Add docs on implementing PyTorch Ops (and CumOp) (#837)

doc/conf.py

@@ -32,8 +32,16 @@ extensions = [
     "sphinx.ext.napoleon",
     "sphinx.ext.linkcode",
     "sphinx.ext.mathjax",
+    "sphinx_design",
+    "sphinx.ext.intersphinx"
 ]
 
+intersphinx_mapping = {
+    "jax": ("https://jax.readthedocs.io/en/latest", None),
+    "numpy": ("https://numpy.org/doc/stable", None),
+    "torch": ("https://pytorch.org/docs/stable", None),
+}
+
 needs_sphinx = "3"
 
 todo_include_todos = True

doc/environment.yml

@@ -13,6 +13,7 @@ dependencies:
   - mock
   - pillow
   - pymc-sphinx-theme
+  - sphinx-design
   - pip
   - pip:
     - -e ..

doc/extending/creating_a_numba_jax_op.rst

-Adding JAX and Numba support for `Op`\s
+Adding JAX, Numba and Pytorch support for `Op`\s
 =======================================
 
-PyTensor is able to convert its graphs into JAX and Numba compiled functions. In order to do
-this, each :class:`Op` in an PyTensor graph must have an equivalent JAX/Numba implementation function.
+PyTensor is able to convert its graphs into JAX, Numba and Pytorch compiled functions. In order to do
+this, each :class:`Op` in an PyTensor graph must have an equivalent JAX/Numba/Pytorch implementation function.
 
-This tutorial will explain how JAX and Numba implementations are created for an :class:`Op`.  It will
-focus specifically on the JAX case, but the same mechanisms are used for Numba as well.
+This tutorial will explain how JAX, Numba and Pytorch implementations are created for an :class:`Op`. 
 
-Step 1: Identify the PyTensor :class:`Op` you'd like to implement in JAX
+Step 1: Identify the PyTensor :class:`Op` you'd like to implement
 ------------------------------------------------------------------------
 
-Find the source for the PyTensor :class:`Op` you'd like to be supported in JAX, and
+Find the source for the PyTensor :class:`Op` you'd like to be supported and
 identify the function signature and return values. These can be determined by
 looking at the :meth:`Op.make_node` implementation. In general, one needs to be familiar
 with PyTensor :class:`Op`\s in order to provide a conversion implementation, so first read
@@ -46,7 +45,7 @@ which currently has an :meth:`Op.make_node` as follows:
         return Apply(self, [x], [out_type])
 
 The :class:`Apply` instance that's returned specifies the exact types of inputs that
-our JAX implementation will receive and the exact types of outputs it's expected to
+our implementation will receive and the exact types of outputs it's expected to
 return--both in terms of their data types and number of dimensions/shapes.
 The actual inputs our implementation will receive are necessarily numeric values
 or NumPy :class:`ndarray`\s; all that :meth:`Op.make_node` tells us is the
@@ -57,7 +56,7 @@ automatically converted to PyTensor variables via :func:`as_tensor_variable`.
 There is another parameter, `axis`, that is used to determine the direction
 of the operation, hence shape of the output. The check that follows imply that
 `axis` must refer to a dimension in the input tensor. The input's elements
-could also have any data type (e.g. floats, ints), so our JAX implementation
+could also have any data type (e.g. floats, ints), so our implementation
 must be able to handle all the possible data types.
 
 It also tells us that there's only one return value, that it has a data type
@@ -89,42 +88,175 @@ as :class:`CumsumOp`\ :class:`Op`. The difference lies in that the `mode` attrib
         c_axis = property(lambda self: np.MAXDIMS if self.axis is None else self.axis)
 
 `__props__` is used to parametrize the general behavior of the :class:`Op`. One need to
-pay attention to this to decide whether the JAX implementation should support all variants
+pay attention to this to decide whether the implementation should support all variants
 or raise an explicit NotImplementedError for cases that are not supported e.g., when
 :class:`CumsumOp` of :class:`CumOp("add")` is supported but not :class:`CumprodOp` of
 :class:`CumOp("mul")`.
 
 Next, we look at the :meth:`Op.perform` implementation to see exactly
 how the inputs and outputs are used to compute the outputs for an :class:`Op`
-in Python. This method is effectively what needs to be implemented in JAX.
+in Python. This method is effectively what needs to be implemented.
 
