提交 151d3f44 authored 作者: Frederic Bastien's avatar Frederic Bastien

docstring fix

上级 7cf15f44
......@@ -498,8 +498,8 @@ import theano and print the config variable, as in:
The sched parameter passed for context creation to pygpu. With
CUDA, using "multi" mean using the parameter
cudaDeviceScheduleYield. This is useful to lower the CPU overhead
when waiting for GPU. One user found that it speed up its other
process that was doing data augmentation.
when waiting for GPU. One user found that it speeds up his other
processes that was doing data augmentation.
.. attribute:: config.gpuarray.single_stream
......
......@@ -244,13 +244,12 @@ AddConfigVar('gpuarray.preallocate',
AddConfigVar('gpuarray.sched',
"""The sched parameter passed for context creation to pygpu.
With CUDA, using "multi" mean using the parameter
With CUDA, using "multi" is equivalent to using the parameter
cudaDeviceScheduleYield. This is useful to lower the
CPU overhead when waiting for GPU. One user found that it
speed up its other process that was doing data augmentation.
speeds up his other processes that was doing data augmentation.
""",
EnumStr("default", "multi", "single"),
in_c_key=True)
EnumStr("default", "multi", "single"))
AddConfigVar('gpuarray.single_stream',
"""
......
Markdown 格式
0%
您添加了 0 到此讨论。请谨慎行事。
请先完成此评论的编辑!
注册 或者 后发表评论