Fastest way to remove N random objects

My question is as follows, I am currently working with a generated list of length m. However the list is supposed to be the result of an algorithm taking n as an argument for the final length. m is always much large than n. Currently I am running a while loop where m is the result of len(list). ie:

from numpy import random as rnd
m = 400000
n = 3000
list = range(0,m)
while len(list) > n:
    rmi = rnd.randint(0, len(list))
    del list[rmi]
    print('%s/%s' %(len(list), n))

This approach certainly works but takes an incredibly long time to run. Is there a more efficient and less time consuming way of removing m-n random entries from my list? The entries removed must be random or the resulting list will no longer represent what it should be.