-Step 2: Find the relevant JAX method (or something close)
+Step 2: Find the relevant method in JAX/Numba/Pytorch (or something close)
 ---------------------------------------------------------
 
 With a precise idea of what the PyTensor :class:`Op` does we need to figure out how
-to implement it in JAX. In the best case scenario, JAX has a similarly named
+to implement it in JAX, Numba or Pytorch. In the best case scenario, there is a similarly named
 function that performs exactly the same computations as the :class:`Op`. For
 example, the :class:`Eye` operator has a JAX equivalent: :func:`jax.numpy.eye`
-(see `the documentation <https://jax.readthedocs.io/en/latest/_autosummary/jax.numpy.eye.html?highlight=eye>`_).
+and a Pytorch equivalent: :func:`torch.eye`.
 
-If we wanted to implement an :class:`Op` like :class:`IfElse`, we might need to
+If we wanted to implement an :class:`Op` like :class:`DimShuffle`, we might need to
 recreate the functionality with some custom logic.  In many cases, at least some
 custom logic is needed to reformat the inputs and outputs so that they exactly
 match the `Op`'s.
 
-Here's an example for :class:`IfElse`:
+Here's an example for :class:`DimShuffle`:
 
-.. code:: python
 
-   def ifelse(cond, *args, n_outs=n_outs):
-       res = jax.lax.cond(
-           cond, lambda _: args[:n_outs], lambda _: args[n_outs:], operand=None
-       )
-       return res if n_outs > 1 else res[0]
+.. tab-set::
+
+        .. tab-item:: JAX     
+
+            .. code:: python
+
+                def dimshuffle(x, op):
+                    res = jnp.transpose(x, op.transposition)
+
+                    shape = list(res.shape[: len(op.shuffle)])
+
+                    for augm in op.augment:
+                        shape.insert(augm, 1)
+
+                    res = jnp.reshape(res, shape)
+
+                    if not op.inplace:
+                        res = jnp.copy(res)
+
+                    return res
+        
+        .. tab-item:: Numba
+
+            .. code:: python
+
+                def numba_funcify_DimShuffle(op, node, **kwargs):
+                    shuffle = tuple(op.shuffle)
+                    transposition = tuple(op.transposition)
+                    augment = tuple(op.augment)
+                    inplace = op.inplace
+
+                    ndim_new_shape = len(shuffle) + len(augment)
+
+                    no_transpose = all(i == j for i, j in enumerate(transposition))
+                    if no_transpose:
+
+                        @numba_basic.numba_njit
+                        def transpose(x):
+                            return x
+
+                    else:
+
+                        @numba_basic.numba_njit
+                        def transpose(x):
+                            return np.transpose(x, transposition)
+
+                    shape_template = (1,) * ndim_new_shape
+
+                    # When `len(shuffle) == 0`, the `shuffle_shape[j]` expression below
+                    # is typed as `getitem(Tuple(), int)`, which has no implementation
+                    # (since getting an item from an empty sequence doesn't make sense).
+                    # To avoid this compile-time error, we omit the expression altogether.
+                    if len(shuffle) > 0:
+                        # Use the statically known shape if available
+                        if all(length is not None for length in node.outputs[0].type.shape):
+                            shape = node.outputs[0].type.shape
+
+                            @numba_basic.numba_njit
+                            def find_shape(array_shape):
+                                return shape
+
+                        else:
+
+                            @numba_basic.numba_njit
+                            def find_shape(array_shape):
+                                shape = shape_template
+                                j = 0
+                                for i in range(ndim_new_shape):
+                                    if i not in augment:
+                                        length = array_shape[j]
+                                        shape = numba_basic.tuple_setitem(shape, i, length)
+                                        j = j + 1
+                                return shape
+
+                    else:
+
+                        @numba_basic.numba_njit
+                        def find_shape(array_shape):
+                            return shape_template
+
+                    if ndim_new_shape > 0:
+
+                        @numba_basic.numba_njit
+                        def dimshuffle_inner(x, shuffle):
+                            x = transpose(x)
+                            shuffle_shape = x.shape[: len(shuffle)]
+                            new_shape = find_shape(shuffle_shape)
+
+                            # FIXME: Numba's `array.reshape` only accepts C arrays.
+                            res_reshape = np.reshape(np.ascontiguousarray(x), new_shape)
+
+                            if not inplace:
+                                return res_reshape.copy()
+                            else:
+                                return res_reshape
+
+                    else:
+
+                        @numba_basic.numba_njit
+                        def dimshuffle_inner(x, shuffle):
+                            return np.reshape(np.ascontiguousarray(x), ())
+
+                    # Without the following wrapper function we would see this error:
+                    # E   No implementation of function Function(<built-in function getitem>) found for signature:
+                    # E
+                    # E    >>> getitem(UniTuple(int64 x 2), slice<a:b>)
+                    # E
+                    # E   There are 22 candidate implementations:
+                    # E      - Of which 22 did not match due to:
+                    # E      Overload of function 'getitem': File: <numerous>: Line N/A.
+                    # E        With argument(s): '(UniTuple(int64 x 2), slice<a:b>)':
+                    # E       No match.
+                    # ...(on this line)...
+                    # E           shuffle_shape = res.shape[: len(shuffle)]
+                    @numba_basic.numba_njit(inline="always")
+                    def dimshuffle(x):
+                        return dimshuffle_inner(np.asarray(x), shuffle)
+
+                    return dimshuffle
+
+        .. tab-item:: Pytorch
+
+            .. code:: python
+
+                def dimshuffle(x, op):
+                    res = torch.permute(x, op.transposition)
+
+                    shape = list(res.shape[: len(op.shuffle)])
+
+                    for augm in op.augment:
+                        shape.insert(augm, 1)
+
+                    res = torch.reshape(res, shape)
+
+                    if not op.inplace:
+                        res = res.clone()
+
+                    return res
 
