Skip to content

P
pytensor
提交 cae78759 authored 作者: Brandon T. Willard's avatar Brandon T. Willard 提交者： Brandon T. Willard
浏览文件
操作

Add missing formatting to extending_aesara.txt and inplace.txt

上级 5057f618
隐藏空白字符变更
内嵌 并排
正在显示 包含 106 行增加 和 113 行删除
doc/extending/extending_aesara.txt
浏览文件 @ cae78759
......@@ -7,14 +7,14 @@ Creating a new Op: Python implementation
So suppose you have looked through the library documentation and you don't see
a function that does what you want.
If you can implement something in terms of existing Ops, you should do that.
If you can implement something in terms of existing ``Op``s, you should do that.
Odds are your function that uses existing Aesara expressions is short,
has no bugs, and potentially profits from optimizations that have already been
implemented.
However, if you cannot implement an Op in terms of existing Ops, you have to
However, if you cannot implement an ``Op`` in terms of existing ``Op``s, you have to
write a new one. Don't worry, Aesara was designed to make it easy to add new
Ops, Types, and Optimizations.
``Op``s, Types, and Optimizations.
.. These first few pages will walk you through the definition of a new :ref:`type`,
.. ``double``, and a basic arithmetic :ref:`operations <op>` on that `Type`.
......@@ -23,23 +23,23 @@ As an illustration, this tutorial shows how to write a simple Python-based
:ref:`operations <op>` which performs operations on
:ref:`type`, ``double<Double>``.
.. It also shows how to implement tests that
.. ensure the proper working of an op.
.. ensure the proper working of an ``Op``.
.. note::
This is an introductury tutorial and as such it does not cover how to make
an op that returns a view or modifies the values in its inputs. Thus, all
ops created with the instructions described here MUST return newly
an ``Op`` that returns a view or modifies the values in its inputs. Thus, all
``Op``s created with the instructions described here MUST return newly
allocated memory or reuse the memory provided in the parameter
``output_storage`` of the :func:`perform` function. See
:ref:`views_and_inplace` for an explanation on how to do this.
If your op returns a view or changes the value of its inputs
If your ``Op`` returns a view or changes the value of its inputs
without doing as prescribed in that page, Aesara will run, but will
return correct results for some graphs and wrong results for others.
It is recommended that you run your tests in DebugMode (Aesara *flag*
``mode=DebugMode``) since it verifies if your op behaves correctly in this
``mode=DebugMode``) since it verifies if your ``Op`` behaves correctly in this
regard.
......@@ -57,7 +57,7 @@ intermediary values. As such, Inputs and Outputs of a graph are lists of Aesara
:ref:`variable` nodes. :ref:`apply` nodes perform computation on these
variables to produce new variables. Each :ref:`apply` node has a link to an
instance of :ref:`Op` which describes the computation to perform. This tutorial
details how to write such an Op instance. Please refers to
details how to write such an ``Op`` instance. Please refers to
:ref:`graphstructures` for a more detailed explanation about the graph
structure.
......@@ -65,9 +65,9 @@ structure.
Op's basic methods
------------------
An op is any Python object which inherits from :class:`Op`.
An ``Op`` is any Python object which inherits from :class:`Op`.
This section provides an overview of the basic methods you typically have to
implement to make a new op. It does not provide extensive coverage of all the
implement to make a new ``Op``. It does not provide extensive coverage of all the
possibilities you may encounter or need. For that refer to
:ref:`op_contract`.
......@@ -119,46 +119,46 @@ possibilities you may encounter or need. For that refer to
def infer_shape(self, fgraph, node, input_shapes):
pass
An op has to implement some methods defined in the the interface of
:class:`Op`. More specifically, it is mandatory for an op to define either
An ``Op`` has to implement some methods defined in the the interface of
:class:`Op`. More specifically, it is mandatory for an ``Op`` to define either
the method :func:`make_node` or :attr:`itypes`, :attr:`otypes` and one of the
implementation methods, either :func:`perform`, :meth:`Op.c_code`
implementation methods, either :func:`perform`, :meth:`COp.c_code`
or :func:`make_thunk`.
