CODEWITHSUNDEEP
pythonnumpy
Peter
Peter

Reputation: 33

float values error after numpy vstack


I have a problem while using numpy vstack for float values with exponent e-5.
Both item_array and date_array are numpy.ndarray.
When I use data = np.vstack((date_array,item_array)), the element 3.04871703e-05 become to 3.0487170344. Other elements are good. Does anyone can help me to fix this issue? Thanks

In [2]: item_array
Out[2]:
array([ 0.00000000e+00, -1.81992510e-03, -9.44964473e-03,
-3.12464669e-03, -5.42864845e-03, -1.67769866e-03,
3.04871703e-05, -7.81295968e-03, -1.12972557e-02,
-1.69797339e-02, -1.22161657e-02, -1.93931514e-02,
-1.11389637e-02, -7.59505250e-03, 5.65141213e-03,
4.81559901e-03, -1.37724956e-02, -1.51201763e-02,
-2.55894748e-02, -2.48333169e-02, -2.56770574e-02,
-3.21192961e-02, -2.71028609e-02, -2.84357450e-02])

In [3]: date_array
Out[3]:
array(['"2016-05-03"', '"2016-05-04"', '"2016-05-05"', '"2016-05-06"',
'"2016-05-07"', '"2016-05-08"', '"2016-05-09"', '"2016-05-10"',
'"2016-05-11"', '"2016-05-12"', '"2016-05-13"', '"2016-05-14"',
'"2016-05-15"', '"2016-05-16"', '"2016-05-17"', '"2016-05-18"',
'"2016-05-19"', '"2016-05-20"', '"2016-05-21"', '"2016-05-22"',
'"2016-05-23"', '"2016-05-24"', '"2016-05-25"', '"2016-05-26"'],
dtype='|S12')

In [4]: data = np.vstack((date_arry,item_array))

In [5]: data
Out[5]:
array([['"2016-05-03"', '"2016-05-04"', '"2016-05-05"', '"2016-05-06"',
'"2016-05-07"', '"2016-05-08"', '"2016-05-09"', '"2016-05-10"',
'"2016-05-11"', '"2016-05-12"', '"2016-05-13"', '"2016-05-14"',
'"2016-05-15"', '"2016-05-16"', '"2016-05-17"', '"2016-05-18"',
'"2016-05-19"', '"2016-05-20"', '"2016-05-21"', '"2016-05-22"',
'"2016-05-23"', '"2016-05-24"', '"2016-05-25"', '"2016-05-26"'],
['0.0', '-0.001819925', '-0.009449644', '-0.003124646',
'-0.005428648', '-0.001677698', '3.0487170344', '-0.007812959',
'-0.011297255', '-0.016979733', '-0.012216165', '-0.019393151',
'-0.011138963', '-0.007595052', '0.0056514121', '0.0048155990',
'-0.013772495', '-0.015120176', '-0.025589474', '-0.024833316',
'-0.025677057', '-0.032119296', '-0.027102860', '-0.028435744']],
dtype='|S12')

Upvotes: 3

Views: 676

Answers (1)

Essex
Essex

Reputation: 6128

I wrote your script like that :

import numpy as np

item_array = np.array([0.00000000e+00, -1.81992510e-03, -9.44964473e-03,
                       -3.12464669e-03, -5.42864845e-03, -1.67769866e-03,
                       3.04871703e-05, -7.81295968e-03, -1.12972557e-02,
                       -1.69797339e-02, -1.22161657e-02, -1.93931514e-02,
                       -1.11389637e-02, -7.59505250e-03, 5.65141213e-03,
                       4.81559901e-03, -1.37724956e-02, -1.51201763e-02,
                       -2.55894748e-02, -2.48333169e-02, -2.56770574e-02,
                       -3.21192961e-02, -2.71028609e-02, -2.84357450e-02])

date_array = np.array(['"2016-05-03"', '"2016-05-04"', '"2016-05-05"', '"2016-05-06"',
                       '"2016-05-07"', '"2016-05-08"', '"2016-05-09"', '"2016-05-10"',
                       '"2016-05-11"', '"2016-05-12"', '"2016-05-13"', '"2016-05-14"',
                       '"2016-05-15"', '"2016-05-16"', '"2016-05-17"', '"2016-05-18"',
                       '"2016-05-19"', '"2016-05-20"', '"2016-05-21"', '"2016-05-22"',
                       '"2016-05-23"', '"2016-05-24"', '"2016-05-25"', '"2016-05-26"'])

data = np.vstack((date_array,item_array))

print data

And I get a good result : 

[['"2016-05-03"' '"2016-05-04"' '"2016-05-05"' '"2016-05-06"'
  '"2016-05-07"' '"2016-05-08"' '"2016-05-09"' '"2016-05-10"'
  '"2016-05-11"' '"2016-05-12"' '"2016-05-13"' '"2016-05-14"'
  '"2016-05-15"' '"2016-05-16"' '"2016-05-17"' '"2016-05-18"'
  '"2016-05-19"' '"2016-05-20"' '"2016-05-21"' '"2016-05-22"'
  '"2016-05-23"' '"2016-05-24"' '"2016-05-25"' '"2016-05-26"']
 ['0.0' '-0.0018199251' '-0.00944964473' '-0.00312464669' '-0.00542864845'
  '-0.00167769866' '3.04871703e-05' '-0.00781295968' '-0.0112972557'
  '-0.0169797339' '-0.0122161657' '-0.0193931514' '-0.0111389637'
  '-0.0075950525' '0.00565141213' '0.00481559901' '-0.0137724956'
  '-0.0151201763' '-0.0255894748' '-0.0248333169' '-0.0256770574'
  '-0.0321192961' '-0.0271028609' '-0.028435745']]

If you don't write dtype, it should work well ;)

Upvotes: 1

Related Questions