Multi-dimensional arrays¶
The functionality in this module provides something of a work-alike for
numpy
arrays, but with all operations executed on the CL compute device.
Data Types¶
PyOpenCL provides some amount of integration between the numpy
type system, as represented by numpy.dtype
, and the types
available in OpenCL. All the simple scalar types map straightforwardly
to their CL counterparts.
Vector Types¶
- class pyopencl.array.vec¶
All of OpenCL’s supported vector types, such as
float3
andlong4
are available asnumpy
data types within this class. Thesenumpy.dtype
instances have field names ofx
,y
,z
, andw
just like their OpenCL counterparts. They will work both for parameter passing to kernels as well as for passing data back and forth between kernels and Python code. For each type, amake_type
function is also provided (e.g.make_float3(x, y, z)
).If you want to construct a pre-initialized vector type you have three new functions to choose from:
zeros_type()
ones_type()
filled_type(fill_value)
Added in version 2014.1.
Changed in version 2014.1: The
make_type
functions have a default value (0) for each component. Relying on the default values has been deprecated. Either specify all components or use one of th new flavors mentioned above for constructing a vector.
Custom data types¶
If you would like to use your own (struct/union/whatever) data types in array
operations where you supply operation source code, define those types in the
preamble passed to pyopencl.elementwise.ElementwiseKernel
,
pyopencl.reduction.ReductionKernel
(or similar), and let PyOpenCL know
about them using this function:
- pyopencl.tools.get_or_register_dtype(c_names, dtype=None)[source]¶
Get or register a
numpy.dtype
associated with the C type names in the string list c_names. If dtype is None, no registration is performed, and thenumpy.dtype
must already have been registered. If so, it is returned. If not,TypeNameNotKnown
is raised.If dtype is not None, registration is attempted. If the c_names are already known and registered to identical
numpy.dtype
objects, then the previously dtype object of the previously registered type is returned. If the c_names are not yet known, the type is registered. If one of the c_names is known but registered to a different type, an error is raised. In this latter case, the type may end up partially registered and any further behavior is undefined.Added in version 2012.2.
- pyopencl.tools.register_dtype(dtype, name)[source]¶
Changed in version 2013.1: This function has been deprecated. It is recommended that you develop against the new interface,
get_or_register_dtype()
.
This function helps with producing C/OpenCL declarations for structured
numpy.dtype
instances:
- pyopencl.tools.match_dtype_to_c_struct(device, name, dtype, context=None)[source]¶
Return a tuple
(dtype, c_decl)
such that the C struct declaration inc_decl
and the structurenumpy.dtype
instancedtype
have the same memory layout.Note that dtype may be modified from the value that was passed in, for example to insert padding.
(As a remark on implementation, this routine runs a small kernel on the given device to ensure that
numpy
and C offsets and sizes match.)Added in version 2013.1.
This example explains the use of this function:
>>> import numpy as np >>> import pyopencl as cl >>> import pyopencl.tools >>> ctx = cl.create_some_context() >>> dtype = np.dtype([("id", np.uint32), ("value", np.float32)]) >>> dtype, c_decl = pyopencl.tools.match_dtype_to_c_struct( ... ctx.devices[0], 'id_val', dtype) >>> print c_decl typedef struct { unsigned id; float value; } id_val; >>> print dtype [('id', '<u4'), ('value', '<f4')] >>> cl.tools.get_or_register_dtype('id_val', dtype)
As this example shows, it is important to call
get_or_register_dtype()
on the modifieddtype
returned by this function, not the original one.
A more complete example of how to use custom structured types can be
found in examples/demo-struct-reduce.py
in the PyOpenCL
distribution.
Complex Numbers¶
PyOpenCL’s Array
type supports complex numbers out of the box, by
simply using the corresponding numpy
types.
If you would like to use this support in your own kernels, here’s how to proceed: Since OpenCL 1.2 (and earlier) do not specify native complex number support, PyOpenCL works around that deficiency. By saying:
#include <pyopencl-complex.h>
in your kernel, you get complex types cfloat_t
and cdouble_t
, along with
functions defined on them such as cfloat_mul(a, b)
or cdouble_log(z)
.
