MAINT - Removing some deprecated features

CHANGES.rst

@@ -120,7 +120,14 @@ Changes
   :pr:`1628` by :user:`Jérôme Dockès <jeromedockes>`.
 - The parameter ``splitter`` of :meth:`DataOp.skb.train_test_split` has been
   renamed ``split_func``. :pr:`1630` by :user:`Jérôme Dockès <jeromedockes>`.
-
+- KEN embeddings and all the relevant functions have been removed from skrub.
+  :pr:`1567` by :user:`Riccardo Cappuzzo<rcap107>`.
+- The objects ``tabular_learner`` and ``DropIfTooManyNulls`` were removed. Use
+  :func:`tabular_pipeline` and :class:`DropUninformative` instead.
+  :pr:`1567` by :user:`Riccardo Cappuzzo<rcap107>`.
+- The skrub global configuration now includes a parameter for setting the default
+  verbosity of the :class:`TableReport`.
+  :pr:`1567` by :user:`Riccardo Cappuzzo<rcap107>`.
 
 Bugfixes
 --------

doc/modules/configuration_and_utils/customizing_configuration.rst

@@ -40,7 +40,7 @@ are available by using
 >>> import skrub
 >>> config = skrub.get_config()
 >>> config.keys()
-dict_keys(['use_table_report', 'use_table_report_data_ops', 'max_plot_columns', 'max_association_columns', 'subsampling_seed', 'enable_subsampling', 'float_precision', 'cardinality_threshold'])
+dict_keys(['use_table_report', 'use_table_report_data_ops', 'table_report_verbosity', 'max_plot_columns', 'max_association_columns', 'subsampling_seed', 'enable_subsampling', 'float_precision', 'cardinality_threshold'])
 
 These are the parameters currently available in the global configuration:
 

skrub/__init__.py

@@ -43,7 +43,7 @@
 from ._squashing_scaler import SquashingScaler
 from ._string_encoder import StringEncoder
 from ._table_vectorizer import Cleaner, TableVectorizer
-from ._tabular_pipeline import tabular_learner, tabular_pipeline
+from ._tabular_pipeline import tabular_pipeline
 from ._text_encoder import TextEncoder
 from ._to_categorical import ToCategorical
 from ._to_datetime import ToDatetime, to_datetime
@@ -64,10 +64,7 @@
     "deferred",
     "eval_mode",
     "TableReport",
-    "patch_display",
-    "unpatch_display",
     "tabular_pipeline",
-    "tabular_learner",
     "DatetimeEncoder",
     "ToDatetime",
     "Joiner",
@@ -100,6 +97,8 @@
     "StringEncoder",
     "column_associations",
     "SquashingScaler",
+    "patch_display",
+    "unpatch_display",
     "get_config",
     "set_config",
     "config_context",

skrub/_config.py

@@ -25,6 +25,7 @@ def _parse_env_bool(env_variable_name, default):
 _global_config = {
     "use_table_report": _parse_env_bool("SKB_USE_TABLE_REPORT", False),
     "use_table_report_data_ops": _parse_env_bool("SKB_USE_TABLE_REPORT_DATA_OPS", True),
+    "table_report_verbosity": int(os.environ.get("SKB_TABLE_REPORT_VERBOSITY", 1)),
     "max_plot_columns": int(os.environ.get("SKB_MAX_PLOT_COLUMNS", 30)),
     "max_association_columns": int(os.environ.get("SKB_MAX_ASSOCIATION_COLUMNS", 30)),
     "subsampling_seed": int(os.environ.get("SKB_SUBSAMPLING_SEED", 0)),
@@ -74,6 +75,7 @@ def _apply_external_patches(config):
         _patching._patch_display(
             max_plot_columns=config["max_plot_columns"],
             max_association_columns=config["max_plot_columns"],
+            verbose=config["table_report_verbosity"],
         )
     else:
         # No-op if dispatch haven't been previously enabled
@@ -83,6 +85,7 @@ def _apply_external_patches(config):
 def set_config(
     use_table_report=None,
     use_table_report_data_ops=None,
+    table_report_verbosity=None,
     max_plot_columns=None,
     max_association_columns=None,
     subsampling_seed=None,
@@ -116,6 +119,11 @@ def set_config(
         This configuration can also be set with the ``SKB_USE_TABLE_REPORT_DATA_OPS``
         environment variable.
 
