Concatenate many arrays in python

Question

I have this code for concatenate two arrays.

import numpy as np
from hmmlearn import hmm
model = hmm.MultinomialHMM(n_components=3, n_iter=10,algorithm='map',tol=0.00001)
sequence3 = np.array([[2, 1, 0, 1]]).T
sequence4 = np.array([[2, 1, 0, 1, 1]]).T
sample = np.concatenate([sequence3, sequence4])
lengths = [len(sequence3), len(sequence4)]
model.fit(sample,lengths)

and it is working correctly. but now if I have more than two array. let us to say I have 10 arrays. how I can make the same process?

import numpy as np
from hmmlearn import hmm
model = hmm.MultinomialHMM(n_components=3, n_iter=10,algorithm='map',tol=0.00001)
sample = np.array([])
lengths = []
for i in range(1:10)
    ?????????????
model.fit(sample,lengths)

So, you'd probably need the new array sample and the lengths of them in list lengths? — kmario23
– kmario23, Commented Mar 26, 2017 at 8:24
You can concatenate many arrays at once. Just put them all in the argument list. But the dimensions have to match. — hpaulj
– hpaulj, Commented Mar 26, 2017 at 8:59

Aaron Brock · Accepted Answer · 2017-03-26 08:25:33Z

2

In order to concatenate more than one array, you simply concatenate the array with the concatenation of all the previous arrays.

# Create arrays
arrays=[
    np.array([1,2,3]),
    np.array([4,5,6]),
    np.array([7,8,9])
]

# Create an array to return to
sample = np.array([])

for array in arrays:
    sample = np.concatenate([sample, array])

# Print results
print('sample', sample)
print('length', len(sample))

answered Mar 26, 2017 at 8:25

Aaron Brock

4,5762 gold badges28 silver badges46 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

user1927468 Over a year ago

it still give me this error "ValueError: expected a sample from a Multinomial distribution." at this line model.fit(sample,lengths)

hpaulj Over a year ago

Just use concatenate(arrays)

user1927468 Over a year ago

now I have this error "ValueError: more than 1 samples in lengths array [3, 3, 3]" in this line model.fit(np.concatenate(arrays),lengths)

user1927468 Over a year ago

it is working with me now. The problem was arrays should be. arrays=[ np.array([[1,2,3]]), np.array([[4,5,6]]), np.array([[7,8,9]]) ]

user1927468 Over a year ago

and I used np.concatenate(arrays) direct without for

Keerthana Prabhakaran · Accepted Answer · 2017-03-26 08:24:26Z

1

You can use vstack

That is,

Equivalent to np.concatenate(tup, axis=0) if tup contains arrays that are at least 2-dimensional.

store your arrays as a list,say array_list

print np.vstack(array_list)

Sample:

import numpy as np
sequence3 = np.array([[2, 1]]).T
sequence4 = np.array([[2, 5]]).T
sequence5 = np.array([[4, 5]]).T
sequence6 = np.array([[6, 7]]).T
array_list=[sequence3,sequence4,sequence5,sequence6]
sample = np.concatenate([sequence3, sequence4])
lengths = [len(sequence3), len(sequence4)]
print np.vstack(array_list)

[[2]
 [1]
 [2]
 [5]
 [4]
 [5]
 [6]
 [7]]

Hope it helps!

answered Mar 26, 2017 at 8:24

Keerthana Prabhakaran

3,8071 gold badge16 silver badges25 bronze badges

1 Comment

hpaulj Over a year ago

Since the arrays are already 2d, I'd suggest concatenate with axis=0.

Collectives™ on Stack Overflow

Concatenate many arrays in python

2 Answers 2

5 Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

5 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related