:func:`make_node` method creates an Apply node representing the application
of the op on the inputs provided. This method is reponsible for three things:
of the ``Op`` on the inputs provided. This method is reponsible for three things:
- it first checks that the input Variables types are compatible
with the current op. If the op cannot be applied on the provided
- it first checks that the input ``Variable``s types are compatible
with the current ``Op``. If the ``Op`` cannot be applied on the provided
input types, it must raises an exception (such as :class:`TypeError`).
- it operates on the Variables found in
- it operates on the ``Variable``s found in
``*inputs`` in Aesara's symbolic language to infer the type of
the symbolic output Variables. It creates output Variables of a suitable
symbolic `Type` to serve as the outputs of this op's
the symbolic output ``Variable``s. It creates output ``Variable``s of a suitable
symbolic `Type` to serve as the outputs of this ``Op``'s
application.
- it creates an Apply instance with the input and output Variable, and
- it creates an Apply instance with the input and output ``Variable``, and
return the Apply instance.
:func:`perform` method defines the Python implementation of an op.
:func:`perform` method defines the Python implementation of an ``Op``.
It takes several arguments:
- ``node`` is a reference to an Apply node which was previously
obtained via the ``Op``'s :func:`make_node` method. It is typically not
used in simple ops, but it contains symbolic information that
could be required for complex ops.
used in simple ``Op``s, but it contains symbolic information that
could be required for complex ``Op``s.
- ``inputs`` is a list of references to data which can be operated on using
non-symbolic statements, (i.e., statements in Python, Numpy).
- ``output_storage`` is a list of storage cells where the output
is to be stored. There is one storage cell for each output of the op.
is to be stored. There is one storage cell for each output of the ``Op``.
The data put in ``output_storage`` must match the type of the
symbolic output. It is forbidden to change the length of the list(s)
contained in ``output_storage``.
A function Mode may allow ``output_storage`` elements to persist
between evaluations, or it may reset ``output_storage`` cells to
hold a value of ``None``. It can also pre-allocate some memory
for the op to use. This feature can allow ``perform`` to reuse
for the ``Op`` to use. This feature can allow ``perform`` to reuse
memory between calls, for example. If there is something
preallocated in the ``output_storage``, it will be of the good
dtype, but can have the wrong shape and have any stride pattern.
......@@ -166,17 +166,17 @@ or :func:`make_thunk`.
:func:`perform` method must be determined by the inputs. That is to say,
when applied to identical inputs the method must return the same outputs.
:class:`Op` allows some other way to define the op implentation.
For instance, it is possible to define :meth:`Op.c_code` to provide a
C-implementation to the op. Please refers to tutorial
:ref:`extending_aesara_c` for a description of :meth:`Op.c_code` and other
related c_methods. Note that an op can provide both Python and C
:class:`Op` allows some other way to define the ``Op`` implentation.
For instance, it is possible to define :meth:`COp.c_code` to provide a
C-implementation to the ``Op``. Please refers to tutorial
:ref:`extending_aesara_c` for a description of :meth:`COp.c_code` and other
related c_methods. Note that an ``Op`` can provide both Python and C
implementation.
:func:`make_thunk` method is another alternative to :func:`perform`.
It returns a thunk. A thunk is defined as a zero-arguments
function which encapsulates the computation to be performed by an
op on the arguments of its corresponding node. It takes several parameters:
``Op`` on the arguments of its corresponding node. It takes several parameters:
- ``node`` is the Apply instance for which a thunk is requested,
- ``storage_map`` is a dict of lists which maps variables to a one-element
......@@ -198,28 +198,28 @@ or :func:`make_thunk`.
:func:`make_thunk` is useful if you want to generate code and compile
it yourself.
