Mailman 3 Numpy newbie question: matrix creation - NumPy-Discussion - python.org

newer
Heads up re: new memory tracking...

Numpy newbie question: matrix creation

older
Removal of numarray and oldnumeric...

Edmondo Porcu

25 Sep 2013 25 Sep '13

9:06 a.m.

Dear all, I am a Newbie with Numpy and I would need some advice on how to create a matrix with certain characteristics : - Every entry should be minimum 0 maximum 1 with a step of 0.1 (legal values are 0,0.1,0.2,0.3 etc) - The number of columns of the matrix is a parameter of this matrix creation algorithm - Only the rows where the sum is 1 must be kept Would great appreciate your advice and suggestions Best Regards Edmondo

Attachments:

attachment.htm (text/html — 629 bytes)

Reply

Sign in to reply online Use email software

Show replies by date

Stéfan van der Walt

25 Sep 25 Sep

2:04 p.m.

On Wed, Sep 25, 2013 at 9:06 AM, Edmondo Porcu wrote:

I am a Newbie with Numpy and I would need some advice on how to create a matrix with certain characteristics :

- Every entry should be minimum 0 maximum 1 with a step of 0.1 (legal values are 0,0.1,0.2,0.3 etc)

You can generate random integers between 0 and 10, and divide them by 10: np.random.random_integers(0, 10, (5, 5)) / 10.

- Only the rows where the sum is 1 must be kept

I'm not sure I understand the exact structure you describe. Perhaps an example would help. Stéfan

Reply

Sign in to reply online Use email software

Robert Kern

2:10 p.m.

On Wed, Sep 25, 2013 at 8:06 AM, Edmondo Porcu wrote:

Dear all,

I am a Newbie with Numpy and I would need some advice on how to create a

matrix with certain characteristics :

- Every entry should be minimum 0 maximum 1 with a step of 0.1 (legal

values are 0,0.1,0.2,0.3 etc)

- The number of columns of the matrix is a parameter of this matrix

creation algorithm

- Only the rows where the sum is 1 must be kept

Would great appreciate your advice and suggestions

Use a multinomial distribution with n=10, then divide by 10.0. For example, with 5 columns and 20 samples: [~] |4> np.random.multinomial(10, np.ones(5)/5.0, size=20) / 10.0 array([[ 0.1, 0.1, 0.4, 0.1, 0.3], [ 0.2, 0.4, 0.1, 0.3, 0. ], [ 0.3, 0.2, 0. , 0.2, 0.3], [ 0. , 0.4, 0.4, 0. , 0.2], [ 0.2, 0.3, 0.1, 0.2, 0.2], [ 0.1, 0.5, 0.1, 0.1, 0.2], [ 0.4, 0. , 0.2, 0.3, 0.1], [ 0.3, 0.3, 0.1, 0. , 0.3], [ 0. , 0.2, 0.5, 0.2, 0.1], [ 0.5, 0.1, 0. , 0.2, 0.2], [ 0.1, 0.3, 0.2, 0.2, 0.2], [ 0.3, 0.1, 0.2, 0.2, 0.2], [ 0.2, 0.3, 0. , 0.3, 0.2], [ 0.2, 0.3, 0.2, 0.3, 0. ], [ 0.1, 0.3, 0.2, 0. , 0.4], [ 0.1, 0.1, 0.2, 0.5, 0.1], [ 0.1, 0.6, 0. , 0. , 0.3], [ 0.1, 0.2, 0.4, 0.1, 0.2], [ 0.3, 0.1, 0.2, 0.2, 0.2], [ 0.2, 0.1, 0.4, 0.2, 0.1]]) -- Robert Kern

Reply

Sign in to reply online Use email software

Edmondo Porcu

2:12 p.m.

