为什么 floatX 的标志会影响 Theano 中是否使用 GPU?
Why does the floatX's flag impact whether GPU is used in Theano?
我正在使用 GPU 测试 Theano script provided in the tutorial for that purpose:
# Start gpu_test.py
# From http://deeplearning.net/software/theano/tutorial/using_gpu.html#using-gpu
from theano import function, config, shared, sandbox
import theano.tensor as T
import numpy
import time
vlen = 10 * 30 * 768 # 10 x #cores x # threads per core
iters = 1000
rng = numpy.random.RandomState(22)
x = shared(numpy.asarray(rng.rand(vlen), config.floatX))
f = function([], T.exp(x))
print(f.maker.fgraph.toposort())
t0 = time.time()
for i in xrange(iters):
r = f()
t1 = time.time()
print("Looping %d times took %f seconds" % (iters, t1 - t0))
print("Result is %s" % (r,))
if numpy.any([isinstance(x.op, T.Elemwise) for x in f.maker.fgraph.toposort()]):
print('Used the cpu')
else:
print('Used the gpu')
# End gpu_test.py
如果我指定 floatX=float32
,它在 GPU 上运行:
francky@here:/fun$ THEANO_FLAGS='mode=FAST_RUN,device=gpu2,floatX=float32' python gpu_test.py
Using gpu device 2: GeForce GTX TITAN X (CNMeM is disabled)
[GpuElemwise{exp,no_inplace}(<CudaNdarrayType(float32, vector)>), HostFromGpu(Gp
Looping 1000 times took 1.458473 seconds
Result is [ 1.23178029 1.61879349 1.52278066 ..., 2.20771813 2.29967761
1.62323296]
Used the gpu
如果我不指定 floatX=float32
,它会在 CPU:
上运行
francky@here:/fun$ THEANO_FLAGS='mode=FAST_RUN,device=gpu2'
Using gpu device 2: GeForce GTX TITAN X (CNMeM is disabled)
[Elemwise{exp,no_inplace}(<TensorType(float64, vector)>)]
Looping 1000 times took 3.086261 seconds
Result is [ 1.23178032 1.61879341 1.52278065 ..., 2.20771815 2.29967753
1.62323285]
Used the cpu
如果我指定 floatX=float64
,它会在 CPU:
上运行
francky@here:/fun$ THEANO_FLAGS='mode=FAST_RUN,device=gpu2,floatX=float64' python gpu_test.py
Using gpu device 2: GeForce GTX TITAN X (CNMeM is disabled)
[Elemwise{exp,no_inplace}(<TensorType(float64, vector)>)]
Looping 1000 times took 3.148040 seconds
Result is [ 1.23178032 1.61879341 1.52278065 ..., 2.20771815 2.29967753
1.62323285]
Used the cpu
为什么 floatX
标志会影响 Theano 中是否使用 GPU?
我使用:
- Theano 0.7.0(根据
pip freeze
),
- Python 2.7.6 64 位(根据
import platform; platform.architecture()
),
- Nvidia-smi 361.28(根据
nvidia-smi
),
- CUDA 7.5.17(根据
nvcc --version
),
- GeForce GTX Titan X(根据
nvidia-smi
),
- Ubuntu 14.04.4 LTS x64(根据
lsb_release -a
和 uname -i
)。
我阅读了 floatX
上的文档,但没有帮助。它只是说:
config.floatX
String value: either ‘float64’ or ‘float32’
Default: ‘float64’
This sets the default dtype returned by tensor.matrix(),
tensor.vector(), and similar functions. It also sets the default
theano bit width for arguments passed as Python floating-point
numbers.
据我所知,这是因为他们还没有为 GPU 实现 float64。
http://deeplearning.net/software/theano/tutorial/using_gpu.html :
Only computations with float32 data-type can be accelerated. Better support for float64 is expected in upcoming hardware but float64 computations are still relatively slow (Jan 2010).
来自 http://deeplearning.net/software/theano/tutorial/using_gpu.html#gpuarray-backend 我读到可以在 GPU 上执行 float64 计算,但是你必须从源代码安装 libgpuarray
。
我设法安装了它,见 this script, I used virtualenv,你甚至不需要 sudo
。
安装后,您可以使用带有 config flag device=gpu
的旧后端和带有 device=cuda
.
