Shell Programming and Scripting

I need to write a simple bash shell script that will recurse a directory and print out the top 2-3 directories with the most files in it.

I know how to use find & wc, but cannot seem to figure out how to store the count in each directory found, sort the count by the number and print out the top few directories that have the most files.

I would pass the script a single directory and it would make a list of all directories below it and count the number of files in each directory, store the count and then sort by descending order & print out the top 2-3 directories with the most files in it (excluding subdirectories in that directory - just want a count of files).

If anyone has an idea, that would be great,

Code:

find /path -type f |awk  '{$NF="";a[$0]++}END{for (i in a) print a[i],i }' FS=\/ OFS=\/ |sort -rn|head -3