From 4c59fd732808b593cc600709fc27cfb636215ec4 Mon Sep 17 00:00:00 2001
From: Andreas Kloeckner <inform@tiker.net>
Date: Sat, 8 Jun 2013 13:43:55 -0400
Subject: [PATCH] PEP8 clrandom
---
pyopencl/clrandom.py | 71 ++++++++++++++++++++------------------------
1 file changed, 33 insertions(+), 38 deletions(-)
diff --git a/pyopencl/clrandom.py b/pyopencl/clrandom.py
index 04d3a59a..04ebe99e 100644
--- a/pyopencl/clrandom.py
+++ b/pyopencl/clrandom.py
@@ -23,6 +23,7 @@ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
"""
+
# {{{ documentation
__doc__ = u"""
@@ -33,7 +34,8 @@ available in any piece of code compiled through PyOpenCL by::
#include <pyopencl-ranluxcl.cl>
-See the `source <https://github.com/inducer/pyopencl/blob/master/src/cl/pyopencl-ranluxcl.cl>`_
+See the `source
+<https://github.com/inducer/pyopencl/blob/master/src/cl/pyopencl-ranluxcl.cl>`_
for some documentation if you're planning on using RANLUXCL directly.
The RANLUX generator is described in the following two articles. If you use the
@@ -58,11 +60,6 @@ from pytools import memoize_method
import numpy as np
-
-
-
-
-
class RanluxGenerator(object):
"""
.. versionadded:: 2011.2
@@ -73,9 +70,10 @@ class RanluxGenerator(object):
.. attribute:: nskip
- nskip is an integer which can (optionally) be defined in the kernel code
- as RANLUXCL_NSKIP. If this is done the generator will be faster for luxury setting
- 0 and 1, or when the p-value is manually set to a multiple of 24.
+ nskip is an integer which can (optionally) be defined in the kernel
+ code as RANLUXCL_NSKIP. If this is done the generator will be faster
+ for luxury setting 0 and 1, or when the p-value is manually set to a
+ multiple of 24.
"""
def __init__(self, queue, num_work_items=None,
@@ -83,19 +81,21 @@ class RanluxGenerator(object):
use_legacy_init=False, max_work_items=None):
"""
:param queue: :class:`pyopencl.CommandQueue`, only used for initialization
- :param luxury: the "luxury value" of the generator, and should be 0-4, where 0 is fastest
- and 4 produces the best numbers. It can also be >=24, in which case it directly
- sets the p-value of RANLUXCL.
- :param num_work_items: is the number of generators to initialize, usually corresponding
- to the number of work-items in the NDRange RANLUXCL will be used with.
- May be `None`, in which case a default value is used.
- :param max_work_items: should reflect the maximum number of work-items that will be used
- on any parallel instance of RANLUXCL. So for instance if we are launching 5120
- work-items on GPU1 and 10240 work-items on GPU2, GPU1's RANLUXCLTab would be
- generated by calling ranluxcl_intialization with numWorkitems = 5120 while
- GPU2's RANLUXCLTab would use numWorkitems = 10240. However maxWorkitems must
- be at least 10240 for both GPU1 and GPU2, and it must be set to the same value
- for both. (may be `None`)
+ :param luxury: the "luxury value" of the generator, and should be 0-4,
+ where 0 is fastest and 4 produces the best numbers. It can also be
+ >=24, in which case it directly sets the p-value of RANLUXCL.
+ :param num_work_items: is the number of generators to initialize,
+ usually corresponding to the number of work-items in the NDRange
+ RANLUXCL will be used with. May be `None`, in which case a default
+ value is used.
+ :param max_work_items: should reflect the maximum number of work-items
+ that will be used on any parallel instance of RANLUXCL. So for
+ instance if we are launching 5120 work-items on GPU1 and 10240
+ work-items on GPU2, GPU1's RANLUXCLTab would be generated by
+ calling ranluxcl_intialization with numWorkitems = 5120 while
+ GPU2's RANLUXCLTab would use numWorkitems = 10240. However
+ maxWorkitems must be at least 10240 for both GPU1 and GPU2, and it
+ must be set to the same value for both. (may be `None`)
.. versionchanged:: 2013.1
Added default value for `num_work_items`.
@@ -112,7 +112,7 @@ class RanluxGenerator(object):
if seed is None:
from time import time
- seed = int(time()*1e6) % 2<<30
+ seed = int(time()*1e6) % 2 << 30
self.context = queue.context
self.luxury = luxury
@@ -130,7 +130,8 @@ class RanluxGenerator(object):
#include <pyopencl-ranluxcl.cl>
- kernel void init_ranlux(unsigned seeds, global ranluxcl_state_t *ranluxcltab)
+ kernel void init_ranlux(unsigned seeds,
+ global ranluxcl_state_t *ranluxcltab)
{
if (get_global_id(0) < %(num_work_items)d)
ranluxcl_initialization(seeds, ranluxcltab);
@@ -180,7 +181,8 @@ class RanluxGenerator(object):
lines.append("#define RANLUXCL_USE_LEGACY_INITIALIZATION")
if self.max_work_items:
- lines.append("#define RANLUXCL_MAXWORKITEMS %d" % self.max_work_items)
+ lines.append(
+ "#define RANLUXCL_MAXWORKITEMS %d" % self.max_work_items)
return "\n".join(lines)
@@ -215,8 +217,7 @@ class RanluxGenerator(object):
c_type = "int"
rng_expr = ("(shift "
"+ convert_int4(scale * gen) "
- "+ convert_int4((scale / (1<<24)) * gen))"
- )
+ "+ convert_int4((scale / (1<<24)) * gen))")
else:
raise TypeError("unsupported RNG data type '%s'" % dtype)
@@ -248,7 +249,9 @@ class RanluxGenerator(object):
unsigned long idx = get_global_id(0)*4;
while (idx + 4 < out_size)
{
- vstore4(GET_RANDOM_NUM(RANLUX_FUNC(&ranluxclstate)), idx >> 2, output);
+ vstore4(
+ GET_RANDOM_NUM(RANLUX_FUNC(&ranluxclstate)),
+ idx >> 2, output);
idx += 4*NUM_WORKITEMS;
}
@@ -355,11 +358,8 @@ class RanluxGenerator(object):
efficiency.
"""
- self.get_sync_kernel()(queue, (self.num_work_items,), self.wg_size, self.state.data)
-
-
-
-
+ self.get_sync_kernel()(queue, (self.num_work_items,),
+ self.wg_size, self.state.data)
@first_arg_dependent_memoize
@@ -369,8 +369,6 @@ def _get_generator(queue, luxury=None):
return gen
-
-
def fill_rand(result, queue=None, luxury=4, a=0, b=1):
"""Fill *result* with random values of `dtype` in the range [0,1).
"""
@@ -380,8 +378,6 @@ def fill_rand(result, queue=None, luxury=4, a=0, b=1):
gen.fill_uniform(result, a=a, b=b)
-
-
def rand(queue, shape, dtype, luxury=None, a=0, b=1):
"""Return an array of `shape` filled with random values of `dtype`
in the range [a,b).
@@ -394,5 +390,4 @@ def rand(queue, shape, dtype, luxury=None, a=0, b=1):
return result
-
# vim: filetype=pyopencl:foldmethod=marker
--
GitLab