If :func:`make_thunk()` is defined by an op, it will be used by Aesara
to obtain the op's implementation.
:func:`perform` and :meth:`Op.c_code` will be ignored.
If :func:`make_thunk()` is defined by an ``Op``, it will be used by Aesara
to obtain the ``Op``'s implementation.
:func:`perform` and :meth:`COp.c_code` will be ignored.
If :func:`make_node` is not defined, the :attr:`itypes` and :attr:`otypes`
are used by the Op's :func:`make_node` method to implement the functionality
are used by the ``Op``'s :func:`make_node` method to implement the functionality
of :func:`make_node` method mentioned above.
Op's auxiliary methods
----------------------
There are other methods that can be optionally defined by the op:
There are other methods that can be optionally defined by the ``Op``:
The :func:`__str__` method provides a meaningful string representation of
your op.
your ``Op``.
:func:`__eq__` and :func:`__hash__` define respectivelly equality
between two ops and the hash of an op instance.
between two ``Op``s and the hash of an ``Op`` instance.
They will be used by the optimization
phase to merge nodes that are doing equivalent computations (same
inputs, same operation).
Two ops that are equal according :func:`__eq__`
Two ``Op``s that are equal according :func:`__eq__`
should return the same output when they are applied on the same inputs.
The :attr:`__props__` lists the properties
......@@ -231,19 +231,19 @@ There are other methods that can be optionally defined by the op:
:attr:`__props__` enables the automatic generation of appropriate
:func:`__eq__` and :func:`__hash__`.
Given the method :func:`__eq__`, automatically generated from
:attr:`__props__`, two ops will be equal if they have the same values for all
:attr:`__props__`, two ``Op``s will be equal if they have the same values for all
the properties listed in :attr:`__props__`.
Given to the method :func:`__hash__` automatically generated from
:attr:`__props__`, two ops will be have the same hash if they have the same
:attr:`__props__`, two ``Op``s will be have the same hash if they have the same
values for all the properties listed in :attr:`__props__`.
:attr:`__props__` will also generate a suitable :func:`__str__` for your op.
:attr:`__props__` will also generate a suitable :func:`__str__` for your ``Op``.
This requires development version after September 1st, 2014 or version 0.7.
The :func:`infer_shape` method allows an `Op` to infer the shape of its
output variables without actually computing them.
It takes as input ``fgraph``, a `FunctionGraph`; ``node``, a reference to the op Apply node;
It takes as input ``fgraph``, a `FunctionGraph`; ``node``, a reference to the ``Op`` Apply node;
and a list of Aesara symbolic Varables (``i0_shape``, ``i1_shape``, ...)
which are the shape of the op input Variables.
which are the shape of the ``Op`` input ``Variable``s.
:func:`infer_shape` returns a list where each element is a tuple representing
the shape of one output.
This could be helpful if one only
......@@ -251,12 +251,12 @@ There are other methods that can be optionally defined by the op:
can be useful, for instance, for optimization procedures.
The :func:`grad` method is required if you want to differentiate some cost
whose expression includes your op. The gradient may be
whose expression includes your ``Op``. The gradient may be
specified symbolically in this method. It takes two arguments ``inputs`` and
``output_gradients`` which are both lists of symbolic Aesara Variables and
``output_gradients`` which are both lists of symbolic Aesara ``Variable``s and
those must be operated on using Aesara's symbolic language. The grad
method must return a list containing one Variable for each
input. Each returned Variable represents the gradient with respect
method must return a list containing one ``Variable`` for each
input. Each returned ``Variable`` represents the gradient with respect
to that input computed based on the symbolic gradients with respect
to each output.
If the output is not differentiable with respect to an input then
......@@ -275,8 +275,8 @@ There are other methods that can be optionally defined by the op:
point, namely: :math:`\frac{\partial f}{\partial x} v`.
