Sync KerasTuner from GitHub

Update of keras_tuner from https://github.com/keras-team/keras-tuner.

PiperOrigin-RevId: 480459137

Author: haifeng-jin

src/python/tensorflow_cloud/tuner/tests/unit/utils_test.py

@@ -46,19 +46,20 @@
     "metrics": [{"goal": "MAXIMIZE", "metric": "accuracy"}],
     "parameters": [
         {
-            "integer_value_spec": {"max_value": 4, "min_value": 1},
             "parameter": "units",
             "type": "INTEGER",
+            "integer_value_spec": {"min_value": 1, "max_value": 4},
+            "scale_type": "UNIT_LINEAR_SCALE",
         }
     ],
 }
 STUDY_CONFIG_INT_STEP = {
     "metrics": [{"goal": "MAXIMIZE", "metric": "accuracy"}],
     "parameters": [
         {
-            "discrete_value_spec": {"values": [32, 64, 96, 128]},
             "parameter": "units",
             "type": "DISCRETE",
+            "discrete_value_spec": {"values": [32, 64, 96, 128]},
         }
     ],
 }
@@ -69,6 +70,7 @@
             "double_value_spec": {"max_value": 0.5, "min_value": 0.1},
             "parameter": "learning_rate",
             "type": "DOUBLE",
+            "scale_type": "UNIT_LINEAR_SCALE",
         }
     ],
 }
@@ -119,14 +121,16 @@
     "metrics": [{"goal": "MAXIMIZE", "metric": "accuracy"}],
     "parameters": [
         {
-            "double_value_spec": {"max_value": 1.57, "min_value": 0.0},
             "parameter": "theta",
             "type": "DOUBLE",
+            "double_value_spec": {"min_value": 0.0, "max_value": 1.57},
+            "scale_type": "UNIT_LINEAR_SCALE",
         },
         {
-            "double_value_spec": {"max_value": 1.0, "min_value": 0.0},
             "parameter": "r",
             "type": "DOUBLE",
+            "double_value_spec": {"min_value": 0.0, "max_value": 1.0},
+            "scale_type": "UNIT_LINEAR_SCALE",
         },
     ],
 }
@@ -245,8 +249,8 @@ def test_convert_study_config_int(self, min_value, max_value, step,
         self._assert_hps_equal(actual_hps, hps)
 
     @parameterized.parameters(
-        (0.1, 0.5, None, None, STUDY_CONFIG_FLOAT),
-        (1, 2, 0.25, None, STUDY_CONFIG_FLOAT_STEP),
+        (0.1, 0.5, None, "linear", STUDY_CONFIG_FLOAT),
+        (1, 2, 0.25, "linear", STUDY_CONFIG_FLOAT_STEP),
         (0.1, 0.8, None, "linear", STUDY_CONFIG_FLOAT_LINEAR_SCALE),
         (1e-4, 1e-1, None, "log", STUDY_CONFIG_FLOAT_LOG_SCALE),
         (1e-4, 1e-1, None, "reverse_log", STUDY_CONFIG_FLOAT_REVERSE_LOG_SCALE))