I have this dataframe:
I wish to have extract all the data for value column in form of array, which the output should be: [nan, 47, 47.9, 48.6]
Anyone can share me the ideas?
-
Possible duplicate of get list from pandas dataframe columnMcGrady– McGrady2017-05-09 05:11:21 +00:00Commented May 9, 2017 at 5:11
Add a comment
|
1 Answer
Use values for numpy array + numpy.ndarray.tolist for list:
L = df['value'].values.tolist()
#alternative
#L = df['value'].tolist()
Or convert to list:
L = list(df['value'])
EDIT:
It seems you need convert to float first by astype:
df = pd.DataFrame({'value':['nan','47','47.9','48.6']})
print (df)
value
0 nan
1 47
2 47.9
3 48.6
L1 = df['value'].values.tolist()
print (L1)
['nan', '47', '47.9', '48.6']
L2 = df['value'].astype(float).values.tolist()
print (L2)
[nan, 47.0, 47.9, 48.6]
because if float all values get different output - 47.0 not 47:
df = pd.DataFrame({'value':[np.nan,47,47.9,48.6]})
value
0 NaN
1 47.0
2 47.9
3 48.6
L = df['value'].values.tolist()
print (L)
[nan, 47.0, 47.9, 48.6]
EDIT1:
If astype return error e.g:
ValueError: could not convert string to float: 'a'
then need to_numeric with parameter errors='coerce' for convert non numeric to NaN
df = pd.DataFrame({'value':[np.nan,47,47.9,48.6, 'a']})
print (df)
value
0 NaN
1 47
2 47.9
3 48.6
4 a
print (pd.to_numeric(df['value'], errors='coerce'))
0 NaN
1 47.0
2 47.9
3 48.6
4 NaN
Name: value, dtype: float64
L = pd.to_numeric(df['value'], errors='coerce').values.tolist()
print (L)
[nan, 47.0, 47.9, 48.6, nan]
