I've got a column within a df labelled a. It contains a list of int values. I want to return if they are consecutive or not.
I can do it passing in a single list but I want to iterate over each row.
df = pd.DataFrame({'a': [[0,2], [9,11,12], [0,1,2], [10,11,13]]})
def cons(L):
return all(n-i == L[0] for i,n in enumerate(L))
print(cons(df['a'][0])) # works
df['cons'] = df['a'].apply(cons, axis=1) # error
intended:
a cons
0 [0, 2] False
1 [9, 11, 12] False
2 [0, 1, 2] True
3 [10, 11, 13] False
Use list comprehension with test if sorted values are same like generated list by minimal and maximal values of lists:
df['cons'] = [sorted(l) == list(range(min(l), max(l)+1)) for l in df['a']]
#alternative
df['cons'] = df['a'].apply(lambda l: sorted(l) == list(range(min(l), max(l)+1)))
Another idea is use np.diff for test if difference is 1 for all values:
df['cons'] = [np.all(np.diff(sorted(l)) == 1) for l in df['a']]
#alternative
df['cons'] = df['a'].apply(lambda l: np.all(np.diff(sorted(l)) == 1))
If want use your solution:
def cons(L):
return all(n-i==L[0] for i,n in enumerate(L))
df['cons'] = df['a'].apply(cons)
#alternative
df['cons'] = [all(n-i==L[0] for i,n in enumerate(L)) for L in df['a']]
print (df)
a cons
0 [0, 2] False
1 [9, 11, 12] False
2 [0, 1, 2] True
3 [10, 11, 13] False