import gensim
text = df.colomn.apply(gensim.utils.simple_preprocess)
model = gensim.models.Word2Vec(
window=10,
min_count=2,
workers=4,
)
model.build_vocab(text, progress_per=1000)
model.train(text, total_examples=model.corpus_count, epochs=model.epochs)
model.save("./word2vec2.model")
model.wv.most_similar("test")