numpy-discussion

[Numpy-discussion] simple question

[Mathew Y. <my...@jp...>]

What is the typical way of doing the following
starting with a 0 matrix, set all values to 1 when a certain condition 
is met, set to -1 when another condition is met, left alone if neither 
condition is met.

Mathew

Re: [Numpy-discussion] simple question

[Keith G. <kwg...@gm...>]

On 7/5/06, Mathew Yeates <my...@jp...> wrote:
> What is the typical way of doing the following
> starting with a 0 matrix, set all values to 1 when a certain condition
> is met, set to -1 when another condition is met, left alone if neither
> condition is met.

This works on recent versions of numpy:

>> x = asmatrix(zeros((2,2)))

>> x

matrix([[0, 0],
       [0, 0]])

>> y = asmatrix(rand(2,2))

>> y

matrix([[ 0.85219404,  0.48311427],
       [ 0.41026966,  0.2184193 ]])

>> x[y > 0.5] = 1

>> x[y < 0.5] = -1

>> x

matrix([[ 1, -1],
       [-1, -1]])

Re: [Numpy-discussion] simple question

[Mathew Y. <my...@jp...>]

Not working.
A[row,all_dates == 10] = -1 where all_dates is a matrix with column 
length of 14 [[960111,..,..
and A is a matrix with same column length

I get
IndexError: arrays used as indices must be of integer type

when I print out all_dates == 10
I get
[True True True True True True True True True False False False True True]]

I experimented with "<" instead of "==" but I still get boolean values 
as indices.

Any help?
Mathew



Keith Goodman wrote:
> On 7/5/06, Mathew Yeates <my...@jp...> wrote:
>> What is the typical way of doing the following
>> starting with a 0 matrix, set all values to 1 when a certain condition
>> is met, set to -1 when another condition is met, left alone if neither
>> condition is met.
>
> This works on recent versions of numpy:
>
>>> x = asmatrix(zeros((2,2)))
>
>>> x
>
> matrix([[0, 0],
>       [0, 0]])
>
>>> y = asmatrix(rand(2,2))
>
>>> y
>
> matrix([[ 0.85219404,  0.48311427],
>       [ 0.41026966,  0.2184193 ]])
>
>>> x[y > 0.5] = 1
>
>>> x[y < 0.5] = -1
>
>>> x
>
> matrix([[ 1, -1],
>       [-1, -1]])
>

Re: [Numpy-discussion] simple question

[Travis O. <oli...@ee...>]

Mathew Yeates wrote:

>Not working.
>A[row,all_dates == 10] = -1 where all_dates is a matrix with column 
>length of 14 [[960111,..,..
>and A is a matrix with same column length
>
>I get
>IndexError: arrays used as indices must be of integer type
>
>when I print out all_dates == 10
>I get
>[True True True True True True True True True False False False True True]]
>
>I experimented with "<" instead of "==" but I still get boolean values 
>as indices.
>
>Any help?
>  
>
What version are you using?  Can you give an example that shows the 
error.  It's hard to guess the type of all the variables.  The following 
works for me.

import numpy
print numpy.__version__
A = numpy.matrix(rand(3,14))
all_dates = array([10,10,1,10,1,10,0,10,0,10,0,1,10,1])
row = 2
A[row, all_dates == 10]





-Trvis

Re: [Numpy-discussion] simple question

[Keith G. <kwg...@gm...>]

On 7/6/06, Travis Oliphant <oli...@ee...> wrote:
> Mathew Yeates wrote:
>
> >Not working.
> >A[row,all_dates == 10] = -1 where all_dates is a matrix with column
> >length of 14 [[960111,..,..
> >and A is a matrix with same column length
> >
> >I get
> >IndexError: arrays used as indices must be of integer type
> >
> >when I print out all_dates == 10
> >I get
> >[True True True True True True True True True False False False True True]]
> >
> >I experimented with "<" instead of "==" but I still get boolean values
> >as indices.
> >
> >Any help?
> >
> >
> What version are you using?  Can you give an example that shows the
> error.  It's hard to guess the type of all the variables.  The following
> works for me.
>
> import numpy
> print numpy.__version__
> A = numpy.matrix(rand(3,14))
> all_dates = array([10,10,1,10,1,10,0,10,0,10,0,1,10,1])
> row = 2
> A[row, all_dates == 10]


