr(self._sklearn_object, "n_components") and getattr(self._sklearn_object, "n_components") != len(self.output_cols):
+                    expected_dtype = "ARRAY"
+                else:
+                    output_types = [signature.as_snowpark_type() for signature in _infer_signature(dataset[self.input_cols], "output", use_snowflake_identifiers=True)]
+                    # We can only infer the output types from the input types if the following two statemetns are true:
+                    # 1) All of the output types are the same. Otherwise, we still have to fall back to variant because `_sklearn_inference` only accepts one type.
+                    # 2) The length of the input columns equals the length of the output columns. Otherwise the transform will likely result in an `ARRAY`.
+                    if all(x == output_types[0] for x in output_types) and len(output_types) == len(self.output_cols):
+                        expected_dtype = convert_sp_to_sf_type(output_types[0])
+            
             output_df = self._batch_inference(
                 dataset=dataset,
                 inference_method="transform",
@@ -730,8 +775,8 @@ class XGBClassifier(BaseTransformer):
 
         return output_df
     
-    @available_if(original_estimator_has_callable("fit_predict"))  # type: ignore[misc]
-    def fit_predict(self, dataset: Union[DataFrame, pd.DataFrame]) -> npt.NDArray[Any]:
+    @available_if(_is_fit_predict_method_enabled())  # type: ignore[misc]
+    def fit_predict(self, dataset: Union[DataFrame, pd.DataFrame]) -> Union[Any, npt.NDArray[Any]]:
         """ Method not supported for this class.
 
 
@@ -744,13 +789,21 @@ class XGBClassifier(BaseTransformer):
         Returns:
             Predicted dataset.
         """
-        if False:
-            self.fit(dataset)
-            assert self._sklearn_object is not None
-            labels : npt.NDArray[Any] = self._sklearn_object.labels_
-            return labels
-        else:
-            raise NotImplementedError
+        self.fit(dataset)
+        assert self._sklearn_object is not None
+        return self._sklearn_object.labels_
+
+
+    @available_if(_is_fit_transform_method_enabled())  # type: ignore[misc]
+    def fit_transform(self, dataset: Union[DataFrame, pd.DataFrame]) -> Union[Any, npt.NDArray[Any]]:
+        """ 
+        Returns:
+            Transformed dataset.
+        """
+        self.fit(dataset)
+        assert self._sklearn_object is not None
+        return self._sklearn_object.embedding_
+
 
     def _get_output_column_names(self, output_cols_prefix: str, output_cols: Optional[List[str]] = None) -> List[str]:
         """ Returns the list of output columns for predict_proba(), decision_function(), etc.. functions.

snowflake/ml/modeling/xgboost/xgb_regressor.py

@@ -53,6 +53,18 @@ _PROJECT = "ModelDevelopment"
 _SUBPROJECT = "".join([s.capitalize() for s in "xgboost".replace("sklearn.", "").split("_")])
 
 
+def _is_fit_predict_method_enabled() -> Callable[[Any], bool]:
+    def check(self: BaseTransformer) -> TypeGuard[Callable[..., object]]:
+        return False and callable(getattr(self._sklearn_object, "fit_predict", None))
+    return check
+
+
+def _is_fit_transform_method_enabled() -> Callable[[Any], bool]:
+    def check(self: BaseTransformer) -> TypeGuard[Callable[..., object]]:
+        return False and callable(getattr(self._sklearn_object, "fit_transform", None))
+    return check
+
+
 class XGBRegressor(BaseTransformer):
     r"""Implementation of the scikit-learn API for XGBoost regression
     For more details on this class, see [xgboost.XGBRegressor]
@@ -105,6 +117,22 @@ class XGBRegressor(BaseTransformer):
     drop_input_cols: Optional[bool], default=False
         If set, the response of predict(), transform() methods will not contain input columns.
 
