Add model test with explicit call() (#151)

* Test a model with an explicit call to call().

* Add TODO.

* Make it look like the example from TF docs.

* More tests.

Author: khatchad

edu.cuny.hunter.hybridize.tests/resources/HybridizeFunction/testModel/in/A.py

@@ -16,7 +16,7 @@ def __init__(self, **kwargs):
     self.dropout = tf.keras.layers.Dropout(0.2)
     self.dense_2 = tf.keras.layers.Dense(10)
 
-  def call(self, x):
+  def __call__(self, x):
     x = self.flatten(x)
 
     for layer in self.my_layers:

edu.cuny.hunter.hybridize.tests/resources/HybridizeFunction/testModel2/in/A.py

@@ -0,0 +1,41 @@
+import tensorflow as tf
+
+# Create an override model to classify pictures
+
+class SequentialModel(tf.keras.Model):
+  def __init__(self, **kwargs):
+    super(SequentialModel, self).__init__(**kwargs)
+
+    self.flatten = tf.keras.layers.Flatten(input_shape=(28, 28))
+
+    # Add a lot of small layers
+    num_layers = 100
+    self.my_layers = [tf.keras.layers.Dense(64, activation="relu")
+                      for n in range(num_layers)]
+
+    self.dropout = tf.keras.layers.Dropout(0.2)
+    self.dense_2 = tf.keras.layers.Dense(10)
+
+  def call(self, x):
+    x = self.flatten(x)
+
+    for layer in self.my_layers:
+      x = layer(x)
+
+    x = self.dropout(x)
+    x = self.dense_2(x)
+
+    return x
+
+if __name__ == '__main__':
+    input_data = tf.random.uniform([20, 28, 28])
+    print("Input:")
+    print(type(input_data))
+    print(input_data)
+
+    model = SequentialModel()
+    result = model(input_data)
+
+    print("Output:")
+    print(type(input_data))
+    print(result)

edu.cuny.hunter.hybridize.tests/resources/HybridizeFunction/testModel2/in/requirements.txt

@@ -0,0 +1 @@
+tensorflow==2.9.3

edu.cuny.hunter.hybridize.tests/resources/HybridizeFunction/testModel3/in/A.py

@@ -0,0 +1,41 @@
+import tensorflow as tf
+
+# Create an override model to classify pictures
+
+class SequentialModel(tf.keras.Model):
+  def __init__(self, **kwargs):
+    super(SequentialModel, self).__init__(**kwargs)
+
+    self.flatten = tf.keras.layers.Flatten(input_shape=(28, 28))
+
+    # Add a lot of small layers
+    num_layers = 100
+    self.my_layers = [tf.keras.layers.Dense(64, activation="relu")
+                      for n in range(num_layers)]
+
+    self.dropout = tf.keras.layers.Dropout(0.2)
+    self.dense_2 = tf.keras.layers.Dense(10)
+
+  def call(self, x):
+    x = self.flatten(x)
+
+    for layer in self.my_layers:
+      x = layer(x)
+
+    x = self.dropout(x)
+    x = self.dense_2(x)
+
+    return x
+
+if __name__ == '__main__':
+    input_data = tf.random.uniform([20, 28, 28])
+    print("Input:")
+    print(type(input_data))
+    print(input_data)
+
+    model = SequentialModel()
+    result = model.call(input_data)
+
+    print("Output:")
+    print(type(input_data))
+    print(result)

edu.cuny.hunter.hybridize.tests/resources/HybridizeFunction/testModel3/in/requirements.txt

@@ -0,0 +1 @@
+tensorflow==2.9.3

edu.cuny.hunter.hybridize.tests/resources/HybridizeFunction/testModel4/in/A.py

@@ -0,0 +1,41 @@
+import tensorflow as tf
+
+# Create an override model to classify pictures
+
+class SequentialModel(tf.keras.Model):
+  def __init__(self, **kwargs):
+    super(SequentialModel, self).__init__(**kwargs)
+
+    self.flatten = tf.keras.layers.Flatten(input_shape=(28, 28))
+
+    # Add a lot of small layers
+    num_layers = 100
+    self.my_layers = [tf.keras.layers.Dense(64, activation="relu")
+                      for n in range(num_layers)]
+
+    self.dropout = tf.keras.layers.Dropout(0.2)
+    self.dense_2 = tf.keras.layers.Dense(10)
+
+  def __call__(self, x):
+    x = self.flatten(x)
+
+    for layer in self.my_layers:
+      x = layer(x)
+
+    x = self.dropout(x)
+    x = self.dense_2(x)
+
+    return x
+
+if __name__ == '__main__':
+    input_data = tf.random.uniform([20, 28, 28])
+    print("Input:")
+    print(type(input_data))
+    print(input_data)
+
+    model = SequentialModel()
+    result = model.__call__(input_data)
+
+    print("Output:")
+    print(type(input_data))
+    print(result)

edu.cuny.hunter.hybridize.tests/resources/HybridizeFunction/testModel4/in/requirements.txt

@@ -0,0 +1 @@
+tensorflow==2.9.3

edu.cuny.hunter.hybridize.tests/test cases/edu/cuny/hunter/hybridize/tests/HybridizeFunctionRefactoringTest.java

@@ -1608,4 +1608,49 @@ public void testModel() throws Exception {
 		// no hybrids.
 		assertTrue(functions.stream().map(Function::isHybrid).allMatch(b -> b == false));
 	}
+
+	/**
+	 * Test a model. No tf.function in this one. Use call instead of __call__. Ariadne doesn't support __call__.
+	 * See https://github.com/wala/ML/issues/24.
+	 */
+	@Test
+	public void testModel2() throws Exception {
+		Set<Function> functions = this.getFunctions();
+		assertNotNull(functions);
+
+		LOG.info("Found functions: " + functions.size());
+
+		// no hybrids.
+		assertTrue(functions.stream().map(Function::isHybrid).allMatch(b -> b == false));
+	}
+
+	/**
+	 * Test a model. No tf.function in this one. Explicit call method.
+	 */
+	@Test
+	public void testModel3() throws Exception {
+		Set<Function> functions = this.getFunctions();
+		assertNotNull(functions);
+
+		LOG.info("Found functions: " + functions.size());
+
+		// no hybrids.
+		assertTrue(functions.stream().map(Function::isHybrid).allMatch(b -> b == false));
+	}
+
+	/**
+	 * Test a model. No tf.function in this one. Explicit call method.
+	 */
+	@Test
+	public void testModel4() throws Exception {
+		Set<Function> functions = this.getFunctions();
+		assertNotNull(functions);
+
+		LOG.info("Found functions: " + functions.size());
+
+		// no hybrids.
+		assertTrue(functions.stream().map(Function::isHybrid).allMatch(b -> b == false));
+	}
+
+	// TODO: Test models that have tf.functions.
 }