This is what NASA is doing (and what I would like to do):

>> A[row, asmatrix(all_dates == 10)]
---------------------------------------------------------------------------
exceptions.ValueError                                Traceback (most
recent call last)

/home/kwg/<ipython console>

/usr/local/lib/python2.4/site-packages/numpy/core/defmatrix.py in
__getitem__(self, index)
    122
    123     def __getitem__(self, index):
--> 124         out = N.ndarray.__getitem__(self, index)
    125         # Need to swap if slice is on first index
    126         # or there is an integer on the second

ValueError: too many indices for array

[Numpy-discussion] More zaniness Re: simple question

[Mathew Y. <my...@jp...>]

okay, I went back to the binary windows distrib. Based on Keths code I wrote

 >> print numpy.asmatrix(all_dates == start_dates[row],dtype=int)
[[0 0 0 0 0 0 0 0 0 0 0 1 0 0]]
 >> [row,numpy.asmatrix(all_dates == start_dates[row],dtype=int)] = -1
 >> print A[row,:]
[[-1. -1.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.]]

huh? It set the first 2 elements and not the 12'th!!

Mathew

Re: [Numpy-discussion] More zaniness Re: simple question

[Keith G. <kwg...@gm...>]

On 7/6/06, Mathew Yeates <my...@jp...> wrote:
> okay, I went back to the binary windows distrib. Based on Keths code I wrote
>
>  >> print numpy.asmatrix(all_dates == start_dates[row],dtype=int)
> [[0 0 0 0 0 0 0 0 0 0 0 1 0 0]]
>  >> [row,numpy.asmatrix(all_dates == start_dates[row],dtype=int)] = -1
>  >> print A[row,:]
> [[-1. -1.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.]]
>
> huh? It set the first 2 elements and not the 12'th!!

You are assigning -1 to column 0 thirteen times and assinging -1 to
column 1 once.

For now, until boolean indexing works with matrices, I would just use
brute force:

A[row,where(all_dates.A == 10)[0]]

Or you can do all rows at once with

A[all_dates== 10]

where all_dates is the same size as A.

Re: [Numpy-discussion] More zaniness Re: simple question

[Travis O. <oli...@ee...>]

Mathew Yeates wrote:

>okay, I went back to the binary windows distrib. Based on Keths code I wrote
>
> >> print numpy.asmatrix(all_dates == start_dates[row],dtype=int)
>[[0 0 0 0 0 0 0 0 0 0 0 1 0 0]]
> >> [row,numpy.asmatrix(all_dates == start_dates[row],dtype=int)] = -1
> >> print A[row,:]
>[[-1. -1.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.]]
>
>huh? It set the first 2 elements and not the 12'th!!
>  
>

Indexing has changed in SVN python, but in general, index matrices are 
not what you want because the dimensionality of the index arrays means 
something and matrices have two dimensions.   So use arrays for indexing.


-Travis

Re: [Numpy-discussion] simple question

[Mathew Y. <my...@jp...>]

The very example you give produces
IndexError: arrays used as indices must be of integer type

this is with 0.9.8

Also .....while your example says "rand" I had to say numpy.rand
This is on WindowsXP

Mathew


Travis Oliphant wrote:
> Mathew Yeates wrote:
>
>> Not working.
>> A[row,all_dates == 10] = -1 where all_dates is a matrix with column 
>> length of 14 [[960111,..,..
>> and A is a matrix with same column length
>>
>> I get
>> IndexError: arrays used as indices must be of integer type
>>
>> when I print out all_dates == 10
>> I get
>> [True True True True True True True True True False False False True 
>> True]]
>>
>> I experimented with "<" instead of "==" but I still get boolean 
>> values as indices.
>>
>> Any help?
>>  
>>
> What version are you using?  Can you give an example that shows the 
> error.  It's hard to guess the type of all the variables.  The 
> following works for me.
>
> import numpy
> print numpy.__version__
> A = numpy.matrix(rand(3,14))
> all_dates = array([10,10,1,10,1,10,0,10,0,10,0,1,10,1])
> row = 2
> A[row, all_dates == 10]
>
>
>
>
>
> -Trvis
>
>

Re: [Numpy-discussion] simple question

[Mathew Y. <my...@jp...>]

ohhhh. I was looking at using "where"



