import nltk
words = nltk.corpus.gutenberg.words('austen-persuasion')

#one way to do it:
print "Tokens: %s\tTypes: %s" % (len(words),len(set(words)))

#another way:
tokens=0;
types={}
for word in words:
    tokens+=1
    if word not in types:
        types[word]=1
print "Tokens: %s\tTypes: %s" % (tokens,len(types.keys()))