Fix users of NumPy APIs that are removed in NumPy 2.0.

This change migrates users of APIs removed in NumPy 2.0 to their recommended replacements (https://numpy.org/devdocs/numpy_2_0_migration_guide.html).

PiperOrigin-RevId: 655943417

Author: tfx-copybara

tensorflow_data_validation/api/stats_api_test.py

@@ -51,7 +51,7 @@ def test_stats_pipeline(self):
             pa.array([np.linspace(1, 500, 500, dtype=np.int32)]),
         ], ['a', 'b', 'c']),
         pa.RecordBatch.from_arrays([
-            pa.array([[3.0, 4.0, np.NaN, 5.0]]),
+            pa.array([[3.0, 4.0, np.nan, 5.0]]),
             pa.array([['a', 'c', '∞', 'a']]),
             pa.array([np.linspace(501, 1250, 750, dtype=np.int32)]),
         ], ['a', 'b', 'c']),

tensorflow_data_validation/statistics/generators/basic_stats_generator_test.py

@@ -826,7 +826,7 @@ def test_with_weight_feature(self):
         pa.array([[1.0], [2.0]])
     ], ['a', 'b', 'w'])
     b2 = pa.RecordBatch.from_arrays([
-        pa.array([[1.0, np.NaN, np.NaN, np.NaN], None]),
+        pa.array([[1.0, np.nan, np.nan, np.nan], None]),
         pa.array([[1], None]),
         pa.array([[3.0], [2.0]])
     ], ['a', 'b', 'w'])
@@ -1269,7 +1269,7 @@ def test_with_per_feature_weight(self):
         pa.array([[2.0], [1.0]]),
     ], ['a', 'b', 'w_a', 'w_b'])
     b2 = pa.RecordBatch.from_arrays([
-        pa.array([[1.0, np.NaN, np.NaN, np.NaN], None]),
+        pa.array([[1.0, np.nan, np.nan, np.nan], None]),
         pa.array([[1], None]),
         pa.array([[3.0], [2.0]]),
         pa.array([[2.0], [3.0]]),
@@ -1706,8 +1706,8 @@ def test_with_individual_feature_value_missing(self):
     # input with two batches: first batch has two examples and second batch
     # has a single example.
     b1 = pa.RecordBatch.from_arrays(
-        [pa.array([[1.0, 2.0], [3.0, 4.0, np.NaN, 5.0]])], ['a'])
-    b2 = pa.RecordBatch.from_arrays([pa.array([[np.NaN, 1.0]])], ['a'])
+        [pa.array([[1.0, 2.0], [3.0, 4.0, np.nan, 5.0]])], ['a'])
+    b2 = pa.RecordBatch.from_arrays([pa.array([[np.nan, 1.0]])], ['a'])
     batches = [b1, b2]
 
     expected_result = {
@@ -2286,7 +2286,7 @@ def test_no_value_in_batch(self):
 
   def test_only_nan(self):
     b1 = pa.RecordBatch.from_arrays(
-        [pa.array([[np.NaN]], type=pa.list_(pa.float32()))], ['a'])
+        [pa.array([[np.nan]], type=pa.list_(pa.float32()))], ['a'])
     batches = [b1]
     expected_result = {
         types.FeaturePath(['a']): text_format.Parse(

tensorflow_data_validation/statistics/generators/mutual_information_test.py

@@ -81,7 +81,7 @@ def test_encoder_feature_excluded(self):
 
   def test_encoder_multivalent_numerical_with_nulls(self):
     batch = pa.RecordBatch.from_arrays(
-        [pa.array([[1.0, 1.0, np.NaN], None, [2.0, 2.0, 1.0], []])], ["fa"])
+        [pa.array([[1.0, 1.0, np.nan], None, [2.0, 2.0, 1.0], []])], ["fa"])
     expected = {
         types.FeaturePath(["fa"]): [[2, 0, 0], [None, None, None], [1, 0, 2],
                                     [None, None, None]]
@@ -92,7 +92,7 @@ def test_encoder_multivalent_numerical_with_nulls(self):
 