 In this case, :class:`CumOp` is implemented with NumPy's :func:`numpy.cumsum`
 and :func:`numpy.cumprod`, which have JAX equivalents: :func:`jax.numpy.cumsum`
-and :func:`jax.numpy.cumprod`.
+and :func:`jax.numpy.cumprod`. The Pytorch equivalents are :func:`torch.cumsum`
+and :func:`torch.cumprod`
 
 .. code:: python
 
@@ -136,132 +268,368 @@ and :func:`jax.numpy.cumprod`.
         else:
             z[0] = np.cumprod(x, axis=self.axis)
 
-Step 3: Register the function with the `jax_funcify` dispatcher
+Step 3: Register the function with the respective dispatcher
 ---------------------------------------------------------------
 
-With the PyTensor `Op` replicated in JAX, we'll need to register the
-function with the PyTensor JAX `Linker`. This is done through the use of
+With the PyTensor `Op` replicated, we'll need to register the
+function with the backends `Linker`. This is done through the use of
 `singledispatch`. If you don't know how `singledispatch` works, see the
 `Python documentation <https://docs.python.org/3/library/functools.html#functools.singledispatch>`_.
 
-The relevant dispatch functions created by `singledispatch` are :func:`pytensor.link.numba.dispatch.numba_funcify` and
-:func:`pytensor.link.jax.dispatch.jax_funcify`.
+The relevant dispatch functions created by `singledispatch` are :func:`pytensor.link.numba.dispatch.numba_funcify`,
+:func:`pytensor.link.jax.dispatch.jax_funcify` and :func:`pytensor.link.pytorch.dispatch.pytorch_funcify`.
 
 Here's an example for the `CumOp`\ `Op`:
 
-.. code:: python
+.. tab-set::
 
-   import jax.numpy as jnp
+    .. tab-item:: JAX
 
-   from pytensor.tensor.extra_ops import CumOp
-   from pytensor.link.jax.dispatch import jax_funcify
+        .. code:: python
 
+            import jax.numpy as jnp
 
-    @jax_funcify.register(CumOp)
-    def jax_funcify_CumOp(op, **kwargs):
-        axis = op.axis
-        mode = op.mode
+            from pytensor.tensor.extra_ops import CumOp
+            from pytensor.link.jax.dispatch import jax_funcify
 
-        def cumop(x, axis=axis, mode=mode):
-            if mode == "add":
-                return jnp.cumsum(x, axis=axis)
-            else:
-                return jnp.cumprod(x, axis=axis)
 
-        return cumop
+            @jax_funcify.register(CumOp)
+            def jax_funcify_CumOp(op, **kwargs):
+                axis = op.axis
+                mode = op.mode
 
-Suppose `jnp.cumprod` does not exist, we will need to register the function as follows:
+                def cumop(x, axis=axis, mode=mode):
+                    if mode == "add":
+                        return jnp.cumsum(x, axis=axis)
+                    else:
+                        return jnp.cumprod(x, axis=axis)
 