+    use_external_memory_version: bool, default=False
+        If set, external memory version of XGBoost trainer is used. External memory training
+        is done in a two-step process. First,in the preprocessing step, input data is read and
+        parsed into an internal format, which can be CSR, CSC, or sorted CSC, and stored in
+        in-memory buffers. The in-memory buffers are continuously flushed out to disk when
+        predefined memory limit is reached. Second, in the tree construction step, the data
+        pages are streamed from disk via a multi-threaded pre-fetcher as needed for tree construction.
+        Note:'tree_method's 'approx', and 'hist' are supported in the external memory version.
+        Note:'grow_policy=depthwise' is used for optimal performance in the external memory version.
+
+    batch_size: int, default=10000
+        Number of rows in each batch of input data while using external memory training.
+        It is not recommended to set small batch sizes, like 32 samples per batch, as this
+        can seriously hurt performance in gradient boosting. Set the batch_size as large as possible
+        based on the available memory.
+
     n_estimators: int
             Number of gradient boosted trees.  Equivalent to number of boosting
             rounds.
@@ -325,6 +353,8 @@ class XGBRegressor(BaseTransformer):
         passthrough_cols: Optional[Union[str, Iterable[str]]] = None,
         drop_input_cols: Optional[bool] = False,
         sample_weight_col: Optional[str] = None,
+        use_external_memory_version: bool = False,
+        batch_size: int = 10000,
         **kwargs,
     ) -> None:
         super().__init__()
@@ -334,7 +364,9 @@ class XGBRegressor(BaseTransformer):
         self.set_label_cols(label_cols)
         self.set_passthrough_cols(passthrough_cols)
         self.set_drop_input_cols(drop_input_cols)
-        self.set_sample_weight_col(sample_weight_col)        
+        self.set_sample_weight_col(sample_weight_col)
+        self._use_external_memory_version = use_external_memory_version
+        self._batch_size = batch_size        
         deps: Set[str] = set([f'numpy=={np.__version__}', f'xgboost=={xgboost.__version__}', f'cloudpickle=={cp.__version__}'])
         
         self._deps = list(deps)
@@ -411,11 +443,6 @@ class XGBRegressor(BaseTransformer):
         if isinstance(dataset, DataFrame):
             session = dataset._session
             assert session is not None  # keep mypy happy
-            # Validate that key package version in user workspace are supported in snowflake conda channel
-            # If customer doesn't have package in conda channel, replace the ones have the closest versions
-            self._deps = pkg_version_utils.get_valid_pkg_versions_supported_in_snowflake_conda_channel(
-                pkg_versions=self._get_dependencies(), session=session, subproject=_SUBPROJECT)
-            
             # Specify input columns so column pruning will be enforced
             selected_cols = self._get_active_columns()
             if len(selected_cols) > 0:
@@ -443,7 +470,9 @@ class XGBRegressor(BaseTransformer):
             label_cols=self.label_cols,
             sample_weight_col=self.sample_weight_col,
             autogenerated=self._autogenerated,
-            subproject=_SUBPROJECT
+            subproject=_SUBPROJECT,
+            use_external_memory_version=self._use_external_memory_version,
+            batch_size=self._batch_size,
         )
         self._sklearn_object = model_trainer.train()
         self._is_fitted = True
@@ -714,6 +743,22 @@ class XGBRegressor(BaseTransformer):
                 # each row containing a list of values.
                 expected_dtype = "ARRAY"
 
+            # If we were unable to assign a type to this transform in the factory, infer the type here.
+            if expected_dtype == "":
+                # If this is a clustering transformer, if the number of output columns does not equal the number of clusters the response will be an "ARRAY"
+                if hasattr(self._sklearn_object, "n_clusters") and getattr(self._sklearn_object, "n_clusters") != len(self.output_cols):
+                    expected_dtype = "ARRAY"
+                # If this is a decomposition transformer, if the number of output columns does not equal the number of components the response will be an "ARRAY"
+                elif hasattr(self._sklearn_object, "n_components") and getattr(self._sklearn_object, "n_components") != len(self.output_cols):
+                    expected_dtype = "ARRAY"
+                else:
+                    output_types = [signature.as_snowpark_type() for signature in _infer_signature(dataset[self.input_cols], "output", use_snowflake_identifiers=True)]
+                    # We can only infer the output types from the input types if the following two statemetns are true:
+                    # 1) All of the output types are the same. Otherwise, we still have to fall back to variant because `_sklearn_inference` only accepts one type.
+                    # 2) The length of the input columns equals the length of the output columns. Otherwise the transform will likely result in an `ARRAY`.
+                    if all(x == output_types[0] for x in output_types) and len(output_types) == len(self.output_cols):
+                        expected_dtype = convert_sp_to_sf_type(output_types[0])
+            
             output_df = self._batch_inference(
                 dataset=dataset,
                 inference_method="transform",
@@ -729,8 +774,8 @@ class XGBRegressor(BaseTransformer):
 
         return output_df
     
-    @available_if(original_estimator_has_callable("fit_predict"))  # type: ignore[misc]
-    def fit_predict(self, dataset: Union[DataFrame, pd.DataFrame]) -> npt.NDArray[Any]:
+    @available_if(_is_fit_predict_method_enabled())  # type: ignore[misc]
+    def fit_predict(self, dataset: Union[DataFrame, pd.DataFrame]) -> Union[Any, npt.NDArray[Any]]:
         """ Method not supported for this class.
 
