Exporting numpy array + String to CSV

Question

I have two numpy array, each shaped (x,1). I wanted to export the data to CSV file so it looks something like this

Channel- Freq(np Array 1) - FFT(np Array 2) - Class (a string label)

Channel and class column will have 1 fixed value, freq and FFT comes from the array.

I have tried using np.column_stack but that only allows combining freq and fft, nothing else. I have tried looking at pandas DataFrame (which the original data comes from after importing from CSV) I can't find anything useful at the moment.

def combine_data(freq, matrix):
    combine_dt = np.column_stack((freq, matrix))
    return combine_dt

def export_data_csv(freq, matrix):
    combined_data = np.column_stack((freq, matrix))
    np.savetxt("freq_amp.csv", combined_data, delimiter=",")

This only works to make a CSV with only two columns

Have you tried numpy.hstack? This allows you to stack two columns together horizontally. You then then write to a csv normally. — Jkind9
– Jkind9, Commented Jun 19, 2019 at 14:33
Given small examples of your arrays, and the desired output. How picky are you about the format and spacing of the numeric columns? column_stack can join 3 arrays. — hpaulj
– hpaulj, Commented Jun 19, 2019 at 15:08
so the data I need to export will have more columns, It will have e.g. Channel - Frequency - FFT.real() - FFT.imag() - Class, a channel is just 1-4 (one of the channel numbers), class is just what the input signal was to generate the original data, e.g. a frequency of 50Hz. — Abdullah
– Abdullah, Commented Jun 19, 2019 at 16:30

Francesco Zambolin · Accepted Answer · 2019-06-19 18:45:06Z

1

I think it's easier using pandas.

Let's say:

array1=[0,1,2,3,4]
array2=[10,11,12,13,14]

First create a dataframe with the 2 arrays and then set the constant values:

my_df = pd.DataFrame(data={"freq":array1,"FFT":array2})
my_df["channel"]="channel_name"
my_df["class"]="class_name"

#Output
    freq    FFT     channel     class
0     0     10   channel_name   class_name
1     1     11   channel_name   class_name
2     2     12   channel_name   class_name
3     3     13   channel_name   class_name
4     4     14   channel_name   class_name

then you can easily export it to csv:

my_df.to_csv("filename.csv")

if you don't want/need the index just set:

my_df.to_csv("filename.csv",index=False)

For other options just look at the documentation pandas.to_csv()

answered Jun 19, 2019 at 18:45

Francesco Zambolin

6016 silver badges9 bronze badges

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

Collectives™ on Stack Overflow

Exporting numpy array + String to CSV

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related