Implement empty_like function in keras.ops (#21840)

shashaka · web-flow · commit bfde12b26714 · 2025-11-11T15:41:11.000-08:00
* Add empty_like function

* Update empty_like function for openvino

* correct function based on gemini advice

* Update the code based on review
diff --git a/keras/api/_tf_keras/keras/ops/__init__.py b/keras/api/_tf_keras/keras/ops/__init__.py
@@ -179,6 +179,7 @@
 from keras.src.ops.numpy import dot as dot
 from keras.src.ops.numpy import einsum as einsum
 from keras.src.ops.numpy import empty as empty
+from keras.src.ops.numpy import empty_like as empty_like
 from keras.src.ops.numpy import equal as equal
 from keras.src.ops.numpy import exp as exp
 from keras.src.ops.numpy import exp2 as exp2
diff --git a/keras/api/_tf_keras/keras/ops/numpy/__init__.py b/keras/api/_tf_keras/keras/ops/numpy/__init__.py
@@ -65,6 +65,7 @@
 from keras.src.ops.numpy import dot as dot
 from keras.src.ops.numpy import einsum as einsum
 from keras.src.ops.numpy import empty as empty
+from keras.src.ops.numpy import empty_like as empty_like
 from keras.src.ops.numpy import equal as equal
 from keras.src.ops.numpy import exp as exp
 from keras.src.ops.numpy import exp2 as exp2
diff --git a/keras/api/ops/__init__.py b/keras/api/ops/__init__.py
@@ -179,6 +179,7 @@
 from keras.src.ops.numpy import dot as dot
 from keras.src.ops.numpy import einsum as einsum
 from keras.src.ops.numpy import empty as empty
+from keras.src.ops.numpy import empty_like as empty_like
 from keras.src.ops.numpy import equal as equal
 from keras.src.ops.numpy import exp as exp
 from keras.src.ops.numpy import exp2 as exp2
diff --git a/keras/api/ops/numpy/__init__.py b/keras/api/ops/numpy/__init__.py
@@ -65,6 +65,7 @@
 from keras.src.ops.numpy import dot as dot
 from keras.src.ops.numpy import einsum as einsum
 from keras.src.ops.numpy import empty as empty
+from keras.src.ops.numpy import empty_like as empty_like
 from keras.src.ops.numpy import equal as equal
 from keras.src.ops.numpy import exp as exp
 from keras.src.ops.numpy import exp2 as exp2
diff --git a/keras/src/backend/jax/numpy.py b/keras/src/backend/jax/numpy.py
@@ -678,6 +678,10 @@ def empty(shape, dtype=None):
     return jnp.empty(shape, dtype=dtype)
 
 
+def empty_like(x, dtype=None):
+    return jnp.empty_like(x, dtype=dtype)
+
+
 def equal(x1, x2):
     x1 = convert_to_tensor(x1)
     x2 = convert_to_tensor(x2)
diff --git a/keras/src/backend/numpy/numpy.py b/keras/src/backend/numpy/numpy.py
@@ -612,6 +612,10 @@ def empty(shape, dtype=None):
     return np.empty(shape, dtype=dtype)
 
 
+def empty_like(x, dtype=None):
+    return np.empty_like(x, dtype=dtype)
+
+
 def equal(x1, x2):
     return np.equal(x1, x2)
 
diff --git a/keras/src/backend/openvino/excluded_concrete_tests.txt b/keras/src/backend/openvino/excluded_concrete_tests.txt
@@ -5,6 +5,7 @@ NumpyDtypeTest::test_argpartition
 NumpyDtypeTest::test_array
 NumpyDtypeTest::test_bartlett
 NumpyDtypeTest::test_blackman
+NumpyDtypeTest::test_empty_like
 NumpyDtypeTest::test_gcd
 NumpyDtypeTest::test_hamming
 NumpyDtypeTest::test_hanning
@@ -113,6 +114,7 @@ NumpyTwoInputOpsCorrectnessTest::test_vdot
 NumpyOneInputOpsDynamicShapeTest::test_angle
 NumpyOneInputOpsDynamicShapeTest::test_bartlett
 NumpyOneInputOpsDynamicShapeTest::test_blackman
+NumpyOneInputOpsDynamicShapeTest::test_empty_like
 NumpyOneInputOpsDynamicShapeTest::test_cbrt
 NumpyOneInputOpsDynamicShapeTest::test_corrcoef
 NumpyOneInputOpsDynamicShapeTest::test_hamming
diff --git a/keras/src/backend/openvino/numpy.py b/keras/src/backend/openvino/numpy.py
@@ -896,6 +896,12 @@ def empty(shape, dtype=None):
     return OpenVINOKerasTensor(empty_tensor)
 