 
@@ -743,13 +788,21 @@ class XGBRegressor(BaseTransformer):
         Returns:
             Predicted dataset.
         """
-        if False:
-            self.fit(dataset)
-            assert self._sklearn_object is not None
-            labels : npt.NDArray[Any] = self._sklearn_object.labels_
-            return labels
-        else:
-            raise NotImplementedError
+        self.fit(dataset)
+        assert self._sklearn_object is not None
+        return self._sklearn_object.labels_
+
+
+    @available_if(_is_fit_transform_method_enabled())  # type: ignore[misc]
+    def fit_transform(self, dataset: Union[DataFrame, pd.DataFrame]) -> Union[Any, npt.NDArray[Any]]:
+        """ 
+        Returns:
+            Transformed dataset.
+        """
+        self.fit(dataset)
+        assert self._sklearn_object is not None
+        return self._sklearn_object.embedding_
+
 
     def _get_output_column_names(self, output_cols_prefix: str, output_cols: Optional[List[str]] = None) -> List[str]:
         """ Returns the list of output columns for predict_proba(), decision_function(), etc.. functions.

snowflake/ml/modeling/xgboost/xgbrf_classifier.py

@@ -53,6 +53,18 @@ _PROJECT = "ModelDevelopment"
 _SUBPROJECT = "".join([s.capitalize() for s in "xgboost".replace("sklearn.", "").split("_")])
 
 
+def _is_fit_predict_method_enabled() -> Callable[[Any], bool]:
+    def check(self: BaseTransformer) -> TypeGuard[Callable[..., object]]:
+        return False and callable(getattr(self._sklearn_object, "fit_predict", None))
+    return check
+
+
+def _is_fit_transform_method_enabled() -> Callable[[Any], bool]:
+    def check(self: BaseTransformer) -> TypeGuard[Callable[..., object]]:
+        return False and callable(getattr(self._sklearn_object, "fit_transform", None))
+    return check
+
+
 class XGBRFClassifier(BaseTransformer):
     r"""scikit-learn API for XGBoost random forest classification
     For more details on this class, see [xgboost.XGBRFClassifier]
@@ -105,6 +117,22 @@ class XGBRFClassifier(BaseTransformer):
     drop_input_cols: Optional[bool], default=False
         If set, the response of predict(), transform() methods will not contain input columns.
 
+    use_external_memory_version: bool, default=False
+        If set, external memory version of XGBoost trainer is used. External memory training
+        is done in a two-step process. First,in the preprocessing step, input data is read and
+        parsed into an internal format, which can be CSR, CSC, or sorted CSC, and stored in
+        in-memory buffers. The in-memory buffers are continuously flushed out to disk when
+        predefined memory limit is reached. Second, in the tree construction step, the data
+        pages are streamed from disk via a multi-threaded pre-fetcher as needed for tree construction.
+        Note:'tree_method's 'approx', and 'hist' are supported in the external memory version.
+        Note:'grow_policy=depthwise' is used for optimal performance in the external memory version.
+
+    batch_size: int, default=10000
+        Number of rows in each batch of input data while using external memory training.
+        It is not recommended to set small batch sizes, like 32 samples per batch, as this
+        can seriously hurt performance in gradient boosting. Set the batch_size as large as possible
+        based on the available memory.
+
     n_estimators: int
             Number of trees in random forest to fit.
 
