h**t
发帖数: 1678 | 1 现在有一个1MM rows x 13 columns/variabls的数据， 作multinomial, 按说也不大，可是很慢啊。这个是因为作multinomial 的原因吗？作linear 好像就会快很多。。。
R 可以处理最大的数据是多大？有package 推荐吗？ 谢谢！ |
u*****3
发帖数: 796 | |
u*****3
发帖数: 796 | |
h**t
发帖数: 1678 | 4 非常非常感谢！！！
2^32-1 这个数是怎莫来得呢？谢谢！ |
z****g
发帖数: 1978 | 5 depends on the memory. The default compilation is 30M or 40M. You can
compile the source code yourself and set a specific MACRO for the memory
size. Current R source code is stable for 32 bit compiler, and the memory
upper limit is the same as the standard 32 bit application, which is 2G.
64bit compilation is not recommended as there may be unknown type bug.
Also, there is no sparse matrix implementation for R. For further
optimization, you can try to build the ATLAS DLL on your machine to optimize
with your hardware setting. |
o*****h
发帖数: 293 | 6 32位机器寻址空间的上限
【在 h**t 的大作中提到】
: 非常非常感谢！！！
: 2^32-1 这个数是怎莫来得呢？谢谢！
|
h**t
发帖数: 1678 | 7 非常感谢！！！will keep this in mind
optimize
【在 z****g 的大作中提到】
: depends on the memory. The default compilation is 30M or 40M. You can
: compile the source code yourself and set a specific MACRO for the memory
: size. Current R source code is stable for 32 bit compiler, and the memory
: upper limit is the same as the standard 32 bit application, which is 2G.
: 64bit compilation is not recommended as there may be unknown type bug.
: Also, there is no sparse matrix implementation for R. For further
: optimization, you can try to build the ATLAS DLL on your machine to optimize
: with your hardware setting.
|
h**t
发帖数: 1678 | 8 谢谢啦！ 我也明白了。。。
【在 o*****h 的大作中提到】
: 32位机器寻址空间的上限
|
f******d
发帖数: 537 | 9 64-bit version has been around for a while. It still has bugs?
Also what about "SparseM" package for sparse matrix?
optimize
【在 z****g 的大作中提到】
: depends on the memory. The default compilation is 30M or 40M. You can
: compile the source code yourself and set a specific MACRO for the memory
: size. Current R source code is stable for 32 bit compiler, and the memory
: upper limit is the same as the standard 32 bit application, which is 2G.
: 64bit compilation is not recommended as there may be unknown type bug.
: Also, there is no sparse matrix implementation for R. For further
: optimization, you can try to build the ATLAS DLL on your machine to optimize
: with your hardware setting.
|
S******y
发帖数: 1123 | 10 For simple descriptive statistics - 32位机器寻址空间的上限;
For statistical packages (involving matrix computation), it depends...
usually much smaller data size...
Just my 2 cents... |
f***a
发帖数: 329 | 11 ff, bigmemory, bigGLM may help |
