python - Sum based on a criterion - Stack Overflow

admin管理员组

文章数量:1321074

I have a data sheet where I want to do a calculation until I find a zero in one specific column. Then I want to sum all the results of this calculation up to that zero and save the result in an array. I tried with an np array and a list but it does not work:

tst = []
x = data[1:len(data),0]
y = data[1:len(data),1]  
intt = data[1:len(data),2]
for i in range(0,len(data)):
   if intt[i]!=0:         
      tst.append((x[i]**2.0+ y[i]-y[i-1])**2.0)

I want the tst to contain in each position, the sum of the expression in the append().

Thanks!

I have a data sheet where I want to do a calculation until I find a zero in one specific column. Then I want to sum all the results of this calculation up to that zero and save the result in an array. I tried with an np array and a list but it does not work:

tst = []
x = data[1:len(data),0]
y = data[1:len(data),1]  
intt = data[1:len(data),2]
for i in range(0,len(data)):
   if intt[i]!=0:         
      tst.append((x[i]**2.0+ y[i]-y[i-1])**2.0)

I want the tst to contain in each position, the sum of the expression in the append().

Thanks!

• python
• list
• sum

Share Improve this question Follow edited Jan 17 at 21:22 Riri asked Jan 17 at 21:22 RiriRiri 1577 bronze badges 1

• 2 Just add else: break. But also think about what y[i-1] is when i is 0. – trincot Commented Jan 17 at 21:26

Add a comment  | 

1 Answer 1

Sorted by: Reset to default 1

I looks like you skip the first datapoint since you write x = data[1:len(data),0] and so forth. As @trincot mentioned, you also have to care about the y[i-1] case for i=0. Maybe the following will help you:

tst = []
x = data[:,0]
y = data[:,1]  
intt = data[:,2]
total = []
for i in range(1,len(data)):
   if intt[i]!=0:         
      tst.append((x[i]**2.0+ y[i]-y[i-1])**2.0)
   else:
      total.append(sum(tst))
      break
total = np.array(total)

This includes all data points in x, y, and intt, but the first data point will still be skipped since the loop starts with i=1.

Edit: A cleaner solution, based on comment from @Barmar.

tst = []
x = data[:,0]
y = data[:,1]  
intt = data[:,2]
for i in range(1,len(data)):
   if intt[i]!=0:         
      tst.append((x[i]**2.0+ y[i]-y[i-1])**2.0)
   else:
      total = np.array(sum(tst))
      break

Edit: Another solution that goes through all of intt:

tst = []
x = data[:,0]
y = data[:,1]  
intt = data[:,2]
total = []
for i in range(1,len(data)):
   if intt[i]!=0:         
      tst.append((x[i]**2.0+ y[i]-y[i-1])**2.0)
   elif i == len(data) - 1:
      total.append(sum(tst))
   else:
      total.append(sum(tst))
      tst = []
      continue
total = np.array(total)

本文标签： pythonSum based on a criterionStack Overflow