@@ -327,6 +355,8 @@ class XGBRFClassifier(BaseTransformer):
         passthrough_cols: Optional[Union[str, Iterable[str]]] = None,
         drop_input_cols: Optional[bool] = False,
         sample_weight_col: Optional[str] = None,
+        use_external_memory_version: bool = False,
+        batch_size: int = 10000,
         **kwargs,
     ) -> None:
         super().__init__()
@@ -336,7 +366,9 @@ class XGBRFClassifier(BaseTransformer):
         self.set_label_cols(label_cols)
         self.set_passthrough_cols(passthrough_cols)
         self.set_drop_input_cols(drop_input_cols)
-        self.set_sample_weight_col(sample_weight_col)        
+        self.set_sample_weight_col(sample_weight_col)
+        self._use_external_memory_version = use_external_memory_version
+        self._batch_size = batch_size        
         deps: Set[str] = set([f'numpy=={np.__version__}', f'xgboost=={xgboost.__version__}', f'cloudpickle=={cp.__version__}'])
         
         self._deps = list(deps)
@@ -416,11 +448,6 @@ class XGBRFClassifier(BaseTransformer):
         if isinstance(dataset, DataFrame):
             session = dataset._session
             assert session is not None  # keep mypy happy
-            # Validate that key package version in user workspace are supported in snowflake conda channel
-            # If customer doesn't have package in conda channel, replace the ones have the closest versions
-            self._deps = pkg_version_utils.get_valid_pkg_versions_supported_in_snowflake_conda_channel(
-                pkg_versions=self._get_dependencies(), session=session, subproject=_SUBPROJECT)
-            
             # Specify input columns so column pruning will be enforced
             selected_cols = self._get_active_columns()
             if len(selected_cols) > 0:
@@ -448,7 +475,9 @@ class XGBRFClassifier(BaseTransformer):
             label_cols=self.label_cols,
             sample_weight_col=self.sample_weight_col,
             autogenerated=self._autogenerated,
-            subproject=_SUBPROJECT
+            subproject=_SUBPROJECT,
+            use_external_memory_version=self._use_external_memory_version,
+            batch_size=self._batch_size,
         )
         self._sklearn_object = model_trainer.train()
         self._is_fitted = True
@@ -719,6 +748,22 @@ class XGBRFClassifier(BaseTransformer):
                 # each row containing a list of values.
                 expected_dtype = "ARRAY"
 
+            # If we were unable to assign a type to this transform in the factory, infer the type here.
+            if expected_dtype == "":
+                # If this is a clustering transformer, if the number of output columns does not equal the number of clusters the response will be an "ARRAY"
+                if hasattr(self._sklearn_object, "n_clusters") and getattr(self._sklearn_object, "n_clusters") != len(self.output_cols):
+                    expected_dtype = "ARRAY"
+                # If this is a decomposition transformer, if the number of output columns does not equal the number of components the response will be an "ARRAY"
+                elif hasattr(self._sklearn_object, "n_components") and getattr(self._sklearn_object, "n_components") != len(self.output_cols):
+                    expected_dtype = "ARRAY"
+                else:
+                    output_types = [signature.as_snowpark_type() for signature in _infer_signature(dataset[self.input_cols], "output", use_snowflake_identifiers=True)]
+                    # We can only infer the output types from the input types if the following two statemetns are true:
+                    # 1) All of the output types are the same. Otherwise, we still have to fall back to variant because `_sklearn_inference` only accepts one type.
+                    # 2) The length of the input columns equals the length of the output columns. Otherwise the transform will likely result in an `ARRAY`.
+                    if all(x == output_types[0] for x in output_types) and len(output_types) == len(self.output_cols):
+                        expected_dtype = convert_sp_to_sf_type(output_types[0])
+            
             output_df = self._batch_inference(
                 dataset=dataset,
                 inference_method="transform",
@@ -734,8 +779,8 @@ class XGBRFClassifier(BaseTransformer):
 
         return output_df
     
-    @available_if(original_estimator_has_callable("fit_predict"))  # type: ignore[misc]
-    def fit_predict(self, dataset: Union[DataFrame, pd.DataFrame]) -> npt.NDArray[Any]:
+    @available_if(_is_fit_predict_method_enabled())  # type: ignore[misc]
+    def fit_predict(self, dataset: Union[DataFrame, pd.DataFrame]) -> Union[Any, npt.NDArray[Any]]:
         """ Method not supported for this class.
 
