Split: Convert NumPy format into pandas

Asked Oct 15 '19 at 01:27

Active Oct 15 '19 at 01:27

Viewed 33 times

I have NumPy format of split function, what is want to do is convert this format into pandas format. the idea is to split each object separately

input_num = 10
train = np.array([np.float32(x) for x in line.split(",")[0:input_num]])
test  = np.array([np.float32(x) for x in line.split(",")[input_num:]])

edited Oct 15 '19 at 01:27

Dani Mesejo

61,499
6
49
76

asked Oct 15 '19 at 01:27

AbuOmair

You can do `pd.Series(test)` ? – Dani Mesejo Oct 15 '19 at 01:28
Possible duplicate of https://stackoverflow.com/questions/20763012/creating-a-pandas-dataframe-from-a-numpy-array-how-do-i-specify-the-index-colum – Sanil Khurana Oct 15 '19 at 01:30
What do `train` and `test` look like? `shape`? `dtype`? – hpaulj Oct 15 '19 at 01:30
@hpaulj `dtype` is float `train` 80% and `test` 20% – AbuOmair Oct 15 '19 at 01:41
@DanielMesejo How can use `pd.Series(test)` ? where to add it in code ? – AbuOmair Oct 15 '19 at 01:43
`pd_test = pd.Series(test)` after your second line – Dani Mesejo Oct 15 '19 at 01:45
@DanielMesejo I am facing this error ..... File "", line 2 train = np.array([np.float32(x) for x in line.split(",")[0:input_num]]) ^ IndentationError: unexpected indent – AbuOmair Oct 15 '19 at 01:49
`80%` is not a `shape`. – hpaulj Oct 15 '19 at 01:55
@hpaulj the `shape ` is 40300 – AbuOmair Oct 15 '19 at 03:00

Split: Convert NumPy format into pandas

0 Answers0