The optional boolean :attr:`check_input` attribute is used to specify
if you want the types used in your op to check their inputs in their
c_code. It can be used to speed up compilation, reduce overhead
if you want the types used in your ``COp`` to check their inputs in their
``COp.c_code``. It can be used to speed up compilation, reduce overhead
(particularly for scalars) and reduce the number of generated C files.
......@@ -356,22 +356,22 @@ At a high level, the code fragment declares a class (e.g., ``DoubleOp1``) and th
creates one instance of it (e.g., ``doubleOp1``).
We often gloss over this distinction, but will be precise here:
``doubleOp1`` (the instance) is an Op, not ``DoubleOp1`` (the class which is a
``doubleOp1`` (the instance) is an ``Op``, not ``DoubleOp1`` (the class which is a
subclass of ``Op``). You can call ``doubleOp1(tensor.vector())`` on a
Variable to build an expression, and in the expression there will be
``Variable`` to build an expression, and in the expression there will be
a ``.op`` attribute that refers to ``doubleOp1``.
.. The first two methods in the Op are relatively boilerplate: ``__eq__``
.. The first two methods in the ``Op`` are relatively boilerplate: ``__eq__``
.. and ``__hash__``.
.. When two Ops are equal, Aesara will merge their outputs if they are applied to the same inputs.
.. When two ``Op``s are equal, Aesara will merge their outputs if they are applied to the same inputs.
.. The base class (Op) says two objects are equal if (and only if)
.. they are the same object.
.. Writing these boilerplate definitions ensures that the logic of the equality comparison is always explicit.
.. It is an essential part of the :ref:`op_contract` that if two Ops compare
.. It is an essential part of the :ref:`op_contract` that if two ``Op``s compare
.. equal, then they must compute the same result when presented with the same
.. inputs. Here, if we allocated another instance of ``Fibby`` by typing ``fibby2
.. = Fibby()`` then we would have two Ops that behave identically.
.. = Fibby()`` then we would have two ``Op``s that behave identically.
..
.. When should the implementation of ``__eq__`` be more complicated?
.. If ``Fibby.__init__`` had parameters, then we could
......@@ -379,27 +379,27 @@ a ``.op`` attribute that refers to ``doubleOp1``.
.. arguments to the constructor. If we had done that, and if that different
.. configuration made ``fibby2`` compute different results from ``fibby`` (for the
.. same inputs) then we would have to add logic to the ``__eq__`` and ``__hash__``
.. function so that he two ``Fibby`` Ops would *not be equal*. The reason why: Aesara's merge
.. optimization looks for Ops comparing equal and merges them. If two Ops compare
.. function so that he two ``Fibby`` ``Op``s would *not be equal*. The reason why: Aesara's merge
.. optimization looks for ``Op``s comparing equal and merges them. If two ``Op``s compare
.. equal but don't always produce equal results from equal inputs, then you might
.. see wrong calculation.
The ``make_node`` method creates a node to be included in the expression graph.
It runs when we apply our Op (``doubleOp1``) to the Variable (``x``), as
It runs when we apply our ``Op`` (``doubleOp1``) to the ``Variable`` (``x``), as
in ``doubleOp1(tensor.vector())``.
When an Op has multiple inputs, their order in the inputs argument to ``Apply``
When an ``Op`` has multiple inputs, their order in the inputs argument to ``Apply``
is important: Aesara will call ``make_node(*inputs)`` to copy the graph,
so it is important not to change the semantics of the expression by changing
the argument order.
All the ``inputs`` and ``outputs`` arguments to ``Apply`` must be Variables.
All the ``inputs`` and ``outputs`` arguments to ``Apply`` must be ``Variable``s.
A common and easy way to ensure inputs are variables is to run them through
``as_tensor_variable``. This function leaves TensorType variables alone, raises
an error for non-TensorType variables, and copies any ``numpy.ndarray`` into
the storage for a TensorType Constant. The ``make_node`` method dictates the
appropriate `Type` for all output variables.
The ``perform`` method implements the Op's mathematical logic in Python.
