Numpy.savetxt ----- > How to save an array/matrix to a csv in python numpy?

Question

I have created a sort of matrix/data-table in python from multiple arrays containing various kinds of data, which looks like this in the variable view:

phones  starts  mids

sil 308000.0    308195.0

DH  308390.0    308410.0

AH0 308430.0    308445.0

B   308460.0    308525.0

These are three lists that I have combined using numpy.columnstack(). I am trying to output this table/matrix in the same format/structure but as a csv using numpy.savetxt. Is this possible given the different variable types in each row/column?

When I try to run :

np.savetxt('blue.txt', np.column_stack((phones, phone_starts, phone_mids)), ',', fmt="%s %f")

I get the error:

File "<__array_function__ internals>", line 4, in savetxt
TypeError: _savetxt_dispatcher() got multiple values for argument 'fmt'

I have also tried without the 'fmt' option and have gotten a similar error.

Any ideas?

Look at/show the result of the column_stack. What is its dtype and shape? My guess it is 2d with a string dtype. (with 3 columns). In which case, I'd try fmt='%s', savetxt writes a fmt % tuple(row) for each row of the array. — hpaulj
– hpaulj, Commented Mar 29, 2020 at 3:13
The input order for savetxt is filename, X, fmt, delimiter, .... Use delimiter=',', fmt='%s'. — hpaulj
– hpaulj, Commented Mar 29, 2020 at 3:18

mamdf · Accepted Answer · 2020-03-29 05:41:18Z

1

Well, do you want to save np arrays to csv with labels? that is the type of work for which pandas were created.

import numpy as np
import pandas as pd

data = {
"phone_starts": np.array([308000.0, 308390.0, 308430.0, 308460.0]),
"phone_mids": np.array([308195.0, 308410.0, 308445.0, 308525.0])
   }

df = pd.DataFrame(data, index=["sil", "DH", "AH0", "B"])
df.index.name = "Phone"
df.to_csv("blue.csv")

answered Mar 29, 2020 at 5:41

mamdf

111 silver badge1 bronze badge

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

3 Comments

Charlie P Over a year ago

Hello thank you very much for responding. I got the sense that pandas might be necessary/more appropriate. i do want to save an array with labels so I will try implementing the code you suggested.

Charlie P Over a year ago

Am I able to create the data frame from the lists stored in variables or do I have to type them manuall? Because the portion above is only just a small part of the whole data

mamdf Over a year ago

yes, for example, df = pd.DataFrame(list_arrays, columns=list_index, index=list_columns), if you create a list of np.arrays as data, I think DataFrame will infer it as rows, but it's possible to change that after. df = df.T

Collectives™ on Stack Overflow

Numpy.savetxt ----- > How to save an array/matrix to a csv in python numpy?

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related