 
@@ -748,13 +793,21 @@ class XGBRFClassifier(BaseTransformer):
         Returns:
             Predicted dataset.
         """
-        if False:
-            self.fit(dataset)
-            assert self._sklearn_object is not None
-            labels : npt.NDArray[Any] = self._sklearn_object.labels_
-            return labels
-        else:
-            raise NotImplementedError
+        self.fit(dataset)
+        assert self._sklearn_object is not None
+        return self._sklearn_object.labels_
+
+
+    @available_if(_is_fit_transform_method_enabled())  # type: ignore[misc]
+    def fit_transform(self, dataset: Union[DataFrame, pd.DataFrame]) -> Union[Any, npt.NDArray[Any]]:
+        """ 
+        Returns:
+            Transformed dataset.
+        """
+        self.fit(dataset)
+        assert self._sklearn_object is not None
+        return self._sklearn_object.embedding_
+
 
     def _get_output_column_names(self, output_cols_prefix: str, output_cols: Optional[List[str]] = None) -> List[str]:
         """ Returns the list of output columns for predict_proba(), decision_function(), etc.. functions.

snowflake/ml/modeling/xgboost/xgbrf_regressor.py

@@ -53,6 +53,18 @@ _PROJECT = "ModelDevelopment"
 _SUBPROJECT = "".join([s.capitalize() for s in "xgboost".replace("sklearn.", "").split("_")])
 
 
+def _is_fit_predict_method_enabled() -> Callable[[Any], bool]:
+    def check(self: BaseTransformer) -> TypeGuard[Callable[..., object]]:
+        return False and callable(getattr(self._sklearn_object, "fit_predict", None))
+    return check
+
+
+def _is_fit_transform_method_enabled() -> Callable[[Any], bool]:
+    def check(self: BaseTransformer) -> TypeGuard[Callable[..., object]]:
+        return False and callable(getattr(self._sklearn_object, "fit_transform", None))
+    return check
+
+
 class XGBRFRegressor(BaseTransformer):
     r"""scikit-learn API for XGBoost random forest regression
     For more details on this class, see [xgboost.XGBRFRegressor]
@@ -105,6 +117,22 @@ class XGBRFRegressor(BaseTransformer):
     drop_input_cols: Optional[bool], default=False
         If set, the response of predict(), transform() methods will not contain input columns.
 
+    use_external_memory_version: bool, default=False
+        If set, external memory version of XGBoost trainer is used. External memory training
+        is done in a two-step process. First,in the preprocessing step, input data is read and
+        parsed into an internal format, which can be CSR, CSC, or sorted CSC, and stored in
+        in-memory buffers. The in-memory buffers are continuously flushed out to disk when
+        predefined memory limit is reached. Second, in the tree construction step, the data
+        pages are streamed from disk via a multi-threaded pre-fetcher as needed for tree construction.
+        Note:'tree_method's 'approx', and 'hist' are supported in the external memory version.
+        Note:'grow_policy=depthwise' is used for optimal performance in the external memory version.
+
+    batch_size: int, default=10000
+        Number of rows in each batch of input data while using external memory training.
+        It is not recommended to set small batch sizes, like 32 samples per batch, as this
+        can seriously hurt performance in gradient boosting. Set the batch_size as large as possible
+        based on the available memory.
+
     n_estimators: int
             Number of trees in random forest to fit.
 
@@ -327,6 +355,8 @@ class XGBRFRegressor(BaseTransformer):
         passthrough_cols: Optional[Union[str, Iterable[str]]] = None,
         drop_input_cols: Optional[bool] = False,
         sample_weight_col: Optional[str] = None,
+        use_external_memory_version: bool = False,
+        batch_size: int = 10000,
         **kwargs,
     ) -> None:
         super().__init__()
@@ -336,7 +366,9 @@ class XGBRFRegressor(BaseTransformer):
         self.set_label_cols(label_cols)
         self.set_passthrough_cols(passthrough_cols)
         self.set_drop_input_cols(drop_input_cols)
-        self.set_sample_weight_col(sample_weight_col)        
+        self.set_sample_weight_col(sample_weight_col)
+        self._use_external_memory_version = use_external_memory_version
+        self._batch_size = batch_size        
         deps: Set[str] = set([f'numpy=={np.__version__}', f'xgboost=={xgboost.__version__}', f'cloudpickle=={cp.__version__}'])
         
