Inserting data from dataframe into numpy array

Question

I am inserting data from a dataframe df with 55 rows into a numpy array matrix_of_coupons_and_facevalues with a shape of (55,60). I am doing this using the code below. However, I get the error IndexError: index 55 is out of bounds for axis 0 with size 55. months_to_maturity contains the numbers [6:6:330].

for (i,row) in df.iterrows():
    matrix_of_coupons_and_facevalues[i,0:(row.months_to_maturity/ 6)-1] = 1/2
    matrix_of_coupons_and_facevalues[i,(row.months_to_maturity/6)-1] = 3/2

Thank You

Can you verify that df does indeed have 55 rows? From the looks of the code, it seems like i is getting a value of 55, which could only occur if df had more than 55 rows. Perhaps you could try something like print(len(list(df.iterrows()))) to see if it is indeed 55. — brenns10
– brenns10, Commented Jul 14, 2015 at 18:30
can you please show how your dataframe looks like? Especially the index. — Anand S Kumar
– Anand S Kumar, Commented Jul 14, 2015 at 18:31
Interesting. It occurs to me that if you've deleted any rows from the dataframe, than the index won't go from 0 to 54 like you might expect, it'll have gaps. It would be helpful if you checked, as @AnandSKumar suggested, what the index contains. — brenns10
– brenns10, Commented Jul 14, 2015 at 18:36
I just realized that I need to reset the index for the dataframe. Its sorted now. Thanks. — user131983
– user131983, Commented Jul 14, 2015 at 18:37

brenns10 · Accepted Answer · 2015-07-14 18:57:20Z

For any future visitors, here's what happened:

A DataFrame's index serves to label each row uniquely, so when you delete a row, that index is removed and you have a "gap" in the index. This is very good when you have a meaningful index. But, when you just want the index to number your rows, it's not what you want. In this case, df contained 55 rows, but the index had holes, so the largest index was something larger than 55, causing an IndexError in the matrix. As an example:

In [1]: import pandas as pd

In [2]: df = pd.DataFrame([[1,2],[3,4],[5,6]], columns=['x','y'])

In [3]: df
Out[3]: 
   x  y
0  1  2
1  3  4
2  5  6

In [4]: df = df.drop(1)

In [5]: df
Out[5]: 
   x  y
0  1  2
2  5  6

In order to remedy this situation, you can simply reassign the index to be a list containing the correct range of numbers:

In [6]: df.index = list(range(len(df.index)))

In [7]: df
Out[7]: 
   x  y
0  1  2
1  5  6

heaven00 · Accepted Answer · 2015-07-15 10:25:06Z

0

or you could use pandas reset_index

In [18]: df.drop(1).reset_index()
Out[18]:
   index  x  y
0      0  1  2
1      2  5  6

answered Jul 15, 2015 at 10:25

heaven00

2003 silver badges9 bronze badges

Collectives™ on Stack Overflow

Inserting data from dataframe into numpy array

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related