+    table_report_verbosity : int, default=None
+        Set the level of verbosity of the :class:`~skrub.TableReport`.
+        Default is 1 (print the progress bar). Refer to the ``TableReport``
+        documentation for more details.
+
     max_plot_columns : int, default=None
         Set the ``max_plot_columns`` argument of :class:`~skrub.TableReport`.
         Default is 30. If "all", all columns will be plotted.
@@ -191,6 +199,17 @@ def set_config(
             )
         local_config["use_table_report_data_ops"] = use_table_report_data_ops
 
+    if table_report_verbosity is not None:
+        if (
+            not isinstance(table_report_verbosity, numbers.Integral)
+            or table_report_verbosity < 0
+        ):
+            raise ValueError(
+                "'table_report_verbosity' must be a non-negative integer, got"
+                f" {table_report_verbosity!r}"
+            )
+        local_config["table_report_verbosity"] = table_report_verbosity
+
     if max_plot_columns is not None:
         if not isinstance(max_plot_columns, numbers.Real) and max_plot_columns != "all":
             raise ValueError(
@@ -247,6 +266,7 @@ def config_context(
     *,
     use_table_report=None,
     use_table_report_data_ops=None,
+    table_report_verbosity=None,
     max_plot_columns=None,
     max_association_columns=None,
     subsampling_seed=None,
@@ -280,6 +300,11 @@ def config_context(
         This configuration can also be set with the ``SKB_USE_TABLE_REPORT_DATA_OPS``
         environment variable.
 
+    table_report_verbosity : int, default=None
+        Set the level of verbosity of the :class:`~skrub.TableReport`.
+        Default is 0 (no verbosity). Refer to the ``TableReport`` documentation for
+        more details.
+
     max_plot_columns : int, default=None
         Set the ``max_plot_columns`` argument of :class:`~skrub.TableReport`.
         Default is 30. If "all", all columns will be plotted.
@@ -348,6 +373,7 @@ def config_context(
     set_config(
         use_table_report=use_table_report,
         use_table_report_data_ops=use_table_report_data_ops,
+        table_report_verbosity=table_report_verbosity,
         max_plot_columns=max_plot_columns,
         max_association_columns=max_association_columns,
         subsampling_seed=subsampling_seed,

skrub/_reporting/_patching.py

@@ -1,5 +1,4 @@
 import importlib
-import warnings
 
 from ._table_report import TableReport
 
@@ -66,10 +65,6 @@ def patch_display(
 ):
     """Replace the default DataFrame HTML displays with ``skrub.TableReport``.
 
-    .. deprecated:: 0.6.0
-        The functionality provided by this function is now implemented in
-        :func:`~skrub.set_config`.
-
     This function replaces the HTML displays (what is shown when an object is
     the output of a jupyter notebook cell) of pandas and polars DataFrames
     with a TableReport.
@@ -112,13 +107,6 @@ def patch_display(
         max_plot_columns=max_plot_columns,
         max_association_columns=max_association_columns,
     )
-    warnings.warn(
-        (
-            "patch_display will be deprecated in the next release. "
-            "Equivalent functionality is available in skrub.set_config."
-        ),
-        category=FutureWarning,
-    )
 
 
 def _patch_display(
@@ -136,10 +124,6 @@ def _patch_display(
 def unpatch_display(pandas=True, polars=True):
     """Undo the effect of ``skrub.patch_display()``.
 
-    .. deprecated:: 0.6.0
-        The functionality provided by this function is now implemented in
-        :func:`~skrub.set_config`.
-
     This function restores the default HTML displays of pandas and polars
     DataFrames.
 
@@ -159,13 +143,6 @@ def unpatch_display(pandas=True, polars=True):
         Directly create a report from a dataframe.
     """
     _change_display(_unpatch, _get_to_patch(pandas=pandas, polars=polars))
-    warnings.warn(
-        (
-            "unpatch_display will be deprecated in the next release. "
-            "Equivalent functionality is available in skrub.set_config."
-        ),
-        category=FutureWarning,
-    )
 
 
 def _unpatch_display(pandas=True, polars=True):

skrub/_reporting/_table_report.py

@@ -193,7 +193,7 @@ def __init__(
         order_by=None,
         title=None,
         column_filters=None,
-        verbose=1,
+        verbose=None,
         max_plot_columns=None,
         max_association_columns=None,
         open_tab="table",
@@ -214,6 +214,10 @@ def __init__(
                 )
 
         n_rows = max(1, n_rows)
+        if verbose is None:
+            self.verbose = _config.get_config()["table_report_verbosity"]
+        else:
+            self.verbose = verbose
 
         # Validate open_tab parameter
         valid_tabs = ["table", "stats", "distributions", "associations"]
@@ -227,12 +231,11 @@ def __init__(
             "order_by": order_by,
             "max_top_slice_size": -(n_rows // -2),
             "max_bottom_slice_size": n_rows // 2,
-            "verbose": verbose,
+            "verbose": self.verbose,
         }
         self._to_html_kwargs = {}
         self.title = title
         self.column_filters = column_filters
-        self.verbose = verbose
         self.max_plot_columns, self.max_association_columns = _check_max_cols(
             max_plot_columns, max_association_columns
         )