的新后端
新后端可以执行 64 位计算,但对我来说它的工作方式不同。一些操作停止工作。 ABSOLUTELY NO WARRANTY, to the extent permitted by applicable law
:)
我正在使用 GPU 测试 Theano script provided in the tutorial for that purpose:
# Start gpu_test.py
# From http://deeplearning.net/software/theano/tutorial/using_gpu.html#using-gpu
from theano import function, config, shared, sandbox
import theano.tensor as T
import numpy
import time
vlen = 10 * 30 * 768 # 10 x #cores x # threads per core
iters = 1000
rng = numpy.random.RandomState(22)
x = shared(numpy.asarray(rng.rand(vlen), config.floatX))
f = function([], T.exp(x))
print(f.maker.fgraph.toposort())
t0 = time.time()
for i in xrange(iters):
r = f()
t1 = time.time()
print("Looping %d times took %f seconds" % (iters, t1 - t0))
print("Result is %s" % (r,))
if numpy.any([isinstance(x.op, T.Elemwise) for x in f.maker.fgraph.toposort()]):
print('Used the cpu')
else:
print('Used the gpu')
# End gpu_test.py
如果我指定 floatX=float32
,它在 GPU 上运行:
francky@here:/fun$ THEANO_FLAGS='mode=FAST_RUN,device=gpu2,floatX=float32' python gpu_test.py
Using gpu device 2: GeForce GTX TITAN X (CNMeM is disabled)
[GpuElemwise{exp,no_inplace}(<CudaNdarrayType(float32, vector)>), HostFromGpu(Gp
Looping 1000 times took 1.458473 seconds
Result is [ 1.23178029 1.61879349 1.52278066 ..., 2.20771813 2.29967761
1.62323296]
Used the gpu
如果我不指定 floatX=float32
,它会在 CPU:
francky@here:/fun$ THEANO_FLAGS='mode=FAST_RUN,device=gpu2'
Using gpu device 2: GeForce GTX TITAN X (CNMeM is disabled)
[Elemwise{exp,no_inplace}(<TensorType(float64, vector)>)]
Looping 1000 times took 3.086261 seconds
Result is [ 1.23178032 1.61879341 1.52278065 ..., 2.20771815 2.29967753
1.62323285]
Used the cpu
如果我指定 floatX=float64
,它会在 CPU:
francky@here:/fun$ THEANO_FLAGS='mode=FAST_RUN,device=gpu2,floatX=float64' python gpu_test.py
Using gpu device 2: GeForce GTX TITAN X (CNMeM is disabled)
[Elemwise{exp,no_inplace}(<TensorType(float64, vector)>)]
Looping 1000 times took 3.148040 seconds
Result is [ 1.23178032 1.61879341 1.52278065 ..., 2.20771815 2.29967753
1.62323285]
Used the cpu
为什么 floatX
标志会影响 Theano 中是否使用 GPU?
我使用:
- Theano 0.7.0(根据
pip freeze
), - Python 2.7.6 64 位(根据
import platform; platform.architecture()
), - Nvidia-smi 361.28(根据
nvidia-smi
), - CUDA 7.5.17(根据
nvcc --version
), - GeForce GTX Titan X(根据
nvidia-smi
), - Ubuntu 14.04.4 LTS x64(根据
lsb_release -a
和uname -i
)。
我阅读了 floatX
上的文档,但没有帮助。它只是说:
config.floatX
String value: either ‘float64’ or ‘float32’
Default: ‘float64’This sets the default dtype returned by tensor.matrix(), tensor.vector(), and similar functions. It also sets the default theano bit width for arguments passed as Python floating-point numbers.
据我所知,这是因为他们还没有为 GPU 实现 float64。
http://deeplearning.net/software/theano/tutorial/using_gpu.html :
Only computations with float32 data-type can be accelerated. Better support for float64 is expected in upcoming hardware but float64 computations are still relatively slow (Jan 2010).
来自 http://deeplearning.net/software/theano/tutorial/using_gpu.html#gpuarray-backend 我读到可以在 GPU 上执行 float64 计算,但是你必须从源代码安装 libgpuarray
。
我设法安装了它,见 this script, I used virtualenv,你甚至不需要 sudo
。
安装后,您可以使用带有 config flag device=gpu
的旧后端和带有 device=cuda
.
新后端可以执行 64 位计算,但对我来说它的工作方式不同。一些操作停止工作。 ABSOLUTELY NO WARRANTY, to the extent permitted by applicable law
:)