Elementwise kernels automatically include the header if your kernel has
complex input or output.
See the source file
for a precise list of what’s available.
If you need double precision support, please:
#define PYOPENCL_DEFINE_CDOUBLE
before including the header, as DP support apparently cannot be reliably autodetected.
Under the hood, the complex types are struct types as defined in the header. Ideally, you should only access the structs through the provided functions, never directly.
Added in version 2012.1.
Changed in version 2015.2: [INCOMPATIBLE] Changed PyOpenCL’s complex numbers from float2
and
double2
OpenCL vector types to custom struct
. This was changed
because it very easily introduced bugs where
complex*complex
real+complex
look like they may do the right thing, but silently do the wrong thing.
The Array
Class¶
- class pyopencl.array.Array(cq: Context | CommandQueue | None, shape: Tuple[int, ...] | int, dtype: Any, order: str = 'C', allocator: AllocatorBase | None = None, data: Any = None, offset: int = 0, strides: Tuple[int, ...] | None = None, events: List[Event] | None = None, _flags: Any = None, _fast: bool = False, _size: int | None = None, _context: Context | None = None, _queue: CommandQueue | None = None)[source]¶
A
numpy.ndarray
work-alike that stores its data and performs its computations on the compute device.shape
anddtype
work exactly as innumpy
. Arithmetic methods inArray
support the broadcasting of scalars. (e.g.array + 5
).cq must be a
CommandQueue
or aContext
.If it is a queue, cq specifies the queue in which the array carries out its computations by default. If a default queue (and thereby overloaded operators and many other niceties) are not desired, pass a
Context
.allocator may be None or a callable that, upon being called with an argument of the number of bytes to be allocated, returns a
pyopencl.Buffer
object. (Apyopencl.tools.MemoryPool
instance is one useful example of an object to pass here.)Changed in version 2011.1: Renamed context to cqa, made it general-purpose.
All arguments beyond order should be considered keyword-only.
Changed in version 2015.2: Renamed context to cq, disallowed passing allocators through it.
- data¶
The
pyopencl.MemoryObject
instance created for the memory that backs thisArray
.Changed in version 2013.1: If a non-zero
offset
has been specified for this array, this will fail withArrayHasOffsetError
.
- base_data¶
The
pyopencl.MemoryObject
instance created for the memory that backs thisArray
. Unlikedata
, the base address of base_data is allowed to be different from the beginning of the array. The actual beginning is the base address of base_data plusoffset
bytes.Unlike
data
, retrievingbase_data
always succeeds.Added in version 2013.1.
- shape¶
A tuple of lengths of each dimension in the array.
- dtype¶
The
numpy.dtype
of the items in the GPU array.
- size¶
The number of meaningful entries in the array. Can also be computed by multiplying up the numbers in
shape
.
- strides¶
A tuple of bytes to step in each dimension when traversing an array.
- flags¶
An object with attributes
c_contiguous
,f_contiguous
andforc
, which may be used to query contiguity properties in analogy tonumpy.ndarray.flags
.
Methods
- with_queue(queue)[source]¶
Return a copy of self with the default queue set to queue.
None is allowed as a value for queue.
Added in version 2013.1.
- view(dtype=None)[source]¶
Returns view of array with the same data. If dtype is different from current dtype, the actual bytes of memory will be reinterpreted.
- squeeze()[source]¶
Returns a view of the array with dimensions of length 1 removed.
Added in version 2015.2.
- transpose(axes=None)[source]¶
Permute the dimensions of an array.
- Parameters:
axes – list of ints, optional. By default, reverse the dimensions, otherwise permute the axes according to the values given.
- Returns:
Array
A view of the array with its axes permuted.
Added in version 2015.2.
- T¶
- set(ary, queue=None, async_=None, **kwargs)[source]¶
Transfer the contents the
numpy.ndarray
object ary onto the device.ary must have the same dtype and size (not necessarily shape) as self.
async_ is a Boolean indicating whether the function is allowed to return before the transfer completes. To avoid synchronization bugs, this defaults to False.
Changed in version 2017.2.1: Python 3.7 makes
async
a reserved keyword. On older Pythons, we will continue to accept async as a parameter, however this should be considered deprecated. async_ is the new, official spelling.