Keith Goodman wrote:
> On 7/5/06, Mathew Yeates <my...@jp...> wrote:
>> What is the typical way of doing the following
>> starting with a 0 matrix, set all values to 1 when a certain condition
>> is met, set to -1 when another condition is met, left alone if neither
>> condition is met.
>
> This works on recent versions of numpy:
>
>>> x = asmatrix(zeros((2,2)))
>
>>> x
>
> matrix([[0, 0],
>       [0, 0]])
>
>>> y = asmatrix(rand(2,2))
>
>>> y
>
> matrix([[ 0.85219404,  0.48311427],
>       [ 0.41026966,  0.2184193 ]])
>
>>> x[y > 0.5] = 1
>
>>> x[y < 0.5] = -1
>
>>> x
>
> matrix([[ 1, -1],
>       [-1, -1]])
>

Re: [Numpy-discussion] simple question

[Bill B. <wb...@gm...>]

That's the kind of thing that should maybe be on the NumpyExample list:
http://www.scipy.org/Numpy_Example_List
But currently little syntactical tricks like that aren't listed there.   I'm
not sure how you'd list that even.

There is an example like that ( a[a<0.5]=0 ) on the Numpy for Matlab Users
page, though.   http://www.scipy.org/NumPy_for_Matlab_Users
It's supposed to be for matlab users, but if you just ignore the Matlab
column, the main table there is still a pretty good list of examples of
numpy syntax and basic functions.

--bb

On 7/6/06, Mathew Yeates <my...@jp...> wrote:
>
> ohhhh. I was looking at using "where"
>
>
>
> Keith Goodman wrote:
> > On 7/5/06, Mathew Yeates <my...@jp...> wrote:
> >> What is the typical way of doing the following
> >> starting with a 0 matrix, set all values to 1 when a certain condition
> >> is met, set to -1 when another condition is met, left alone if neither
> >> condition is met.
> >
> > This works on recent versions of numpy:
> >
> >>> x = asmatrix(zeros((2,2)))
> >
> >>> x
> >
> > matrix([[0, 0],
> >       [0, 0]])
> >
> >>> y = asmatrix(rand(2,2))
> >
> >>> y
> >
> > matrix([[ 0.85219404,  0.48311427],
> >       [ 0.41026966,  0.2184193 ]])
> >
> >>> x[y > 0.5] = 1
> >
> >>> x[y < 0.5] = -1
> >
> >>> x
> >
> > matrix([[ 1, -1],
> >       [-1, -1]])
> >
>
>
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job
> easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Numpy-discussion mailing list
> Num...@li...
> https://lists.sourceforge.net/lists/listinfo/numpy-discussion
>



-- 
William V. Baxter III
OLM Digital
Kono Dens Building Rm 302
1-8-8 Wakabayashi Setagaya-ku
Tokyo, Japan  154-0023
+81 (3) 3422-3380

Re: [Numpy-discussion] simple question

[Pau G. <pau...@gm...>]

On 7/6/06, Bill Baxter <wb...@gm...> wrote:
> That's the kind of thing that should maybe be on the NumpyExample list:
> http://www.scipy.org/Numpy_Example_List
> But currently little syntactical tricks like that aren't listed there.   I'm
> not sure how you'd list that even.
>
There is a (tentative) introduction to 'fancy' indexing in the
tentative numpy tutorial.
http://www.scipy.org/Tentative_NumPy_Tutorial#head-0dffc419afa7d77d51062d40d2d84143db8216c2

i'm going to add the '[ ]' operator to the Numpy example list.

pau

Re: [Numpy-discussion] simple question

[Christopher B. <Chr...@no...>]

Mathew Yeates wrote:
> ohhhh. I was looking at using "where"

There's nothing wrong with where:

 >>> y = N.asmatrix(N.rand(3,3))
 >>> y
matrix([[ 0.29741635,  0.78756994,  0.641378  ],
        [ 0.0198837 ,  0.71677631,  0.76068183],
        [ 0.84904382,  0.80169706,  0.23877389]])
 >>> x = N.asmatrix(N.where(y > 0.5, 1, 0))
 >>> x
matrix([[0, 1, 1],
        [0, 1, 1],
        [1, 1, 0]])

And did you really mean matrix? or would 2-d arrays be fine?

-CHB

-- 
Christopher Barker, Ph.D.
Oceanographer
                                     		
NOAA/OR&R/HAZMAT         (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chr...@no...