CODEWITHSUNDEEP
python
BAKKAB
BAKKAB

Reputation: 61

What is a more efficient way of calculating means with in an array?

I have some data that is stored as daily means over 20 years, and I would like to create a new array that tells me the monthly means over the same period.

I am not very experienced with python, so best I could figure was something like the following:

 dis2 = "array of daily means"
 a2 = np.sum(dis2[:365])
 b2 = np.sum(dis2[365:731])
 c2 = np.sum(dis2[731:1096])
 d2 = np.sum(dis2[1096:1461])
 e2 = np.sum(dis2[1461:1826])
 f2 = np.sum(dis2[1826:2191])
 g2 = np.sum(dis2[2191:2556])
 h2 = np.sum(dis2[2556: 2921])
 i2 = np.sum(dis2[2921:3286])
 j2 = np.sum(dis2[3286:3651])
 k2 = np.sum(dis2[3651:4016])
 l2 = np.sum(dis2[4016:4381])
 m2 = np.sum(dis2[4381:4746])
 n2 = np.sum(dis2[4746:5111])
 o2 = np.sum(dis2[5111:5476])
 p2 = np.sum(dis2[5476:5841])
 q2 = np.sum(dis2[5841:6206])
 r2 = np.sum(dis2[6206:6571])
 s2 = np.sum(dis2[6571:6936])
 t2 = np.sum(dis2[6936:7301])

 z2 = [a2,b2,c2,d2,e2,f2,g2,h2,i2,j2, k2, l2, m2, n2, o2, p2, q2, r2, s2, t2]
 z2 = [i/365 for i in z2]

The above method over course only gives me the yearly means, and to get the monthly means using this methods I'd need well over a hundred variables and I am certain there must be a simpler more efficient way of doing so, but I have not the experience to determine what it is.

In case it is at all relevant, here is how I loaded my data:

 filename = 'LOWL.txt'
 f2 = open(filename, 'r')


 date2 = []
 discharge2 = []


 lines = f2.readlines()


 import pandas as pd

 data2 = pd.read_csv('LOWL.txt',sep='\t',header=None,usecols=[2,3])
 date2 = data2[2].values
 discharge2 = data2[3].values


 date2 = np.array(date2, dtype = "datetime64")
 dis2 = [float(i) for i in discharge2]

Upvotes: 3

Views: 42

Answers (1)

Lukas Thaler
Lukas Thaler

Reputation: 2720

Combining np.mean(), slicing and a list comprehension makes for a more efficient way of doing your calculation:

z2 = [np.mean(dis2[i:(i+365)]) for i in range(0, len(dis2), 365)]

Upvotes: 1

Related Questions