在合并了几个data.tables之后，我得到了一个看起来有点像这样的表：

dt<- data.table(age = rep(c("5-16", "17-39"), each = 3), dur = rep(c("short","medium","long"), 2), short.count = rep(c(23,45),each = 3), medium.count = rep(c(12,34), each = 3), long.count = rep(c(3,5), each = 3))

> dt

     age    dur short.count medium.count long.count
1:  5-16  short          23           12          3
2:  5-16 medium          23           12          3
3:  5-16   long          23           12          3
4: 17-39  short          45           34          5
5: 17-39 medium          45           34          5
6: 17-39   long          45           34          5

我想把它融化掉，这样每行就有一个（合适的）计数值。

age    dur       count 
1:  5-16  short          23           
2:  5-16 medium          12   
3:  5-16   long          3    
4: 17-39  short          45          
5: 17-39 medium          34           
6: 17-39   long          5

有没有一个优雅的方式来做到这一点？