-.. code:: python
+                return cumop
 
-   import jax.numpy as jnp
+        Suppose `jnp.cumprod` does not exist, we will need to register the function as follows:
 
-   from pytensor.tensor.extra_ops import CumOp
-   from pytensor.link.jax.dispatch import jax_funcify
+        .. code:: python
 
+            import jax.numpy as jnp
 
-    @jax_funcify.register(CumOp)
-    def jax_funcify_CumOp(op, **kwargs):
-        axis = op.axis
-        mode = op.mode
+            from pytensor.tensor.extra_ops import CumOp
+            from pytensor.link.jax.dispatch import jax_funcify
 
-        def cumop(x, axis=axis, mode=mode):
-            if mode == "add":
-                return jnp.cumsum(x, axis=axis)
-            else:
-                raise NotImplementedError("JAX does not support cumprod function at the moment.")
 
-        return cumop
+                @jax_funcify.register(CumOp)
+                def jax_funcify_CumOp(op, **kwargs):
+                    axis = op.axis
+                    mode = op.mode
 
-Step 4: Write tests
--------------------
+                    def cumop(x, axis=axis, mode=mode):
+                        if mode == "add":
+                            return jnp.cumsum(x, axis=axis)
+                        else:
+                            raise NotImplementedError("JAX does not support cumprod function at the moment.")
 
-Test that your registered `Op` is working correctly by adding tests to the
-appropriate test suites in PyTensor (e.g. in ``tests.link.jax`` and one of
-the modules in ``tests.link.numba``). The tests should ensure that your implementation can
-handle the appropriate types of inputs and produce outputs equivalent to `Op.perform`.
-Check the existing tests for the general outline of these kinds of tests. In
-most cases, a helper function can be used to easily verify the correspondence
-between a JAX/Numba implementation and its `Op`.
+                    return cumop
 
-For example, the :func:`compare_jax_and_py` function streamlines the steps
-involved in making comparisons with `Op.perform`.
+    .. tab-item:: Numba
 
-Here's a small example of a test for :class:`CumOp` above:
+        .. code:: python
 
-.. code:: python
-    
-    import numpy as np
-    import pytensor.tensor as pt
-    from pytensor.configdefaults import config
-    from tests.link.jax.test_basic import compare_jax_and_py
-    from pytensor.graph import FunctionGraph
-    from pytensor.graph.op import get_test_value
+            import numpy as np
 
-    def test_jax_CumOp():
-        """Test JAX conversion of the `CumOp` `Op`."""
+            from pytensor import config
+            from pytensor.graph import Apply
+            from pytensor.link.numba.dispatch import basic as numba_basic
+            from pytensor.tensor import TensorVariable
+            from pytensor.tensor.extra_ops import CumOp,
 
-        # Create a symbolic input for the first input of `CumOp`
-        a = pt.matrix("a")
+            def numba_funcify_CumOp(op: CumOp, node: Apply, **kwargs):
+                axis = op.axis
+                mode = op.mode
+                ndim = cast(TensorVariable, node.outputs[0]).ndim
 
-        # Create test value tag for a
-        a.tag.test_value = np.arange(9, dtype=config.floatX).reshape((3, 3))
+                if axis is not None:
+                    if axis < 0:
+                        axis = ndim + axis
+                    if axis < 0 or axis >= ndim:
+                        raise ValueError(f"Invalid axis {axis} for array with ndim {ndim}")
 
-        # Create the output variable
-        out = pt.cumsum(a, axis=0)
+                    reaxis_first = (axis, *(i for i in range(ndim) if i != axis))
+                    reaxis_first_inv = tuple(np.argsort(reaxis_first))
 
-        # Create a PyTensor `FunctionGraph`
-        fgraph = FunctionGraph([a], [out])
+                if mode == "add":
+                    if axis is None or ndim == 1:
 
-        # Pass the graph and inputs to the testing function
-        compare_jax_and_py(fgraph, [get_test_value(i) for i in fgraph.inputs])
+                        @numba_basic.numba_njit(fastmath=config.numba__fastmath)
+                        def cumop(x):
+                            return np.cumsum(x)
 
-        # For the second mode of CumOp
-        out = pt.cumprod(a, axis=1)
-        fgraph = FunctionGraph([a], [out])
-        compare_jax_and_py(fgraph, [get_test_value(i) for i in fgraph.inputs])
+                    else:
 