That's what I was looking for, except that I want to be sure to generate all the possible combinations, and to have no repeated values. Thanks Edmondo 2013/9/25 Robert Kern

On Wed, Sep 25, 2013 at 8:06 AM, Edmondo Porcu wrote:

...
Dear all,

I am a Newbie with Numpy and I would need some advice on how to create a

matrix with certain characteristics :

...
- Every entry should be minimum 0 maximum 1 with a step of 0.1 (legal

values are 0,0.1,0.2,0.3 etc)

...
- The number of columns of the matrix is a parameter of this matrix

creation algorithm

...
- Only the rows where the sum is 1 must be kept

Would great appreciate your advice and suggestions

Use a multinomial distribution with n=10, then divide by 10.0.

For example, with 5 columns and 20 samples:

[~] |4> np.random.multinomial(10, np.ones(5)/5.0, size=20) / 10.0 array([[ 0.1, 0.1, 0.4, 0.1, 0.3], [ 0.2, 0.4, 0.1, 0.3, 0. ], [ 0.3, 0.2, 0. , 0.2, 0.3], [ 0. , 0.4, 0.4, 0. , 0.2], [ 0.2, 0.3, 0.1, 0.2, 0.2], [ 0.1, 0.5, 0.1, 0.1, 0.2], [ 0.4, 0. , 0.2, 0.3, 0.1], [ 0.3, 0.3, 0.1, 0. , 0.3], [ 0. , 0.2, 0.5, 0.2, 0.1], [ 0.5, 0.1, 0. , 0.2, 0.2], [ 0.1, 0.3, 0.2, 0.2, 0.2], [ 0.3, 0.1, 0.2, 0.2, 0.2], [ 0.2, 0.3, 0. , 0.3, 0.2], [ 0.2, 0.3, 0.2, 0.3, 0. ], [ 0.1, 0.3, 0.2, 0. , 0.4], [ 0.1, 0.1, 0.2, 0.5, 0.1], [ 0.1, 0.6, 0. , 0. , 0.3], [ 0.1, 0.2, 0.4, 0.1, 0.2], [ 0.3, 0.1, 0.2, 0.2, 0.2], [ 0.2, 0.1, 0.4, 0.2, 0.1]])

-- Robert Kern

_______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion

Reply

Sign in to reply online Use email software

Robert Kern

2:38 p.m.

On Wed, Sep 25, 2013 at 1:12 PM, Edmondo Porcu wrote:

That's what I was looking for, except that I want to be sure to generate

all the possible combinations, and to have no repeated values. Okay, then you need to find all of the integer partitions of 10 with `ncols` elements (padding with 0s for those partitions with fewer than `ncols` elements), finding the permutations of each padded partition, then eliminating duplicates. import itertools import numpy as np def gen_partitions(n): """ Generate integer partitions for `n`. http://homepages.ed.ac.uk/jkellehe/partitions.php """ a = [0 for i in range(n + 1)] k = 1 y = n - 1 while k != 0: x = a[k - 1] + 1 k -= 1 while 2*x <= y: a[k] = x y -= x k += 1 l = k + 1 while x <= y: a[k] = x a[l] = y yield a[:k + 2] x += 1 y -= 1 a[k] = x + y y = x + y - 1 yield a[:k + 1] def gen_permuted_partitions(ncols, n=10): for partition in gen_partitions(n): partition = list(partition) if len(partition) > ncols: continue partition.extend([0] * (ncols - len(partition))) for x in itertools.permutations(partition): yield x def compute_matrix(ncols): return np.array(sorted(set(gen_permuted_partitions(ncols)))) / 10.0 -- Robert Kern

Reply

Sign in to reply online Use email software

Alan G Isaac

3 p.m.

On 9/25/2013 3:06 AM, Edmondo Porcu wrote:

advice on how to create a matrix with certain characteristics : - Every entry should be minimum 0 maximum 1 with a step of 0.1 (legal values are 0,0.1,0.2,0.3 etc) - The number of columns of the matrix is a parameter of this matrix creation algorithm - Only the rows where the sum is 1 must be kept

This is not nearly enough information. But since it is all you offer, here is an answer for a square matrix with parameter n: np.identity(n) Alan Isaac

Reply

Sign in to reply online Use email software

3865

Age (days ago)

3865

Last active (days ago)

Download

5 comments

4 participants

tags

participants (4)

Alan G Isaac
Edmondo Porcu
Robert Kern
Stéfan van der Walt