Docs: mention pytest.approx in assert and getting-started documentation (#13788)

Fixes #13662

Author: tanuj-rai

doc/en/getting-started.rst

@@ -199,6 +199,26 @@ This is outlined below:
 
 Note that attributes added at class level are *class attributes*, so they will be shared between tests.
 
+Compare floating-point values with pytest.approx
+--------------------------------------------------------------
+
+``pytest`` also provides a number of utilities to make writing tests easier.
+For example, you can use :func:`pytest.approx` to compare floating-point
+values that may have small rounding errors:
+
+.. code-block:: python
+
+    # content of test_approx.py
+    import pytest
+
+
+    def test_sum():
+        assert (0.1 + 0.2) == pytest.approx(0.3)
+
+This avoids the need for manual tolerance checks or using
+``math.isclose`` and works with scalars, lists, and NumPy arrays.
+
+
 Request a unique temporary directory for functional tests
 --------------------------------------------------------------
 

doc/en/how-to/assert.rst

@@ -66,6 +66,33 @@ See :ref:`assert-details` for more information on assertion introspection.
 
 .. _`assertraises`:
 
+Assertions about approximate equality
+-------------------------------------
+
+When comparing floating point values (or arrays of floats), small rounding
+errors are common. Instead of using ``assert abs(a - b) < tol`` or
+``numpy.isclose``, you can use :func:`pytest.approx`:
+
+.. code-block:: python
+
+    import pytest
+    import numpy as np
+
+
+    def test_floats():
+        assert (0.1 + 0.2) == pytest.approx(0.3)
+
+
+    def test_arrays():
+        a = np.array([1.0, 2.0, 3.0])
+        b = np.array([0.9999, 2.0001, 3.0])
+        assert a == pytest.approx(b)
+
+``pytest.approx`` works with scalars, lists, dictionaries, and NumPy arrays.
+It also supports comparisons involving NaNs.
+
+See :func:`pytest.approx` for details.
+
 Assertions about expected exceptions
 ------------------------------------------