CODEWITHSUNDEEP
pythonpandas
Luca91
Luca91

Reputation: 609

repeat random numbers each n rows

I have a large dataframe with 2 columns, like this:

                    dtm                SoC
      0     2018-03-01 00:00:00 +0000   39
      1     2018-03-01 00:00:01 +0000   39
      2     2018-03-01 00:00:02 +0000   39
      3     2018-03-01 00:00:03 +0000   39
     ...               ...              ...
2678393     2018-04-01 00:59:53 +0100   39
2678394     2018-04-01 00:59:54 +0100   39
2678395     2018-04-01 00:59:55 +0100   39
2678396     2018-04-01 00:59:56 +0100   39
2678397     2018-04-01 00:59:57 +0100   39
2678398     2018-04-01 00:59:58 +0100   39
2678399     2018-04-01 00:59:59 +0100   39

the column SoC is a random generated number between 0 and 40. I would like it to be a different random number repeated each 86400 rows (and not be the same for the entire dataframe).

To be more clear:

-rows 0-86399 1st random number

-rows 86400-172800 2nd random number

-etc

I was trying df['SoC']=np.repeat(random.randint(0,40),len(df)/86400) but there is an error "Length of values does not match length of index"

ideas? thank you in advance

Upvotes: 2

Views: 224

Answers (2)

jezrael
jezrael

Reputation: 862641

First create array with parameter size with floor division for integer and then repeat:

#possible duplicated random values
df['SoC'] = np.repeat(np.random.randint(0,40, size=len(df) // 86400), 86400)

#unique random numbers
df['SoC'] = np.repeat(np.random.choice(np.range(0, 40), 
                      size=len(df) // 86400, replace=False), 86400)

Upvotes: 2

sync11
sync11

Reputation: 1280

Another way:

arr = np.arange(40)
np.random.shuffle(arr)
arr

array([15, 30, 21,  3, 10, 19, 13, 31,  5, 32,  1, 39, 24,  6, 12,  7, 22,
       38, 27, 20, 25, 35, 14, 28, 33, 18, 29, 17, 37, 36, 34,  8,  2,  0,
        4, 11, 16, 23, 26,  9])

df['SoC'] = np.repeat(arr, 86400)

Upvotes: 0

Related Questions