- get(queue=None, ary=None, async_=None, **kwargs)[source]¶
Transfer the contents of self into ary or a newly allocated
numpy.ndarray
. If ary is given, it must have the same shape and dtype.Changed in version 2019.1.2: Calling with
async_=True
was deprecated and replaced byget_async()
. The event returned bypyopencl.enqueue_copy()
is now stored intoevents
to ensure data is not modified before the copy is complete.Changed in version 2015.2: ary with different shape was deprecated.
Changed in version 2017.2.1: Python 3.7 makes
async
a reserved keyword. On older Pythons, we will continue to accept async as a parameter, however this should be considered deprecated. async_ is the new, official spelling.
- get_async(queue=None, ary=None, **kwargs)[source]¶
Asynchronous version of
get()
which returns a tuple(ary, event)
containing the host arrayary
and thepyopencl.NannyEvent
event
returned bypyopencl.enqueue_copy()
.Added in version 2019.1.2.
- copy(queue=<class 'pyopencl.array._copy_queue'>)[source]¶
- Parameters:
queue – The
CommandQueue
for the returned array.
Changed in version 2017.1.2: Updates the queue of the returned array.
Added in version 2013.1.
- real¶
Added in version 2012.1.
- imag¶
Added in version 2012.1.
- __setitem__(subscript, value)[source]¶
Set the slice of self identified subscript to value.
value is allowed to be:
A
Array
of the sameshape
and (for now)strides
, but with potentially differentdtype
.A
numpy.ndarray
of the sameshape
and (for now)strides
, but with potentially differentdtype
.A scalar.
Non-scalar broadcasting is not currently supported.
Added in version 2013.1.
- setitem(subscript, value, queue=None, wait_for=None)[source]¶
Like
__setitem__()
, but with the ability to specify a queue and wait_for.Added in version 2013.1.
Changed in version 2013.2: Added wait_for.
- map_to_host(queue=None, flags=None, is_blocking=True, wait_for=None)[source]¶
If is_blocking, return a
numpy.ndarray
corresponding to the same memory as self.If is_blocking is not true, return a tuple
(ary, evt)
, where ary is the above-mentioned array.The host array is obtained using
pyopencl.enqueue_map_buffer()
. See there for further details.- Parameters:
flags – A combination of
pyopencl.map_flags
. Defaults to read-write.
Added in version 2013.2.
Comparisons, conditionals, any, all
Added in version 2013.2.
Boolean arrays are stored as
numpy.int8
becausebool
has an unspecified size in the OpenCL spec.Event management
If an array is used from within an out-of-order queue, it needs to take care of its own operation ordering. The facilities in this section make this possible.
Added in version 2014.1.1.
- events¶
A list of
pyopencl.Event
instances that the current content of this array depends on. User code may read, but should never modify this list directly. To update this list, instead use the following methods.
- exception pyopencl.array.ArrayHasOffsetError(val='The operation you are attempting does not yet support arrays that start at an offset from the beginning of their buffer.')[source]¶
Added in version 2013.1.
Constructing Array
Instances¶
- pyopencl.array.to_device(queue, ary, allocator=None, async_=None, array_queue=<class 'pyopencl.array._same_as_transfer'>, **kwargs)[source]¶
Return a
Array
that is an exact copy of thenumpy.ndarray
instance ary.- Parameters:
array_queue – The
CommandQueue
which will be stored in the resulting array. Useful to make sure there is no implicit queue associated with the array by passing None.
See
Array
for the meaning of allocator.Changed in version 2015.2: array_queue argument was added.
Changed in version 2017.2.1: Python 3.7 makes
async
a reserved keyword. On older Pythons, we will continue to accept async as a parameter, however this should be considered deprecated. async_ is the new, official spelling.
- pyopencl.array.empty(queue, shape, dtype, order='C', allocator=None, data=None)[source]¶
A synonym for the
Array
constructor.
- pyopencl.array.zeros(queue, shape, dtype, order='C', allocator=None)[source]¶
Same as
empty()
, but theArray
is zero-initialized before being returned.Changed in version 2011.1: context argument was deprecated.