-If the variant :class:`CumprodOp` is not implemented, we can add a test for it as follows:
+                        @numba_basic.numba_njit(boundscheck=False, fastmath=config.numba__fastmath)
+                        def cumop(x):
+                            out_dtype = x.dtype
+                            if x.shape[axis] < 2:
+                                return x.astype(out_dtype)
+
+                            x_axis_first = x.transpose(reaxis_first)
+                            res = np.empty(x_axis_first.shape, dtype=out_dtype)
+
+                            res[0] = x_axis_first[0]
+                            for m in range(1, x.shape[axis]):
+                                res[m] = res[m - 1] + x_axis_first[m]
+
+                            return res.transpose(reaxis_first_inv)
+
+                else:
+                    if axis is None or ndim == 1:
+
+                        @numba_basic.numba_njit(fastmath=config.numba__fastmath)
+                        def cumop(x):
+                            return np.cumprod(x)
+
+                    else:
+
+                        @numba_basic.numba_njit(boundscheck=False, fastmath=config.numba__fastmath)
+                        def cumop(x):
+                            out_dtype = x.dtype
+                            if x.shape[axis] < 2:
+                                return x.astype(out_dtype)
+
+                            x_axis_first = x.transpose(reaxis_first)
+                            res = np.empty(x_axis_first.shape, dtype=out_dtype)
+
+                            res[0] = x_axis_first[0]
+                            for m in range(1, x.shape[axis]):
+                                res[m] = res[m - 1] * x_axis_first[m]
+
+                            return res.transpose(reaxis_first)
+
+                return cumop
+
+
+    .. tab-item:: Pytorch
+
+        .. code:: python
+
+            import torch
+
+            from pytensor.link.pytorch.dispatch.basic import pytorch_funcify
+            from pytensor.tensor.extra_ops import CumOp
 
-.. code:: python
 
-    import pytest
+            @pytorch_funcify.register(CumOp)
+            def pytorch_funcify_Cumop(op, **kwargs):
+                axis = op.axis
+                mode = op.mode
+
+                def cumop(x,):
+                    if axis is None:
+                        x = x.reshape(-1)
+                        dim = 0
+                    else:
+                        dim=axis
+                    if mode == "add":
+                        return torch.cumsum(x, dim=dim)
+                    else:
+                        return torch.cumprod(x, dim=dim)
+
+                return cumop
+
+
+        Suppose `torch.cumprod` does not exist, we will need to register the function as follows:
+
+        .. code:: python
+
+            import torch
+
+            from pytensor.tensor.extra_ops import CumOp
+            from pytensor.link.pytorch.dispatch import pytorch_funcify
+
+
+            @pytorch_funcify.register(CumOp)
+            def pytorch_funcify_Cumop(op, **kwargs):
+                axis = op.axis
+                mode = op.mode
+
+                def cumop(x, axis=axis, mode=mode):
+                    if mode == "add":
+                        return torch.cumsum(x, axis=axis)
+                    else:
+                        raise NotImplementedError("Pytorch does not support cumprod function at the moment.")
+
+                return cumop
+
+Step 4: Write tests
+-------------------
+.. tab-set::
+
+    .. tab-item:: JAX
+
+        Test that your registered `Op` is working correctly by adding tests to the
+        appropriate test suites in PyTensor (e.g. in ``tests.link.jax``). 
+        The tests should ensure that your implementation can
+        handle the appropriate types of inputs and produce outputs equivalent to `Op.perform`.
+        Check the existing tests for the general outline of these kinds of tests. In
+        most cases, a helper function can be used to easily verify the correspondence
+        between a Numba implementation and its `Op`.
+
+        For example, the :func:`compare_jax_and_py` function streamlines the steps
+        involved in making comparisons with `Op.perform`.
+
+        Here's a small example of a test for :class:`CumOp` above:
+
+        .. code:: python
+            
+            import numpy as np
+            import pytensor.tensor as pt
+            from pytensor.configdefaults import config
+            from tests.link.jax.test_basic import compare_jax_and_py
+            from pytensor.graph import FunctionGraph
+            from pytensor.graph.op import get_test_value
+
+            def test_jax_CumOp():
+                """Test JAX conversion of the `CumOp` `Op`."""
+
+                # Create a symbolic input for the first input of `CumOp`
+                a = pt.matrix("a")
+
+                # Create test value tag for a
+                a.tag.test_value = np.arange(9, dtype=config.floatX).reshape((3, 3))
+
+                # Create the output variable
+                out = pt.cumsum(a, axis=0)
+
+                # Create a PyTensor `FunctionGraph`
+                fgraph = FunctionGraph([a], [out])
+
+                # Pass the graph and inputs to the testing function
+                compare_jax_and_py(fgraph, [get_test_value(i) for i in fgraph.inputs])
+
+                # For the second mode of CumOp
+                out = pt.cumprod(a, axis=1)
+                fgraph = FunctionGraph([a], [out])
+                compare_jax_and_py(fgraph, [get_test_value(i) for i in fgraph.inputs])
+
+        If the variant :class:`CumprodOp` is not implemented, we can add a test for it as follows:
+
+        .. code:: python
+
+            import pytest
+            
+            def test_jax_CumOp():
+                """Test JAX conversion of the `CumOp` `Op`."""
+                a = pt.matrix("a")
+                a.tag.test_value = np.arange(9, dtype=config.floatX).reshape((3, 3))
+                
+                with pytest.raises(NotImplementedError):
+                    out = pt.cumprod(a, axis=1)
+                    fgraph = FunctionGraph([a], [out])
+                    compare_jax_and_py(fgraph, [get_test_value(i) for i in fgraph.inputs])
+    
     
