Fixed float NaN error

[tomszekeres]

Fixed "cannot convert float NaN to integer" error for 'week' and 'rank' in billboard.csv

[sprestridge]

#1
I noticed that error as well but this fix did not work for me. If I read the fix correctly it was the addition of two lines of code to the #formatting section of the Billboard 100 example. The two lines were:

df = df[~df['week'].isnull()]

df = df[~df['rank'].isnull()]

Still getting the NaN error:

ValueError: cannot convert float NaN to integer

I believe the a correct fix is the following:

df['week'] = df['week'].str.extract('(\d+)', expand=False).astype(np.float)

verse the line currently ending as astype(int)

[RoboKiteCat]

@sprestridge thanks that helped me. I also noted that I needed to use the same approach to the next line:

df["rank"] = df["rank"].astype(np.float)

were it was previously df["rank"] = df["rank"].astype(int)