transform Count consecutive integers

Create Series for consecutive integers by Series.diff, compare for not equal 1 by Series.ne and add cumulative sum by Series.cumsum:

s = df['id'].diff().ne(1).cumsum()

Then use Series.map with Series.value_counts:

df['count'] = s.map(s.value_counts())

Or GroupBy.transform with GroupBy.size:

df['count'] = s.groupby(s).transform('size') 

print (df)
   id  count
0   1      4
1   2      4
2   3      4
3   4      4
4   7      2
5   8      2