Skip to content

P
pytensor
提交 954b07ab authored 作者: Ian Goodfellow's avatar Ian Goodfellow
浏览文件
操作

merged

上级 e06a39ea 5d1217d7
隐藏空白字符变更
内嵌 并排
正在显示 包含 27 行增加 和 15 行删除
theano/misc/check_blas.py
浏览文件 @ 954b07ab
...@@ -91,38 +91,48 @@ if __name__ == "__main__": ...@@ -91,38 +91,48 @@ if __name__ == "__main__":
Cpu tested: Xeon E5345(2.33Ghz, 8M L2 cache, 1333Mhz FSB), Xeon E5430(2.66Ghz, 12M L2 cache, 1333Mhz FSB), Cpu tested: Xeon E5345(2.33Ghz, 8M L2 cache, 1333Mhz FSB), Xeon E5430(2.66Ghz, 12M L2 cache, 1333Mhz FSB),
Xeon E5450(3Ghz, 12M L2 cache, 1333Mhz FSB), Xeon X5560(2.8Ghz, 12M L2 cache, 6.4GT/s QPI, hyper-threads enabled?) Xeon E5450(3Ghz, 12M L2 cache, 1333Mhz FSB), Xeon X5560(2.8Ghz, 12M L2 cache, 6.4GT/s QPI, hyper-threads enabled?)
Core 2 E8500, Core i7 930(2.8Ghz, hyper-threads enabled), Core i7 950(3.07GHz, hyper-threads enabled) Core 2 E8500, Core i7 930(2.8Ghz, hyper-threads enabled), Core i7 950(3.07GHz, hyper-threads enabled)
Xeon X5550(2.67GHz, 8M l2 cache?, hyper-threads enabled)
Lib tested: Lib tested:
* numpy with ATLAS from distribution(FC9) package (1 thread) * numpy with ATLAS from distribution(FC9) package (1 thread)
* manually compiled numpy and ATLAS with 2 threads * manually compiled numpy and ATLAS with 2 threads
* goto 1.26 with 1, 2, 4 and 8 threads. * goto 1.26 with 1, 2, 4 and 8 threads.
Xeon Xeon Xeon Core2 i7 i7 Xeon * goto2 1.13 compiled with multiple thread enabled.
lib/nb threads E5345 E5430 E5450 E8500 930 950 X5560
Xeon Xeon Xeon Core2 i7 i7 Xeon Xeon
lib/nb threads E5345 E5430 E5450 E8500 930 950 X5560 X5550
numpy 1.3.0 blas 775.92s
numpy_FC9_atlas/1 39.2s 35.0s 30.7s 29.6s 21.5s 19.60s numpy_FC9_atlas/1 39.2s 35.0s 30.7s 29.6s 21.5s 19.60s
goto/1 18.7s 16.1s 14.2s 13.7s 16.1s 14.67s goto/1 18.7s 16.1s 14.2s 13.7s 16.1s 14.67s
numpy_MAN_atlas/2 12.0s 11.6s 10.2s 9.2s 9.0s numpy_MAN_atlas/2 12.0s 11.6s 10.2s 9.2s 9.0s
goto/2 9.5s 8.1s 7.1s 7.3s 8.1s 7.4s goto/2 9.5s 8.1s 7.1s 7.3s 8.1s 7.4s
goto/4 4.9s 4.4s 3.7s - 4.1s 3.8s goto/4 4.9s 4.4s 3.7s - 4.1s 3.8s
goto/8 2.7s 2.4s 2.0s - 4.1s 3.8s goto/8 2.7s 2.4s 2.0s - 4.1s 3.8s
openblas/1 14.04s openblas/1 14.04s
openblas/2 7.16s openblas/2 7.16s
openblas/4 3.71s openblas/4 3.71s
openblas/8 3.70s openblas/8 3.70s
mkl 11.0.083/1 7.97s
mkl 10.2.2.025/1 13.7s mkl 10.2.2.025/1 13.7s
mkl 10.2.2.025/2 7.6s mkl 10.2.2.025/2 7.6s
mkl 10.2.2.025/4 4.0s mkl 10.2.2.025/4 4.0s
mkl 10.2.2.025/8 2.0s mkl 10.2.2.025/8 2.0s
mkl 11.0.083/1 7.97s goto2 1.13/1 14.37s
goto2 1.13/2 7.26s
goto2 1.13/4 3.70s
goto2 1.13/8 1.94s
goto2 1.13/16 3.16s
Test time in float32 with cuda 3.0.14 Test time in float32 with cuda 3.0.14
(cuda version 3.2RC and up are supposed to have faster gemm on the GTX4?? card) (cuda version 3.2RC and up are supposed to have faster gemm on the GTX4?? card)
cpu/cuda version gpu/cuda version
GTX580/3.2 0.20s GTX580/3.2 0.20s
GTX480/3.2 0.24s GTX480/3.2 0.24s
GTX480/3.0 0.27s GTX480/3.0 0.27s
GTX470/3.2 0.29s GTX470/3.2 0.29s
M2070/3.2 0.32s
GTX470/3.0 0.34s GTX470/3.0 0.34s
GTX285/3.0 0.40s GTX285/3.0 0.40s
GT220/3.2RC 5.15s GT220/3.2RC 5.15s
......
theano/sandbox/cuda/cuda_ndarray.cu
浏览文件 @ 954b07ab
...@@ -636,7 +636,9 @@ PyObject * CudaNdarray_Reshape(CudaNdarray * self, PyObject * shape) ...@@ -636,7 +636,9 @@ PyObject * CudaNdarray_Reshape(CudaNdarray * self, PyObject * shape)
} }
if (rval_size==0) if (rval_size==0)
{ {
return CudaNdarray_NewDims(rval_nd, rval_dims); PyObject * rval = CudaNdarray_NewDims(rval_nd, rval_dims);
free(rval_dims);
return rval;
} }
if(CudaNdarray_is_c_contiguous(self)) if(CudaNdarray_is_c_contiguous(self))
......
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