   def test_encoder_univalent_with_nulls(self):
     batch = pa.RecordBatch.from_arrays(
-        [pa.array([None, [2.0], [], [None], [np.NaN]])], ["fa"])
+        [pa.array([None, [2.0], [], [None], [np.nan]])], ["fa"])
     expected = {
         types.FeaturePath(["fa"]): [[None], [2], [None], [None], [None]]
     }
@@ -205,7 +205,7 @@ def test_mi_with_univalent_features(self):
     label_array = pa.array([[0.1], [0.2], [0.7], [0.2], None, [0.9], [0.4],
                             [0.8]])
     # Random floats that do not map onto the label
-    terrible_feat_array = pa.array([[0.4], [0.1], [0.4], [np.NaN], [0.8], [0.2],
+    terrible_feat_array = pa.array([[0.4], [0.1], [0.4], [np.nan], [0.8], [0.2],
                                     [0.5], [0.1]])
     batch = pa.RecordBatch.from_arrays(
         [label_array, label_array, terrible_feat_array],
@@ -318,7 +318,7 @@ def test_mi_batch_smaller_than_k(self):
   def test_mi_normalized(self):
     label_array = pa.array([[0.1], [0.2], [0.7], [0.2], None, [0.9], [0.4],
                             [0.8]])
-    terrible_feat_array = pa.array([[0.4], [0.1], [0.4], [np.NaN], [0.8], [0.2],
+    terrible_feat_array = pa.array([[0.4], [0.1], [0.4], [np.nan], [0.8], [0.2],
                                     [0.5], [0.1]])
     batch = pa.RecordBatch.from_arrays(
         [label_array, label_array, terrible_feat_array],
@@ -468,7 +468,7 @@ def test_mi_with_unicode_labels(self):
 
   def test_mi_with_univalent_feature_all_null(self):
     label_array = pa.array([[0.1], [0.2], [0.7], [0.7]])
-    null_feat_array = pa.array([[np.NaN], [np.NaN], [np.NaN], [np.NaN]])
+    null_feat_array = pa.array([[np.nan], [np.nan], [np.nan], [np.nan]])
     batch = pa.RecordBatch.from_arrays([label_array, null_feat_array],
                                        ["label_key", "null_feature"])
 
@@ -510,7 +510,7 @@ def test_mi_with_univalent_feature_all_null(self):
 
   def test_mi_with_multivalent_feature_all_null(self):
     label_array = pa.array([[0.1], [0.2], [0.7], [0.7]])
-    null_feat_array = pa.array([[np.NaN], [np.NaN], [np.NaN], [np.NaN]])
+    null_feat_array = pa.array([[np.nan], [np.nan], [np.nan], [np.nan]])
     batch = pa.RecordBatch.from_arrays([label_array, null_feat_array],
                                        ["label_key", "null_feature"])
 
@@ -551,7 +551,7 @@ def test_mi_with_multivalent_feature_all_null(self):
 
   def test_mi_with_multivalent_feature_all_empty(self):
     label_array = pa.array([[0.1], [0.2], [0.7], [0.7]])
-    empty_feat_array = pa.array([[np.NaN], [], [], []])
+    empty_feat_array = pa.array([[np.nan], [], [], []])
     batch = pa.RecordBatch.from_arrays([label_array, empty_feat_array],
                                        ["label_key", "empty_feature"])
 
@@ -633,7 +633,7 @@ def test_mi_with_multivalent_feature_univalent_label(self):
                                   types.FeaturePath(["label_key"]))
 
   def test_mi_with_multivalent_numeric_feature(self):
-    feat_array = pa.array([[3.1], None, [4.0], [np.NaN], [1.2, 8.5], [2.3],
+    feat_array = pa.array([[3.1], None, [4.0], [np.nan], [1.2, 8.5], [2.3],
                            [1.2, 3.2, 3.9]])
     label_array = pa.array([[3.3], None, [4.0], [2.0, 8.0], [1.3, 8.5], [2.3],
                             [1.0, 3.1, 4]])
@@ -1395,7 +1395,7 @@ def setUp(self):
         ], ["fa", "fb", "fc", "fd", "label_key"]),
         pa.RecordBatch.from_arrays([
             pa.array([["1"]]),
-            pa.array([[np.NaN]], type=pa.list_(pa.float64())),
+            pa.array([[np.nan]], type=pa.list_(pa.float64())),
             pa.array([["0", "0"]]),
             pa.array([[0.0, 0.2]]),
             pa.array([["label"]]),

tensorflow_data_validation/statistics/generators/partitioned_stats_generator_test.py

@@ -567,7 +567,7 @@ def setUp(self):
             pa.array([['Green']]),
             pa.array([[2.3]]),
             pa.array([[0]]),
-            pa.array([[np.NaN]], type=pa.list_(pa.float64())),
+            pa.array([[np.nan]], type=pa.list_(pa.float64())),
             pa.array([['Label']]),
         ], ['fa', 'fb', 'fc', 'fd', 'label_key']),
         pa.RecordBatch.from_arrays([

tensorflow_data_validation/statistics/generators/sklearn_mutual_information_test.py