- pyopencl.array.empty_like(ary, queue=<class 'pyopencl.array._copy_queue'>, allocator=None)[source]¶
Make a new, uninitialized
Array
having the same properties as other_ary.
- pyopencl.array.zeros_like(ary)[source]¶
Make a new, zero-initialized
Array
having the same properties as other_ary.
- pyopencl.array.arange(queue, [start, ]stop, [step, ]**kwargs)[source]¶
Create a
Array
filled with numbers spaced step apart, starting from start and ending at stop. If not given, start defaults to 0, step defaults to 1.For floating point arguments, the length of the result is
ceil((stop - start)/step)
. This rule may result in the last element of the result being greater than stop.dtype is a required keyword argument.
Changed in version 2011.1: context argument was deprecated.
Changed in version 2011.2: allocator keyword argument was added.
- pyopencl.array.take(a, indices, out=None, queue=None, wait_for=None)[source]¶
Return the
Array
[a[indices[0]], ..., a[indices[n]]]
. For the moment, a must be a type that can be bound to a texture.
Manipulating Array
instances¶
Conditionals¶
- pyopencl.array.if_positive(criterion, then_, else_, out=None, queue=None)[source]¶
Return an array like then_, which, for the element at index i, contains then_[i] if criterion[i]>0, else else_[i].
Logical Operations¶
- pyopencl.array.logical_and(x1, x2, /, out=None, queue=None)[source]¶
Returns the element-wise logical AND of x1 and x2.
Reductions¶
- pyopencl.array.sum(a, dtype=None, queue=None, slice=None, initial=<no value>)[source]¶
Added in version 2011.1.
- pyopencl.array.vdot(a, b, dtype=None, queue=None, slice=None)[source]¶
Like
numpy.vdot()
.Added in version 2013.1.
- pyopencl.array.subset_dot(subset, a, b, dtype=None, queue=None, slice=None)[source]¶
Added in version 2011.1.
See also Sums and counts (“reduce”).
Elementwise Functions on Array
Instances¶
The pyopencl.clmath
module contains exposes array versions of the C
functions available in the OpenCL standard. (See table 6.8 in the spec.)
- pyopencl.clmath.fmod(arg, mod, queue=None)[source]¶
Return the floating point remainder of the division
arg / mod
, for each element inarg
andmod
.
- pyopencl.clmath.frexp(arg, queue=None)[source]¶
Return a tuple
(significands, exponents)
such thatarg == significand * 2**exponent
.
- pyopencl.clmath.ldexp(significand, exponent, queue=None)[source]¶
Return a new array of floating point values composed from the entries of
significand
andexponent
, paired together asresult = significand * 2**exponent
.
Generating Arrays of Random Numbers¶
PyOpenCL includes and uses some of the Random123 random number generators by D.E. Shaw Research. In addition to being usable through the convenience functions above, they are available in any piece of code compiled through PyOpenCL by:
#include <pyopencl-random123/philox.cl>
#include <pyopencl-random123/threefry.cl>
See the Philox source and the Threefry source for some documentation if you’re planning on using Random123 directly.
- class pyopencl.clrandom.PhiloxGenerator(context, key=None, counter=None, seed=None)[source]¶
Added in version 2016.2.
- uniform(*args, **kwargs)[source]¶
Make a new empty array, apply
fill_uniform()
to it.
- fill_normal(ary, mu=0, sigma=1, queue=None)[source]¶
Fill ary with normally distributed numbers with mean mu and standard deviation sigma.
- normal(*args, **kwargs)[source]¶
Make a new empty array, apply
fill_normal()
to it.
- class pyopencl.clrandom.ThreefryGenerator(context, key=None, counter=None, seed=None)[source]¶
Added in version 2016.2.
- uniform(*args, **kwargs)[source]¶
Make a new empty array, apply
fill_uniform()
to it.
- fill_normal(ary, mu=0, sigma=1, queue=None)[source]¶
Fill ary with normally distributed numbers with mean mu and standard deviation sigma.
- normal(*args, **kwargs)[source]¶
Make a new empty array, apply
fill_normal()
to it.