The ``perform`` method implements the ``Op``'s mathematical logic in Python.
The inputs (here ``x``) are passed by value, but a single output is returned
indirectly as the first element of single-element lists. If ``doubleOp1`` had
a second output, it would be stored in ``output_storage[1][0]``.
......@@ -408,9 +408,9 @@ a second output, it would be stored in ``output_storage[1][0]``.
In some execution modes, the output storage might contain the return value of
a previous call. That old value can be reused to avoid memory re-allocation,
but it must not influence the semantics of the Op output.
but it must not influence the semantics of the ``Op`` output.
You can try the new Op as follows:
You can try the new ``Op`` as follows:
.. testcode:: example
......@@ -482,8 +482,8 @@ Example: __props__ definition
We can modify the previous piece of code in order to demonstrate
the usage of the :attr:`__props__` attribute.
We create an Op that takes a variable ``x`` and returns ``a*x+b``.
We want to say that two such ops are equal when their values of ``a``
We create an ``Op`` that takes a variable ``x`` and returns ``a*x+b``.
We want to say that two such ``Op``s are equal when their values of ``a``
and ``b`` are equal.
.. testcode:: properties
......@@ -556,7 +556,7 @@ in a file and execute it with the ``pytest`` program.
Basic Tests
^^^^^^^^^^^
Basic tests are done by you just by using the op and checking that it
Basic tests are done by you just by using the ``Op`` and checking that it
returns the right answer. If you detect an error, you must raise an
*exception*. You can use the ``assert`` keyword to automatically raise an
``AssertionError``.
......@@ -593,8 +593,8 @@ Testing the infer_shape
^^^^^^^^^^^^^^^^^^^^^^^
When a class inherits from the ``InferShapeTester`` class, it gets the
``self._compile_and_check`` method that tests the op's ``infer_shape``
method. It tests that the op gets optimized out of the graph if only
``self._compile_and_check`` method that tests the ``Op``'s ``infer_shape``
method. It tests that the ``Op`` gets optimized out of the graph if only
the shape of the output is needed and not the output
itself. Additionally, it checks that the optimized graph computes
the correct shape, by comparing it to the actual shape of the computed
......@@ -603,7 +603,7 @@ output.
``self._compile_and_check`` compiles an Aesara function. It takes as
parameters the lists of input and output Aesara variables, as would be
provided to ``aesara.function``, and a list of real values to pass to the
compiled function. It also takes the op class as a parameter
compiled function. It also takes the ``Op`` class as a parameter
in order to verify that no instance of it appears in the shape-optimized graph.
If there is an error, the function raises an exception. If you want to
......@@ -617,7 +617,7 @@ same value have been mixed up. For instance, if the infer_shape uses
the width of a matrix instead of its height, then testing with only
square matrices will not detect the problem. This is why the
``self._compile_and_check`` method prints a warning in such a case. If
your op works only with such matrices, you can disable the warning with the
your ``Op`` works only with such matrices, you can disable the warning with the
``warn=False`` parameter.
.. testcode:: tests
......@@ -641,7 +641,7 @@ Testing the gradient
^^^^^^^^^^^^^^^^^^^^
The function :ref:`verify_grad <validating_grad>`
verifies the gradient of an op or Aesara graph. It compares the
verifies the gradient of an ``Op`` or Aesara graph. It compares the
analytic (symbolically computed) gradient and the numeric
gradient (computed through the Finite Difference Method).
......@@ -663,7 +663,7 @@ Testing the Rop
The class :class:`RopLop_checker` defines the functions
:func:`RopLop_checker.check_mat_rop_lop`, :func:`RopLop_checker.check_rop_lop` and
:func:`RopLop_checker.check_nondiff_rop`. These allow to test the
implementation of the Rop method of a particular op.
implementation of the Rop method of a particular ``Op``.
For instance, to verify the Rop method of the DoubleOp, you can use this:
......@@ -745,9 +745,9 @@ as_op
-----
as_op is a python decorator that converts a python function into a
basic Aesara op that will call the supplied function during execution.
basic Aesara ``Op`` that will call the supplied function during execution.
