FEAT - add support for numpy arrays to the TableReport (#1676)

Co-authored-by: Nismamjad1 <nisma.amjad1@gmail.com>
Co-authored-by: Riccardo Cappuzzo <7548232+rcap107@users.noreply.github.com>

Author: 3 people

CHANGES.rst

@@ -26,6 +26,7 @@ New features
   environments produced by a CV splitter -- similar to
   :meth:`DataOp.skb.train_test_split` but for multiple cross-validation splits.
   :pr:`1653` by :user:`Jérôme Dockès <jeromedockes>`.
+- :class:`TableReport` now supports ``np.array``. :pr:`1676` by :user: `Nisma Amjad <Nismamjad1>`.
 - :meth:`DataOp.skb.full_report` now accepts a new parameter, title, that is displayed
   in the html report.
   :pr:`1654` by :user:`Marie Sacksick <MarieSacksick>`.

skrub/_apply_to_cols.py

@@ -220,6 +220,10 @@ class ApplyToCols(TransformerMixin, BaseEstimator):
     """Map a transformer to columns in a dataframe.
 
     A separate clone of the transformer is applied to each column separately.
+
+    Columns that are not selected in the ``cols`` parameter are passed through
+    without modification.
+
     All columns not listed in ``cols`` remain unmodified in the output.
     Moreover, if ``allow_reject`` is ``True`` and the transformers'
     ``fit_transform`` raises a ``RejectColumn`` exception for a particular

skrub/_reporting/_table_report.py

@@ -4,6 +4,9 @@
 import numbers
 from pathlib import Path
 
+import numpy as np
+import pandas as pd
+
 from .. import _config
 from .. import _dataframe as sbd
 from ._html import to_html
@@ -46,9 +49,9 @@ def _check_max_cols(max_plot_columns, max_association_columns):
 class TableReport:
     r"""Summarize the contents of a dataframe.
 
-    This class summarizes a dataframe, providing information such as the type
-    and summary statistics (mean, number of missing values, etc.) for each
-    column.
+    This class summarizes a dataframe or numpy array, providing information such as
+    the type and summary statistics (mean, number of missing values, etc.) for each
+    column. Numpy arrays are converted to pandas DataFrame or Series.
 
     Parameters
     ----------
@@ -186,6 +189,21 @@ def __init__(
         max_plot_columns=None,
         max_association_columns=None,
     ):
+        if isinstance(dataframe, np.ndarray):
+            if dataframe.ndim == 1:
+                dataframe = pd.Series(dataframe, name="0")
+
+            elif dataframe.ndim == 2:
+                dataframe = pd.DataFrame(
+                    dataframe, columns=[str(i) for i in range(dataframe.shape[1])]
+                )
+
+            else:
+                raise ValueError(
+                    f"Input NumPy array has {dataframe.ndim} dimensions. "
+                    "TableReport only supports 1D and 2D arrays"
+                )
+
         n_rows = max(1, n_rows)
         self._summary_kwargs = {
             "order_by": order_by,

skrub/_reporting/tests/test_table_report.py

@@ -337,3 +337,45 @@ def test_bad_cols_parameter(pd_module, arg):
     df = pd_module.example_dataframe
     with pytest.raises(ValueError):
         TableReport(df, **{arg: -1})
+
+
+def test_array_dim_check():
+    array_3d = np.array([[[1, 2], [3, 4]], [[5, 6], [7, 8]]])
+    assert array_3d.ndim == 3
+    with pytest.raises(ValueError, match=r"Input (NumPy )?array has 3 dimensions"):
+        TableReport(array_3d)
+
+    array_4d = np.array([[[[1]]]])
+    assert array_4d.ndim == 4
+    with pytest.raises(ValueError, match=r"Input (NumPy )?array has 4 dimensions"):
+        TableReport(array_4d)
+
+    array_1d = np.array([1, 2, 3])
+    assert array_1d.ndim == 1
+
+    TableReport(array_1d)
+
+
+numpy_test_cases = [
+    (
+        np.array(
+            [
+                [1, 2, 3],
+                [
+                    4,
+                    5,
+                    6,
+                ],
+            ]
+        ),
+        3,
+    ),
+    (np.array([[10, 20], [30, 40], [50, 60], [60, 70]]), 2),
+]
+
+
+@pytest.mark.parametrize("input_array, expected_columns", numpy_test_cases)
+def test_numpy_array_columns(input_array, expected_columns):
+    report = TableReport(input_array, max_association_columns=0)
+
+    assert report._summary["n_columns"] == expected_columns