-    def test_jax_CumOp():
-        """Test JAX conversion of the `CumOp` `Op`."""
-        a = pt.matrix("a")
-        a.tag.test_value = np.arange(9, dtype=config.floatX).reshape((3, 3))
+    .. tab-item:: Numba
+
+        Test that your registered `Op` is working correctly by adding tests to the
+        appropriate test suites in PyTensor (e.g. in ``tests.link.numba``). 
+        The tests should ensure that your implementation can
+        handle the appropriate types of inputs and produce outputs equivalent to `Op.perform`.
+        Check the existing tests for the general outline of these kinds of tests. In
+        most cases, a helper function can be used to easily verify the correspondence
+        between a Numba implementation and its `Op`.
+
+        For example, the :func:`compare_numba_and_py` function streamlines the steps
+        involved in making comparisons with `Op.perform`.
+
+        Here's a small example of a test for :class:`CumOp` above:
+
+        .. code:: python
+            
+            from tests.link.numba.test_basic import compare_numba_and_py
+            from pytensor.graph import FunctionGraph
+            from pytensor.compile.sharedvalue import SharedVariable
+            from pytensor.graph.basic import Constant
+            from pytensor.tensor import extra_ops
+
+            def test_CumOp(val, axis, mode):
+                g = extra_ops.CumOp(axis=axis, mode=mode)(val)
+                g_fg = FunctionGraph(outputs=[g])
+
+                compare_numba_and_py(
+                    g_fg,
+                    [
+                        i.tag.test_value
+                        for i in g_fg.inputs
+                        if not isinstance(i, SharedVariable | Constant)
+                    ],
+                )
+    
+
+
+    .. tab-item:: Pytorch
         
-        with pytest.raises(NotImplementedError):
-            out = pt.cumprod(a, axis=1)
-            fgraph = FunctionGraph([a], [out])
-            compare_jax_and_py(fgraph, [get_test_value(i) for i in fgraph.inputs])
+        Test that your registered `Op` is working correctly by adding tests to the
+        appropriate test suites in PyTensor (``tests.link.pytorch``). The tests should ensure that your implementation can
+        handle the appropriate types of inputs and produce outputs equivalent to `Op.perform`.
+        Check the existing tests for the general outline of these kinds of tests. In
+        most cases, a helper function can be used to easily verify the correspondence
+        between a Pytorch implementation and its `Op`.
+
+        For example, the :func:`compare_pytorch_and_py` function streamlines the steps
+        involved in making comparisons with `Op.perform`.
+
+        Here's a small example of a test for :class:`CumOp` above:
+
+        .. code:: python
+            
+            import numpy as np
+            import pytest
+            import pytensor.tensor as pt
+            from pytensor.configdefaults import config
+            from tests.link.pytorch.test_basic import compare_pytorch_and_py
+            from pytensor.graph import FunctionGraph
+
+            @pytest.mark.parametrize(
+                "dtype",
+                ["float64", "int64"],
+            )
+            @pytest.mark.parametrize(
+                "axis", 
+                [None, 1, (0,)],
+            )
+            def test_pytorch_CumOp(axis, dtype):
+                """Test PyTorch conversion of the `CumOp` `Op`."""
+
+                # Create a symbolic input for the first input of `CumOp`
+                a = pt.matrix("a", dtype=dtype)
+
+                # Create test value
+                test_value = np.arange(9, dtype=dtype).reshape((3, 3))
+
+                # Create the output variable
+                if isinstance(axis, tuple):
+                    with pytest.raises(TypeError, match="axis must be an integer or None."):
+                        out = pt.cumsum(a, axis=axis)
+                    with pytest.raises(TypeError, match="axis must be an integer or None."):
+                        out = pt.cumprod(a, axis=axis)
+                else:
+                    out = pt.cumsum(a, axis=axis)
+                    # Create a PyTensor `FunctionGraph`
+                    fgraph = FunctionGraph([a], [out])
+
+                    # Pass the graph and inputs to the testing function
+                    compare_pytorch_and_py(fgraph, [test_value])
+
+                    # For the second mode of CumOp
+                    out = pt.cumprod(a, axis=axis)
+                    fgraph = FunctionGraph([a], [out])
+                    compare_pytorch_and_py(fgraph, [test_value])
+
 