This isn't the recommended way to build an op, but allows for a quick
This isn't the recommended way to build an ``Op``, but allows for a quick
implementation.
It takes an optional :func:`infer_shape` parameter that must have this
......@@ -766,14 +766,14 @@ signature:
.. note::
Not providing the `infer_shape` method prevents shape-related
optimizations from working with this op. For example
`your_op(inputs, ...).shape` will need the op to be executed just
optimizations from working with this ``Op``. For example
`your_op(inputs, ...).shape` will need the ``Op`` to be executed just
to get the shape.
.. note::
As no grad is defined, this means you won't be able to
differentiate paths that include this op.
differentiate paths that include this ``Op``.
.. note::
......@@ -818,12 +818,11 @@ You can try it as follows:
Exercise
^^^^^^^^
Run the code of the *numpy_dot* example above.
Run the code of the *``numpy_dot``* example above.
Modify and execute to compute: numpy.add and numpy.subtract.
Modify and execute to compute: ``numpy.add`` and ``numpy.subtract``.
Modify and execute the example to return two outputs: x + y
and x - y.
Modify and execute the example to return two outputs: ``x + y`` and ``x - y``.
.. _Documentation:
......@@ -835,14 +834,14 @@ will not be accepted.
NanGuardMode and AllocEmpty
---------------------------
NanGuardMode help users find where in the graph NaN appear. But
``NanGuardMode`` help users find where in the graph NaN appear. But
sometimes, we want some variables to not be checked. For example, in
the old GPU back-end, we use a float32 CudaNdarray to store the MRG
random number generator state (they are integers). So if NanGuardMode
random number generator state (they are integers). So if ``NanGuardMode``
check it, it will generate false positive. Another case is related to
[Gpu]AllocEmpty or some computation on it (like done by Scan).
``[Gpu]AllocEmpty`` or some computation on it (like done by ``Scan``).
You can tell NanGuardMode to do not check a variable with:
You can tell ``NanGuardMode`` to do not check a variable with:
``variable.tag.nan_guard_mode_check``. Also, this tag automatically
follow that variable during optimization. This mean if you tag a
variable that get replaced by an inplace version, it will keep that
......@@ -855,7 +854,7 @@ Final Note
A more extensive discussion of this section's content may be found in
the advanced tutorial :ref:`Extending Aesara<extending>`.
The section :ref:`Other ops <other_ops>` includes more instructions for
The section :ref:`Other ``Op``s <other_ops>` includes more instructions for
the following specific cases:
- :ref:`scalar_ops`
......
doc/extending/inplace.txt
浏览文件 @ cae78759
......@@ -5,23 +5,17 @@
Views and inplace operations
============================
Aesara allows the definition of Ops which return a :term:`view` on one
Aesara allows the definition of ``Op``s which return a :term:`view` on one
of their inputs or operate :term:`inplace` on one or several
inputs. This allows more efficient operations on numpy's ``ndarray``
inputs. This allows more efficient operations on NumPy's ``ndarray``
data type than would be possible otherwise.
However, in order to work correctly, these Ops need to
However, in order to work correctly, these ``Op``s need to
implement an additional interface.
Aesara recognizes views and inplace operations specially. It ensures
that they are used in a consistent manner and it ensures that
operations will be carried in a compatible order.
An unfortunate fact is that it is impossible to return a view on an
input with the ``double`` type or to operate inplace on it (Python
floats are immutable). Therefore, we can't make examples of these
concepts out of what we've just built. Nonetheless, we will present
the concepts:
.. _views:
Views
......@@ -50,7 +44,7 @@ range ``0xDEADBEFF - 0xDEADBFDF`` and z the range ``0xCAFEBABE -
0xCAFEBBBE``. Since the ranges for ``x`` and ``y`` overlap, ``y`` is
considered to be a view of ``x`` and vice versa.
