Hi, List and Laurent,
For the chisq function I tried to used through rpy2, I just did
more test with the exact code as follows:
----------------------------------------------
import rpy2.robjects as robjects
p1dist = []
p2dist = []
num = 17
for x in range(num):
p1dist.append(x)
p2dist.append(x-1)
print p1dist
print p2dist
rp1dist = robjects.FloatVector(p1dist)
rp2dist = robjects.FloatVector(p2dist)
res = robjects.r['chisq.test'](rp1dist, rp2dist)
print(str(res.r['p.value']).split()[-1])
-----------------------------------------------------
When I set num to 17, it works
"Warning message: In function (x, y = NULL, correct = TRUE, p =
rep(1/length(x), length(x)), :
Chi-squared approximation may be incorrect
0.2350833"
However, if I set num to 18 (add one more element to the two vectors to
be compared by chisq.test), it doesn't work.
The error message is similar to my previous post:
File "/usr/lib64/python2.6/site-packages/rpy2/robjects/__init__.py",
line 422, in __call__
res = super(RFunction, self).__call__(*new_args, **new_kwargs)
rinterface.RRuntimeError: Error in names(dimnames(x)) <- DNAME :
'names' attribute [4] must be the same length as the vector [2]
Could this mean that rpy2 doesn't take a long vector or I made some
stupid mistake?
Thanks!
Hao
> Hi Hao,
>
> The exact example that triggers the error may matter.
>
> I just tried the following with rpy2-2.1dev and it worked.
>
> x = robjects.FloatVector((1,2,3))
> y = robjects.FloatVector((2,3,4))
>
> res = robjects.r['chisq.test'](x, y)
>
> I only get the following
> Warning message:
> In function (x, y = NULL, correct = TRUE, p = rep(1/length(x),
> length(x)), :
> Chi-squared approximation may be incorrect
>
> L.
>
>
>
> Hao Fan wrote:
>> Hi, list
>>
>> I would like to calculate the fitness of two histograms, so I
>> tried to use the chisq.test() in R through rpy2.
>> The python code I have is as follows:
>> -------------------
>> import rpy2.robjects as robjects
>>
>> p1dist = [X1, X2, ... Xm]
>> p2dist = [Y1, Y2, ... Ym]
>>
>> rp1dist = robjects.FloatVector(p1dist)
>> rp2dist = robjects.FloatVector(p2dist)
>>
>> robjects.r['chisq.test'](rp1dist, rp2dist)
>> -------------------
>>
>> Then I got the following error:
>> File "/usr/lib64/python2.6/site-packages/rpy2/robjects/__init__.py",
>> in __call__
>> res = super(RFunction, self).__call__(*new_args, **new_kwargs)
>> rinterface.RRuntimeError: Error in names(dimnames(x)) <- DNAME :
>> 'names' attribute [62] must be the same length as the vector [2]
>>
>>
>> I also tried rpy_classic, so my code change to the follows:
>> --------------------
>> import rpy2.robjects as robjects
>> import rpy2.rpy_classic as rpy
>> rpy.set_default_mode(rpy.NO_CONVERSION)
>>
>> p1dist = [X1, X2, ... Xm]
>> p2dist = [Y1, Y2, ... Ym]
>>
>> rp1dist = robjects.FloatVector(p1dist)
>> rp2dist = robjects.FloatVector(p2dist)
>>
>> robjects.r['chisq.test'](rp1dist, rp2dist)
>> -----------------
>>
>> Then I got similar error:
>> File "/usr/lib64/python2.6/site-packages/rpy2/rpy_classic.py", line
>> 214, in __call__
>> res = self.__sexp(*args_r, **kwargs_r)
>> rinterface.RRuntimeError: Error in names(dimnames(x)) <- DNAME :
>> 'names' attribute [62] must be the same length as the vector [2]
>>
>>
>> I feel this problem is due to my wrong use of the R vector, but I
>> can not find it. Any hints will be mostly appreciated!
>> Thanks a lot in advance!
>>
>>
>> Hao
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
--
Hao Fan, PhD
Department of Biopharmaceutical Sciences
Mission Bay, Byers Hall
1700 4th Street, Suite 501
University of California, San Francisco
San Francisco, CA 94158-2330
Tel +1 (415) 514-4258, 4232, 4233
Fax +1 (415) 514-4231, 4234
------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
rpy-list mailing list
rpy-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rpy-list
