error for pandas select column

Question

I'm a new user to pandas. I run the following code with my titanic csv file (which I got from a tutorial) and the command titanic.sex returns the sex row as expected.

 titanic = pd.read_csv('/Users/eflannery/Dropbox/titanic.csv')
    titanic.Sex
    0       male
    1     female
    2     female
    3     female
    4       male
    5       male
    6       male
    7       male
    8     female
    9     female
    10    female
    11    female
    12      male
    13      male
    14    female
    ...
    876      male
    877      male
    878      male
    879    female
    880    female
    881      male
    882    female
    883      male
    884      male
    885    female
    886      male
    887    female
    888    female
    889      male
    890      male
    Name: Sex, Length: 891, dtype: object

If I run the same code with one of my csv files, it doesn't pull out the cov column as expected. I checked that my file is a csv. Any ideas?

sam = pd.read_csv('/Users/eflannery/Dropbox/Cartika/CNV_data/samFlags_NK65EvolB20_baseCounts2.csv')
sam.cov
<bound method DataFrame.cov of                chrom    pos  cov
0             berg02      1    0
1             berg02      2    0
2             berg02      3    0
3             berg02      4    0
4             berg02      5    2
5             berg02      6    3
6             berg02      7    3
7             berg02      8    3
8             berg02      9    4
9             berg02     10    4
10            berg02     11    4
11            berg02     12    4
12            berg02     13    4
13            berg02     14    4
14            berg02     15    4
15            berg02     16    4
16            berg02     17    5
17            berg02     18    5
18            berg02     19    5
19            berg02     20    5
20            berg02     21    5
21            berg02     22    5
22            berg02     23    5
23            berg02     24    6
24            berg02     25    6
25            berg02     26    6
26            berg02     27    6
27            berg02     28    6
28            berg02     29    6
29            berg02     30    6
...              ...    ...  ...
18433379  PBANKA_API  30273   25
18433380  PBANKA_API  30274   25
18433381  PBANKA_API  30275   25
18433382  PBANKA_API  30276   25
18433383  PBANKA_API  30277   25
18433384  PBANKA_API  30278   24
18433385  PBANKA_API  30279   24
18433386  PBANKA_API  30280   24
18433387  PBANKA_API  30281   24
18433388  PBANKA_API  30282   24
18433389  PBANKA_API  30283   24
18433390  PBANKA_API  30284   24
18433391  PBANKA_API  30285   18
18433392  PBANKA_API  30286   16
18433393  PBANKA_API  30287   16
18433394  PBANKA_API  30288   16
18433395  PBANKA_API  30289   16
18433396  PBANKA_API  30290   13
18433397  PBANKA_API  30291   13
18433398  PBANKA_API  30292   13
18433399  PBANKA_API  30293   13
18433400  PBANKA_API  30294   10
18433401  PBANKA_API  30295    8
18433402  PBANKA_API  30296    5
18433403  PBANKA_API  30297    5
18433404  PBANKA_API  30298    5
18433405  PBANKA_API  30299    5
18433406  PBANKA_API  30300    5
18433407  PBANKA_API  30301    5
18433408  PBANKA_API  30302    2

[18433409 rows x 3 columns]>

Answer 1

Using the . access syntax can be handy for convenience, but you've tripped across a problem with it: when there's a method of the same name, you get the method, not the column. Use dictionary-style access instead:

>>> df
    chrom  pos  cov
0  berg02    1    0
1  berg02    2    0
2  berg02    3    1
>>> df.cov
<bound method DataFrame.cov of     chrom  pos  cov
0  berg02    1    0
1  berg02    2    0
2  berg02    3    1>
>>> df["cov"]
0    0
1    0
2    1
Name: cov, dtype: int64

Or I guess you could actually compute the covariance :-)

>>> df.cov()
     pos       cov
pos  1.0  0.500000
cov  0.5  0.333333