Make TextVectorization work with list input.

PiperOrigin-RevId: 379498023

keras/layers/preprocessing/text_vectorization.py

@@ -465,7 +465,7 @@ def build(self, input_shape):
     # expression to evaluate to False instead of True if the shape is undefined;
     # the expression needs to evaluate to True in that case.
     if self._split is not None:
-      if input_shape.ndims > 1 and not input_shape[-1] == 1:  # pylint: disable=g-comparison-negation
+      if input_shape and input_shape.ndims > 1 and not input_shape[-1] == 1:  # pylint: disable=g-comparison-negation
         raise RuntimeError(
             "When using TextVectorization to tokenize strings, the innermost "
             "dimension of the input array must be 1, got shape "

keras/layers/preprocessing/text_vectorization_test.py

@@ -453,6 +453,11 @@ def test_layer_dimensionality_handling_with_split(self, data, expected):
     output = vectorization(tf.ragged.constant(data, inner_shape=(1,)))
     self.assertAllEqual(expected, output)
 
+  def test_layer_list_input(self):
+    output = text_vectorization.TextVectorization()(["a", "b", "c"])
+    expected_output = [[1], [1], [1]]
+    self.assertEqual(output.numpy().tolist(), expected_output)
+
 
 @keras_parameterized.run_all_keras_modes(always_skip_v1=True)
 class TextVectorizationPreprocessingTest(