Suppose you had an Op which took ``x`` as input and returned
Suppose you had an ``Op`` which took ``x`` as input and returned
``y``. You would need to tell Aesara that ``y`` is a view of ``x``. For this
purpose, you would set the ``view_map`` field as follows:
......@@ -126,7 +120,7 @@ operation on ``x``.
r4 = log(r2)
Needless to say, this goes for user-defined inplace operations as
well: the modified input must figure in the list of outputs you
well; the modified input must figure in the list of outputs you
give to ``Apply`` in the definition of ``make_node``.
Also, for technical reasons but also because they are slightly
......@@ -140,13 +134,13 @@ operation on ``x``.
introduces inconsistencies.
Take the previous definitions of ``x``, ``y`` and ``z`` and suppose an Op which
Take the previous definitions of ``x``, ``y`` and ``z`` and suppose an ``Op`` which
adds one to every byte of its input. If we give ``x`` as an input to
that Op, it can either allocate a new buffer of the same size as ``x``
that ``Op``, it can either allocate a new buffer of the same size as ``x``
(that could be ``z``) and set that new buffer's bytes to the variable of
the addition. That would be a normal, :term:`pure` Op. Alternatively,
the addition. That would be a normal, :term:`pure` ``Op``. Alternatively,
it could add one to each byte *in* the buffer ``x``, therefore
changing it. That would be an inplace Op.
changing it. That would be an inplace ``Op``.
Aesara needs to be notified of this fact. The syntax is similar to
that of ``view_map``:
......@@ -181,11 +175,11 @@ Destructive Operations
======================
While some operations will operate inplace on their inputs, some might
simply destroy or corrupt them. For example, an Op could do temporary
simply destroy or corrupt them. For example, an ``Op`` could do temporary
calculations right in its inputs. If that is the case, Aesara also
needs to be notified. The way to notify Aesara is to assume that some
output operated inplace on whatever inputs are changed or corrupted by
the Op (even if the output does not technically reuse any of the
the ``Op`` (even if the output does not technically reuse any of the
input(s)'s memory). From there, go to the previous section.
......@@ -203,24 +197,24 @@ input(s)'s memory). From there, go to the previous section.
certainly lead to erroneous computations.
You can often identify an incorrect ``view_map`` or ``destroy_map``
by using :ref:`DebugMode`. *Be sure to use DebugMode when developing
a new Op that uses ``view_map`` and/or ``destroy_map``.*
by using :ref:`DebugMode`. *Be sure to use ``DebugMode`` when developing
a new ``Op`` that uses ``view_map`` and/or ``destroy_map``.*
Inplace optimization and DebugMode
==================================
It is recommended that during the graph construction, all Ops are not inplace.
Then an optimization replaces them with inplace ones. Currently DebugMode checks
It is recommended that during the graph construction, all ``Op``s are not inplace.
Then an optimization replaces them with inplace ones. Currently ``DebugMode`` checks
all optimizations that were tried even if they got rejected. One reason an inplace
optimization can get rejected is when there is another Op that is already being applied
optimization can get rejected is when there is another ``Op`` that is already being applied
inplace on the same input. Another reason to reject an inplace optimization is
if it would introduce a cycle into the graph.
The problem with DebugMode is that it will trigger a useless error when
The problem with ``DebugMode`` is that it will trigger a useless error when
checking a rejected inplace optimization, since it will lead to wrong results.
In order to be able to use DebugMode in more situations, your inplace
In order to be able to use ``DebugMode`` in more situations, your inplace
optimization can pre-check whether it will get rejected by using the
``aesara.graph.destroyhandler.fast_inplace_check()`` function, that will tell
which Ops can be performed inplace. You may then skip the optimization if it is
which ``Op``s can be performed inplace. You may then skip the optimization if it is
incompatible with this check. Note however that this check does not cover all
cases where an optimization may be rejected (it will not detect cycles).
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论