Gensim: topic modelling for humans

✔ Train large-scale semantic NLP models ✔ Represent text as semantic vectors ✔ Find semantically related documents from gensim import corpora, models, similarities, downloader # Stream a training corpus directly from S3. corpus = corpora.MmCorpus("s3://path/to/corpus") # Train Latent Semantic Indexing with 200D vectors. lsi = models.LsiModel(corpus, num_topics=200) # Convert another corpus t