@@ -612,7 +612,7 @@ def test_mi_with_imputed_numerical_feature(self):
         [0.1], [0.2], [0.8], [0.7], [0.2], [0.2], [0.3],
         [0.1], [0.2], [0.8], [0.7], [0.2], [0.2], [0.3]])
     feat_array = pa.array([
-        [0.1], [0.2], [0.8], [0.7], [0.2], [np.NaN], None,
+        [0.1], [0.2], [0.8], [0.7], [0.2], [np.nan], None,
         [0.1], [0.2], [0.8], [0.7], [0.2], [0.2], [0.3]])
     batch = pa.RecordBatch.from_arrays([label_array, feat_array],
                                        ["label_key", "fa"])
@@ -716,7 +716,7 @@ def test_mi_with_imputed_categorical_label(self):
 
   def test_mi_with_imputed_numerical_label(self):
     label_array = pa.array([
-        [0.1], [0.2], [0.8], [0.7], [0.2], [np.NaN], None,
+        [0.1], [0.2], [0.8], [0.7], [0.2], [np.nan], None,
         [0.1], [0.2], [0.8], [0.7], [0.2], [0.2], [0.3]])
     feat_array = pa.array([
         [0.1], [0.2], [0.8], [0.7], [0.2], [0.2], [0.3],

tensorflow_data_validation/statistics/stats_impl_test.py

@@ -117,7 +117,7 @@ def extract_output(self, accumulator):
                 pa.array([np.linspace(1, 500, 500, dtype=np.int64)]),
             ], ['a', 'b', 'c']),
             pa.RecordBatch.from_arrays([
-                pa.array([[3.0, 4.0, np.NaN, 5.0]], type=pa.list_(
+                pa.array([[3.0, 4.0, np.nan, 5.0]], type=pa.list_(
                     pa.float32())),
                 pa.array([[b'a', b'c', b'd', b'a']], type=pa.list_(
                     pa.binary())),
@@ -1891,7 +1891,7 @@ def extract_output(self, accumulator):
         pa.array([np.linspace(1, 500, 500, dtype=np.int64)]),
     ], ['a', 'b', 'c']),
     pa.RecordBatch.from_arrays([
-        pa.array([[3.0, 4.0, np.NaN, 5.0]], type=pa.list_(pa.float32())),
+        pa.array([[3.0, 4.0, np.nan, 5.0]], type=pa.list_(pa.float32())),
         pa.array([[b'a', b'b']], type=pa.list_(pa.binary())),
         pa.array([np.linspace(501, 1250, 750, dtype=np.int64)]),
     ], ['a', 'b', 'c']),
@@ -2015,7 +2015,7 @@ def extract_output(self, accumulator):
                 pa.array([np.linspace(1, 500, 500, dtype=np.int64)]),
             ], ['a', 'b', 'c']),
             pa.RecordBatch.from_arrays([
-                pa.array([[3.0, 4.0, np.NaN, 5.0]], type=pa.list_(
+                pa.array([[3.0, 4.0, np.nan, 5.0]], type=pa.list_(
                     pa.float32())),
                 pa.array([[b'a', b'b']], type=pa.list_(pa.binary())),
                 pa.array([np.linspace(501, 1250, 750, dtype=np.int64)]),
@@ -2114,7 +2114,7 @@ def test_stats_impl_slicing_sql(self):
             pa.array([np.linspace(1, 500, 500, dtype=np.int64)]),
         ], ['a', 'b', 'c']),
         pa.RecordBatch.from_arrays([
-            pa.array([[3.0, 4.0, np.NaN, 5.0]], type=pa.list_(
+            pa.array([[3.0, 4.0, np.nan, 5.0]], type=pa.list_(
                 pa.float32())),
             pa.array([[b'a', b'b']], type=pa.list_(pa.binary())),
             pa.array([np.linspace(501, 1250, 750, dtype=np.int64)]),
@@ -2160,7 +2160,7 @@ def test_stats_impl_slicing_sql_in_config(self):
             pa.array([np.linspace(1, 500, 500, dtype=np.int64)]),
         ], ['a', 'b', 'c']),
         pa.RecordBatch.from_arrays([
-            pa.array([[3.0, 4.0, np.NaN, 5.0]], type=pa.list_(
+            pa.array([[3.0, 4.0, np.nan, 5.0]], type=pa.list_(
                 pa.float32())),
             pa.array([[b'a', b'b']], type=pa.list_(pa.binary())),
             pa.array([np.linspace(501, 1250, 750, dtype=np.int64)]),
@@ -2677,7 +2677,7 @@ def test_tfdv_telemetry(self):
             pa.array([None]),
         ], ['a', 'b', 'c']),
         pa.RecordBatch.from_arrays([
-            pa.array([[3.0, 4.0, np.NaN, 5.0]]),
+            pa.array([[3.0, 4.0, np.nan, 5.0]]),
             pa.array([['d', 'e', 'f']]),
             pa.array([None]),
         ], ['a', 'b', 'c']),