Fix lookup layer oov token check when num_oov_indices > len(vocabulary tokens)

This was always broken for numpy vocabulary inputs, and recently broke for lists as well.

PiperOrigin-RevId: 380946490

keras/layers/preprocessing/index_lookup.py

@@ -411,11 +411,7 @@ class IndexLookup(base_preprocessing_layer.PreprocessingLayer):
     should_have_oov = (self.num_oov_indices > 0)
     expected_oov = [self.oov_token] * self.num_oov_indices
     found_oov = vocabulary[oov_start:token_start]
-    has_oov = should_have_oov and found_oov == expected_oov
-    # If we get a numpy array, then has_oov may end up being a numpy array
-    # instead of a bool. Fix this by collapsing the variable if it's not bool.
-    if not isinstance(has_oov, bool):
-      has_oov = any(has_oov)
+    has_oov = should_have_oov and np.array_equal(found_oov, expected_oov)
 
     if all([should_have_mask, has_mask, should_have_oov]) and not has_oov:
       raise ValueError(

keras/layers/preprocessing/index_lookup_test.py

@@ -1535,6 +1535,31 @@ class IndexLookupVocabularyTest(keras_parameterized.TestCase,
     self.assertAllEqual(returned_vocab, ["wind", "and", "fire"])
     self.assertAllEqual(layer.vocabulary_size(), 5)
 
+  def test_vocab_multi_oov(self):
+    vocab_data = ["", "[OOV]", "[OOV]", "wind", "and", "fire"]
+    layer = index_lookup.IndexLookup(
+        max_tokens=None,
+        num_oov_indices=2,
+        mask_token="",
+        oov_token="[OOV]",
+        dtype=tf.string)
+    layer.set_vocabulary(vocab_data)
+    returned_vocab = layer.get_vocabulary()
+    self.assertAllEqual(returned_vocab, vocab_data)
+
+  def test_vocab_multi_oov_not_present(self):
+    vocab_data = ["wind", "and", "fire"]
+    layer = index_lookup.IndexLookup(
+        max_tokens=None,
+        num_oov_indices=10,
+        mask_token="",
+        oov_token="[OOV]",
+        dtype=tf.string)
+    layer.set_vocabulary(vocab_data)
+    returned_vocab = layer.get_vocabulary()
+    self.assertAllEqual(returned_vocab,
+                        [""] + ["[OOV]"] * 10 + ["wind", "and", "fire"])
+
   def test_vocab_with_max_cap(self):
     vocab_data = ["", "[OOV]", "wind", "and", "fire"]
     layer = index_lookup.IndexLookup(