o1 = "laks laks skj?re deilig".split()
b1 = "laks lus celle alge alge".split()
o2 = "alge nori laks ris eddik".split()
b2 = "celle laks b?rekraftighet celle".split()
o3 = "laks laks laks laks laks laks laks laks laks laks".split()

d1 = "b?rekraftighet deilig laks ris".split()
d2 = "lus lus celle eddik".split()


alle = [o1,o2,b1,b2,o3]

bio = [b1,b2]
opp = [o1,o2,o3]

classes = {"oppskrift":opp,"biologi":bio}

V = []
for document in alle:
    for w in document:
        V.append(w)
#print(V)
V = set(V)
#print(V)


def predict(sentence):
    results = {}
    for name, cl in classes.items():
        p_prior = len(cl)/len(alle)
        c_vocab = [w for doc in cl for w in doc]
        p_sent = 1
        for w in sentence:
            if w in V:
                w_in_c = c_vocab.count(w) + 1
                tot_w_in_c = len(c_vocab) + len(V)
                p_sent *= w_in_c/tot_w_in_c
        p = p_prior * p_sent
        results[name] = p
    for r in results:
        print(r,results[r])
                 

predict(d1)
predict(d2)