 Note
 ----
 In out previous example of extending JAX, :class:`Eye`\ :class:`Op` was used with the test function as follows:
 
 .. code:: python
+
     def test_jax_Eye():
         """Test JAX conversion of the `Eye` `Op`."""
 

pytensor/link/pytorch/dispatch/__init__.py

@@ -4,4 +4,5 @@ from pytensor.link.pytorch.dispatch.basic import pytorch_funcify, pytorch_typify
 # # Load dispatch specializations
 import pytensor.link.pytorch.dispatch.scalar
 import pytensor.link.pytorch.dispatch.elemwise
+import pytensor.link.pytorch.dispatch.extra_ops
 # isort: on

pytensor/link/pytorch/dispatch/extra_ops.py

+import torch
+
+from pytensor.link.pytorch.dispatch.basic import pytorch_funcify
+from pytensor.tensor.extra_ops import CumOp
+
+
+@pytorch_funcify.register(CumOp)
+def pytorch_funcify_Cumop(op, **kwargs):
+    axis = op.axis
+    mode = op.mode
+
+    def cumop(x):
+        if axis is None:
+            x = x.reshape(-1)
+            dim = 0
+        else:
+            dim = axis
+        if mode == "add":
+            return torch.cumsum(x, dim=dim)
+        else:
+            return torch.cumprod(x, dim=dim)
+
+    return cumop

pytensor/tensor/extra_ops.py

@@ -283,6 +283,8 @@ class CumOp(COp):
     def __init__(self, axis: int | None = None, mode="add"):
         if mode not in ("add", "mul"):
             raise ValueError(f'{type(self).__name__}: Unknown mode "{mode}"')
+        if not (isinstance(axis, int) or axis is None):
+            raise TypeError("axis must be an integer or None.")
         self.axis = axis
         self.mode = mode
 

tests/link/pytorch/test_extra_ops.py

+import numpy as np
+import pytest
+
+import pytensor.tensor as pt
+from pytensor.graph import FunctionGraph
+from tests.link.pytorch.test_basic import compare_pytorch_and_py
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    ["float64", "int64"],
+)
+@pytest.mark.parametrize(
+    "axis",
+    [None, 1, (0,)],
+)
+def test_pytorch_CumOp(axis, dtype):
+    """Test PyTorch conversion of the `CumOp` `Op`."""
+
+    # Create a symbolic input for the first input of `CumOp`
+    a = pt.matrix("a", dtype=dtype)
+
+    # Create test value
+    test_value = np.arange(9, dtype=dtype).reshape((3, 3))
+
+    # Create the output variable
+    if isinstance(axis, tuple):
+        with pytest.raises(TypeError, match="axis must be an integer or None."):
+            out = pt.cumsum(a, axis=axis)
+        with pytest.raises(TypeError, match="axis must be an integer or None."):
+            out = pt.cumprod(a, axis=axis)
+    else:
+        out = pt.cumsum(a, axis=axis)
+        # Create a PyTensor `FunctionGraph`
+        fgraph = FunctionGraph([a], [out])
+
+        # Pass the graph and inputs to the testing function
+        compare_pytorch_and_py(fgraph, [test_value])
+
+        # For the second mode of CumOp
+        out = pt.cumprod(a, axis=axis)
+        fgraph = FunctionGraph([a], [out])
+        compare_pytorch_and_py(fgraph, [test_value])