docs: Fix sign in pandas example of column assignment

[jennynz]

In the docs on coming from pandas, under column assignment, the example says:

When the value in column c is equal to 2 then we replace the value in a with the value in b.

But the pandas example shown is:

df.assign(a=lambda df_: df_.a.where(df_.c != 2, df_.b))

This just changes the sign so it's actually when c is equal to 2, rather than not equal.

[jennynz]

Broken link is an existing one it seems:

ERROR: 1 dead links found!
[✖] https://matplotlib.org/ → Status: 403

[rodrigogiraoserrao]

This looks correct to me 👇 the first five rows of df have the value 2 in column c and in the result those five rows had the values of a replaced by the values of b.
Isn't this what's described?

What may be causing the confusion is where replaces values where the condition is False (c.f. the pandas docs on Series.where.

>>> df = pd.DataFrame({
...     "a": list(range(10)),
...     "b": [99] * 10,
...     "c": [2] * 5 + [0] * 5,
... })
>>>
>>> df
   a   b  c
0  0  99  2
1  1  99  2
2  2  99  2
3  3  99  2
4  4  99  2
5  5  99  0
6  6  99  0
7  7  99  0
8  8  99  0
9  9  99  0
>>> df.assign(a=lambda df_: df_.a.where(df_.c != 2, df_.b))
    a   b  c
0  99  99  2
1  99  99  2
2  99  99  2
3  99  99  2
4  99  99  2
5   5  99  0
6   6  99  0
7   7  99  0
8   8  99  0
9   9  99  0

[jennynz]

Ah apologies, you're right - I did get confused about how where works!

[rodrigogiraoserrao]

No worries. I triple checked everything before closing your PR because I was so confused as well. Double negations get me every time 😆