CODEWITHSUNDEEP
pythonlistpython-2.7numpysequence
Cairo Lievore
Cairo Lievore

Reputation: 53

amend code to return True or False for numbers in sequence

    import numpy as np                                          
    import re                                                   

    def validate(seq):                                          
        stl = "".join(np.diff(seq).astype(str))                 
        for x in re.findall("[1]+",stl):                        
            if len(x)>3:                                        
                return False                                    
        return True                                             

    print validate([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 16, 20])
    print validate([1, 2, 3, 6, 7, 8, 9, 11, 12, 16, 17, 18, 19, 22, 23])
    print validate([2, 3, 5, 7, 9, 10, 11, 12, 14, 15, 17, 20, 22, 23, 24])

output

    False
    True
    True

This code check the list and check how many numbers are chained and return False if there are more than 4 together like the first print (1, 2, 3, 4, 5, 6, 7, ...), returning True if there are only 4 in sequence like the second print (6, 7, 8, 9) (16, 17, 18, 19).

So how can I amend the code to return True when the list have only one group with 4 numbers in sequence? Returning False for lists with more than for numbers in sequence?

print validate([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 16, 20])
print validate([1, 2, 3, 6, 7, 8, 9, 11, 12, 16, 17, 18, 19, 22, 23])
print validate([2, 3, 5, 7, 9, 10, 11, 12, 14, 15, 17, 20, 22, 23, 24])

output

False
False
True

Upvotes: 0

Views: 133

Answers (1)

wasserfeder
wasserfeder

Reputation: 486

The way you defined validate function is somewhat hard to follow. I rewrote it in the following way

def validate(seq, counts=1, N=4):
    if len(seq) < N:
        return False

    diff_seq = np.diff(seq)
    sum_seq = np.array([(np.sum(diff_seq[i:i+N-1])==N-1) for i in range(len(diff_seq)-N+2)])
    return np.count_nonzero(sum_seq) == counts

where N is the length of consecutive numbers in a group and counts is the number of such groups that you want to have in seq.

EDIT You can use convolve to compute sum_seq in function validate as follows

    sum_seq = np.convolve(diff_seq, np.ones((N-1,)), mode='valid') == N-1

This should be much faster and it looks more pythonic.

Upvotes: 0

Related Questions