         self._deps = list(deps)
@@ -416,11 +448,6 @@ class XGBRFRegressor(BaseTransformer):
         if isinstance(dataset, DataFrame):
             session = dataset._session
             assert session is not None  # keep mypy happy
-            # Validate that key package version in user workspace are supported in snowflake conda channel
-            # If customer doesn't have package in conda channel, replace the ones have the closest versions
-            self._deps = pkg_version_utils.get_valid_pkg_versions_supported_in_snowflake_conda_channel(
-                pkg_versions=self._get_dependencies(), session=session, subproject=_SUBPROJECT)
-            
             # Specify input columns so column pruning will be enforced
             selected_cols = self._get_active_columns()
             if len(selected_cols) > 0:
@@ -448,7 +475,9 @@ class XGBRFRegressor(BaseTransformer):
             label_cols=self.label_cols,
             sample_weight_col=self.sample_weight_col,
             autogenerated=self._autogenerated,
-            subproject=_SUBPROJECT
+            subproject=_SUBPROJECT,
+            use_external_memory_version=self._use_external_memory_version,
+            batch_size=self._batch_size,
         )
         self._sklearn_object = model_trainer.train()
         self._is_fitted = True
@@ -719,6 +748,22 @@ class XGBRFRegressor(BaseTransformer):
                 # each row containing a list of values.
                 expected_dtype = "ARRAY"
 
+            # If we were unable to assign a type to this transform in the factory, infer the type here.
+            if expected_dtype == "":
+                # If this is a clustering transformer, if the number of output columns does not equal the number of clusters the response will be an "ARRAY"
+                if hasattr(self._sklearn_object, "n_clusters") and getattr(self._sklearn_object, "n_clusters") != len(self.output_cols):
+                    expected_dtype = "ARRAY"
+                # If this is a decomposition transformer, if the number of output columns does not equal the number of components the response will be an "ARRAY"
+                elif hasattr(self._sklearn_object, "n_components") and getattr(self._sklearn_object, "n_components") != len(self.output_cols):
+                    expected_dtype = "ARRAY"
+                else:
+                    output_types = [signature.as_snowpark_type() for signature in _infer_signature(dataset[self.input_cols], "output", use_snowflake_identifiers=True)]
+                    # We can only infer the output types from the input types if the following two statemetns are true:
+                    # 1) All of the output types are the same. Otherwise, we still have to fall back to variant because `_sklearn_inference` only accepts one type.
+                    # 2) The length of the input columns equals the length of the output columns. Otherwise the transform will likely result in an `ARRAY`.
+                    if all(x == output_types[0] for x in output_types) and len(output_types) == len(self.output_cols):
+                        expected_dtype = convert_sp_to_sf_type(output_types[0])
+            
             output_df = self._batch_inference(
                 dataset=dataset,
                 inference_method="transform",
@@ -734,8 +779,8 @@ class XGBRFRegressor(BaseTransformer):
 
         return output_df
     
-    @available_if(original_estimator_has_callable("fit_predict"))  # type: ignore[misc]
-    def fit_predict(self, dataset: Union[DataFrame, pd.DataFrame]) -> npt.NDArray[Any]:
+    @available_if(_is_fit_predict_method_enabled())  # type: ignore[misc]
+    def fit_predict(self, dataset: Union[DataFrame, pd.DataFrame]) -> Union[Any, npt.NDArray[Any]]:
         """ Method not supported for this class.
 
 
@@ -748,13 +793,21 @@ class XGBRFRegressor(BaseTransformer):
         Returns:
             Predicted dataset.
         """
-        if False:
-            self.fit(dataset)
-            assert self._sklearn_object is not None
-            labels : npt.NDArray[Any] = self._sklearn_object.labels_
-            return labels
-        else:
-            raise NotImplementedError
+        self.fit(dataset)
+        assert self._sklearn_object is not None
+        return self._sklearn_object.labels_
+
+
+    @available_if(_is_fit_transform_method_enabled())  # type: ignore[misc]
+    def fit_transform(self, dataset: Union[DataFrame, pd.DataFrame]) -> Union[Any, npt.NDArray[Any]]:
+        """ 
+        Returns:
+            Transformed dataset.
+        """
+        self.fit(dataset)
+        assert self._sklearn_object is not None
+        return self._sklearn_object.embedding_
+
 
     def _get_output_column_names(self, output_cols_prefix: str, output_cols: Optional[List[str]] = None) -> List[str]:
         """ Returns the list of output columns for predict_proba(), decision_function(), etc.. functions.

snowflake/ml/registry/__init__.py

@@ -0,0 +1,3 @@
+from snowflake.ml.registry.registry import Registry
+
+__all__ = ["Registry"]