 
+def empty_like(x, dtype=None):
+    raise NotImplementedError(
+        "`empty_like` is not supported with openvino backend"
+    )
+
+
 def equal(x1, x2):
     element_type = None
     if isinstance(x1, OpenVINOKerasTensor):
diff --git a/keras/src/backend/tensorflow/numpy.py b/keras/src/backend/tensorflow/numpy.py
@@ -1489,6 +1489,10 @@ def empty(shape, dtype=None):
     return tf.zeros(shape, dtype=dtype)
 
 
+def empty_like(x, dtype=None):
+    return tf.zeros_like(x, dtype=dtype)
+
+
 def equal(x1, x2):
     x1 = convert_to_tensor(x1)
     x2 = convert_to_tensor(x2)
diff --git a/keras/src/backend/torch/numpy.py b/keras/src/backend/torch/numpy.py
@@ -770,6 +770,12 @@ def empty(shape, dtype=None):
     return torch.empty(size=shape, dtype=dtype, device=get_device())
 
 
+def empty_like(x, dtype=None):
+    x = convert_to_tensor(x)
+    dtype = to_torch_dtype(dtype or x.dtype)
+    return torch.empty_like(x, dtype=dtype, device=get_device())
+
+
 def equal(x1, x2):
     x1, x2 = convert_to_tensor(x1), convert_to_tensor(x2)
     return torch.eq(x1, x2)
diff --git a/keras/src/ops/numpy.py b/keras/src/ops/numpy.py
@@ -3114,6 +3114,48 @@ def empty(shape, dtype=None):
     return backend.numpy.empty(shape, dtype=dtype)
 
 
+class EmptyLike(Operation):
+    def __init__(self, dtype=None, *, name=None):
+        super().__init__(name=name)
+        self.dtype = None if dtype is None else backend.standardize_dtype(dtype)
+
+    def call(self, x):
+        return backend.numpy.empty_like(x, dtype=self.dtype)
+
+    def compute_output_spec(self, x):
+        dtype = (
+            backend.standardize_dtype(x.dtype)
+            if self.dtype is None
+            else self.dtype
+        )
+        return KerasTensor(x.shape, dtype=dtype)
+
+
+@keras_export(["keras.ops.empty_like", "keras.ops.numpy.empty_like"])
+def empty_like(x, dtype=None):
+    """Return a new uninitialized tensor with the same shape and dtype as `x`.
+
+    Args:
+        x: Input tensor to mimic shape and dtype.
+        dtype: Optional data type. If None, uses `x.dtype`.
+
+    Returns:
+        A tensor with the same shape and dtype as `x`, with arbitrary contents.
+
+    Example:
+    >>> from keras import ops
+    >>> x = ops.ones((2, 3), dtype="float32")
+    >>> y = ops.empty_like(x)
+    >>> y.shape
+    (2, 3)
+    >>> y.dtype
+    dtype('float32')
+    """
+    if any_symbolic_tensors((x,)):
+        return EmptyLike(dtype=dtype).symbolic_call(x)
+    return backend.numpy.empty_like(x, dtype=dtype)
+
+
 class Equal(Operation):
     def call(self, x1, x2):
         return backend.numpy.equal(x1, x2)
diff --git a/keras/src/ops/numpy_test.py b/keras/src/ops/numpy_test.py
@@ -1495,6 +1495,11 @@ def test_dot(self):
         y = KerasTensor((5,))
         self.assertEqual(knp.dot(x, y).shape, ())
 
+    def test_empty_like(self):
+        x = KerasTensor((None, 3))
+        self.assertEqual(knp.empty_like(x).shape, (None, 3))
+        self.assertEqual(knp.empty_like(x).dtype, x.dtype)
+
     def test_exp(self):
         x = KerasTensor((None, 3))
         self.assertEqual(knp.exp(x).shape, (None, 3))
@@ -2139,6 +2144,11 @@ def test_dot(self):
             y = KerasTensor((2, 3))
             knp.dot(x, y)
 
+    def test_empty_like(self):
+        x = KerasTensor((2, 3))
+        self.assertEqual(knp.empty_like(x).shape, (2, 3))
+        self.assertEqual(knp.empty_like(x).dtype, x.dtype)
+
     def test_exp(self):
         x = KerasTensor((2, 3))
         self.assertEqual(knp.exp(x).shape, (2, 3))
@@ -7328,6 +7338,22 @@ def test_empty(self, dtype):
             expected_dtype,
         )
 
+    @parameterized.named_parameters(named_product(dtype=ALL_DTYPES))
+    def test_empty_like(self, dtype):
+        import jax.numpy as jnp
+
+        x = jnp.empty([2, 3, 4], dtype=dtype)
+        expected_dtype = standardize_dtype(jnp.empty_like(x, dtype=dtype).dtype)
+
+        self.assertEqual(
+            standardize_dtype(knp.empty_like(x, dtype=dtype).dtype),
+            expected_dtype,
+        )
+        self.assertEqual(
+            standardize_dtype(knp.EmptyLike().symbolic_call(x).dtype),
+            expected_dtype,
+        )
+
     @parameterized.named_parameters(
         named_product(dtypes=itertools.combinations(ALL_DTYPES, 2))
     )
