randomly shuffle 3d numpy array on specific axis

Question

I have loaded 66 csv files, each containing a time series of 5000 time steps and three features with the following code:

rms = glob(sims_dir+"/rms*.csv")
df = [pd.read_csv(f).values for f in rms]
data = np.asarray(df)

data.shape
(66, 5004, 3)

Here my first axis of size 66 are my 66 unique time series. I would like to shuffle my array so that my first dimension (66) is of random order, but aren't exactly sure of the best way to do this... an alternative approach could be to load each csv randomly from its directory but I was wondering how this could be achieved in numpy.

Imanpal Singh · Accepted Answer · 2020-07-14 22:38:33Z

1

Use np.random.shuffle(data). This function is dedicated for this task. See np.random.shuffle

edited Jul 14, 2020 at 22:38

Imanpal Singh

1,2081 gold badge13 silver badges25 bronze badges

answered Jul 14, 2020 at 11:01

tstanisl

14.3k3 gold badges31 silver badges45 bronze badges

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

2 Comments

ojp Over a year ago

Thanks, but does this only suffle with respect to the first dimension? i want the order of the second and third to be maintained.

ojp Over a year ago

Ahh "Multi-dimensional arrays are only shuffled along the first axis" thank you!

Collectives™ on Stack Overflow

randomly shuffle 3d numpy array on specific axis

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related