"numpy.interp" ValueError: object too deep for desired array

Question

I have a data set that has one set of x values and many sets of y values that have the same set of x values. For instance,

X     Y2     Y3
0     2     1     
2     4     15
4     7     20
6     4     30
8     5     10
10     0.2     1 
12     0.7     2
14     1     1
16     1.2     10
18     2.4     11
20     2.5     5
22     3     6
24     1     7
26     8     7
28     9     5
30     1     1.2
32     1.2     1.5
34     1.5     2

I want to get Y2 and Y3 values for "new_x" by using interpolation. new_x=np.arange(0,34,0.424).

array([ 0.   ,  0.424,  0.848,  1.272,  1.696,  2.12 ,  2.544,  2.968,
        3.392,  3.816,  4.24 ,  4.664,  5.088,  5.512,  5.936,  6.36 ,
        6.784,  7.208,  7.632,  8.056,  8.48 ,  8.904,  9.328,  9.752,
       10.176, 10.6  , 11.024, 11.448, 11.872, 12.296, 12.72 , 13.144,
       13.568, 13.992, 14.416, 14.84 , 15.264, 15.688, 16.112, 16.536,
       16.96 , 17.384, 17.808, 18.232, 18.656, 19.08 , 19.504, 19.928,
       20.352, 20.776, 21.2  , 21.624, 22.048, 22.472, 22.896, 23.32 ,
       23.744, 24.168, 24.592, 25.016, 25.44 , 25.864, 26.288, 26.712,
       27.136, 27.56 , 27.984, 28.408, 28.832, 29.256, 29.68 , 30.104,
       30.528, 30.952, 31.376, 31.8  , 32.224, 32.648, 33.072, 33.496,
       33.92 ])

When I run intrp_CC=np.interp(new_x,old_x,current_Y), I get ValueError: object too deep for desired array.

Any idea as to why this is happening and how I can fix it?

hpaulj · Accepted Answer · 2018-12-04 21:56:11Z

2

Did you read the docs?

Returns the one-dimensional piecewise linear interpolant to a function with given discrete data points (xp, fp), evaluated at x.

fp : 1-D sequence of float or complex

With a 1d fp:

In [397]: np.interp(np.arange(10), np.arange(10), np.arange(20).reshape(2,10)[0,:])
Out[397]: array([0., 1., 2., 3., 4., 5., 6., 7., 8., 9.])

With a 2d, multirow, fp - your error:

In [398]: np.interp(np.arange(10), np.arange(10), np.arange(20).reshape(2,10))
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-398-e1383ba2f0f0> in <module>()
----> 1 np.interp(np.arange(10), np.arange(10), np.arange(20).reshape(2,10))

/usr/local/lib/python3.6/dist-packages/numpy/lib/function_base.py in interp(x, xp, fp, left, right, period)
   1306         fp = np.concatenate((fp[-1:], fp, fp[0:1]))
   1307 
-> 1308     return interp_func(x, xp, fp, left, right)
   1309 
   1310 

ValueError: object too deep for desired array

answered Dec 4, 2018 at 21:56

hpaulj

233k14 gold badges260 silver badges392 bronze badges

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

1 Comment

user7852656 Over a year ago

My bad. What I end up having to do is to create a for loop to do interpolate one y for one x for all sets of y values.

Johann cohen-tanugi · Accepted Answer · 2021-04-07 08:46:32Z

2

I think that it is still a fair question to ask if the explicit for loop is in any way avoidable..... The answer to that question is yes, provided one uses scipy interpolate.interp1d module instead of numpy.

answered Apr 7, 2021 at 8:46

Johann cohen-tanugi

1671 silver badge12 bronze badges

Collectives™ on Stack Overflow

"numpy.interp" ValueError: object too deep for desired array

2 Answers 2

1 Comment

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related