how slice by hybrid stile-CodePudding

having a random df

df = pd.DataFrame([[1,2,3,4],[4,5,6,7],[7,8,9,10],[10,11,12,13],[14,15,16,17]], columns=['A', 'B','C','D'])
cols_in = list(df)[0:2] list(df)[4:]

now:

x = []
for i in range(df.shape[0]):
   x.append(df.iloc[i,cols_in])

obviously in the cycle, x return an error due to col_in assignment in iloc.

How could be possible apply a mixed style slicing of df like in append function ?

CodePudding user response：

It seems like you want to exclude one column? There is no column 4, so depending on which columns you are after, something like this might be what you are after:

df = pd.DataFrame([[1,2,3,4],[4,5,6,7],[7,8,9,10],[10,11,12,13],[14,15,16,17]], columns=['A', 'B','C','D'])

If you want to get the column indeces from column names you can do:

cols = ['A', 'B', 'D']
cols_in = np.nonzero(df.columns.isin(cols))[0]


x = []
for i in range(df.shape[0]):
   x.append(df.iloc[i, cols_in].to_list()) 
x

Output:

[[1, 2, 4], [4, 5, 7], [7, 8, 10], [10, 11, 13], [14, 15, 17]]