Increase test coverage in saving

keras_core/legacy/saving/legacy_h5_format.py

@@ -37,10 +37,10 @@ def save_model_to_hdf5(model, filepath, overwrite=True, include_optimizer=True):
             if not proceed:
                 return
 
-        # Try creating dir if not exist
         dirpath = os.path.dirname(filepath)
-        if not os.path.exists(dirpath):
-            os.path.makedirs(dirpath)
+        if dirpath and not os.path.exists(dirpath):
+            # os.path.makedirs(dirpath)
+            os.makedirs(dirpath)
 
         f = h5py.File(filepath, mode="w")
         opened_new_file = True

keras_core/saving/saving_api_test.py

@@ -0,0 +1,139 @@
+import os
+import unittest
+
+import numpy as np
+
+from keras_core import layers
+from keras_core.models import Sequential
+from keras_core.saving import saving_api
+
+
+class SaveModelTests(unittest.TestCase):
+    def setUp(self):
+        self.model = Sequential(
+            [
+                layers.Dense(5, input_shape=(3,)),
+                layers.Softmax(),
+            ],
+        )
+        self.filepath = "test_model.keras"
+        saving_api.save_model(self.model, self.filepath)
+
+    def test_basic_saving(self):
+        loaded_model = saving_api.load_model(self.filepath)
+        x = np.random.uniform(size=(10, 3))
+        self.assertTrue(
+            np.allclose(self.model.predict(x), loaded_model.predict(x))
+        )
+
+    def test_invalid_save_format(self):
+        with self.assertRaisesRegex(
+            ValueError, "The `save_format` argument is deprecated"
+        ):
+            saving_api.save_model(self.model, "model.txt", save_format=True)
+
+    def test_overwrite_prompt(self):
+        original_mtime = os.path.getmtime(self.filepath)
+        saving_api.io_utils.ask_to_proceed_with_overwrite = lambda x: False
+        saving_api.save_model(self.model, self.filepath, overwrite=False)
+        new_mtime = os.path.getmtime(self.filepath)
+        self.assertEqual(original_mtime, new_mtime)
+
+    def test_unsupported_arguments(self):
+        with self.assertRaises(ValueError):
+            saving_api.save_model(self.model, self.filepath, random_arg=True)
+
+    def test_save_h5_format(self):
+        filepath_h5 = "test_model.h5"
+        saving_api.save_model(self.model, filepath_h5)
+        self.assertTrue(os.path.exists(filepath_h5))
+        os.remove(filepath_h5)  # Cleanup
+
+    def test_save_unsupported_extension(self):
+        with self.assertRaises(ValueError):
+            saving_api.save_model(self.model, "model.png")
+
+    def tearDown(self):
+        if os.path.exists(self.filepath):
+            os.remove(self.filepath)
+
+
+class LoadModelTests(unittest.TestCase):
+    def setUp(self):
+        self.model = Sequential(
+            [
+                layers.Dense(5, input_shape=(3,)),
+                layers.Softmax(),
+            ],
+        )
+        self.filepath = "test_model.keras"
+        saving_api.save_model(self.model, self.filepath)
+
+    def test_basic_load(self):
+        loaded_model = saving_api.load_model(self.filepath)
+        x = np.random.uniform(size=(10, 3))
+        self.assertTrue(
+            np.allclose(self.model.predict(x), loaded_model.predict(x))
+        )
+
+    def test_load_unsupported_format(self):
+        with self.assertRaises(ValueError):
+            saving_api.load_model("model.pkl")
+
+    def test_load_keras_not_zip(self):
+        with self.assertRaises(ValueError):
+            saving_api.load_model("not_a_zip.keras")
+
+    def test_load_h5_format(self):
+        filepath_h5 = "test_model.h5"
+        saving_api.save_model(self.model, filepath_h5)
+        loaded_model = saving_api.load_model(filepath_h5)
+        x = np.random.uniform(size=(10, 3))
+        self.assertTrue(
+            np.allclose(self.model.predict(x), loaded_model.predict(x))
+        )
+        os.remove(filepath_h5)  # Cleanup
+
+    def tearDown(self):
+        if os.path.exists(self.filepath):
+            os.remove(self.filepath)
+
+
+class LoadWeightsTests(unittest.TestCase):
+    def setUp(self):
+        self.model = Sequential(
+            [
+                layers.Dense(5, input_shape=(3,)),
+                layers.Softmax(),
+            ],
+        )
+
+    def test_load_keras_weights(self):
+        filepath = "test_weights.weights.h5"
+        self.model.save_weights(filepath)
+        original_weights = self.model.get_weights()
+        self.model.load_weights(filepath)
+        loaded_weights = self.model.get_weights()
+        for orig, loaded in zip(original_weights, loaded_weights):
+            self.assertTrue(np.array_equal(orig, loaded))
+
+    def test_load_unsupported_format(self):
+        with self.assertRaises(ValueError):
+            self.model.load_weights("weights.pkl")
+
+    def test_load_keras_format_weights(self):
+        filepath_keras = "test_weights.weights.h5"
+        self.model.save_weights(filepath_keras)
+        self.model.load_weights(filepath_keras)
+        os.remove(filepath_keras)  # Cleanup
+
+    def test_load_h5_format_weights(self):
+        filepath_h5 = "test_weights.weights.h5"
+        self.model.save_weights(filepath_h5)
+        self.model.load_weights(filepath_h5)
+        os.remove(filepath_h5)  # Cleanup
+
+    def tearDown(self):
+        filepath = "test_weights.weights.h5"
+        if os.path.exists(filepath):
+            os.remove(filepath)