:star2: New Features
* Enable online training of Poincare models ([koiizukag](https://github.com/koiizukag), [#2505](https://github.com/RaRe-Technologies/gensim/pull/2505))
* Make BM25 more scalable by adding support for generator inputs ([saraswatmks](https://github.com/saraswatmks), [#2479](https://github.com/RaRe-Technologies/gensim/pull/2479))
* Allow the Gensim dataset / pre-trained model downloader `gensim.downloader` to run offline, by introducing a local file cache ([mpenkov](https://github.com/mpenkov), [#2545](https://github.com/RaRe-Technologies/gensim/pull/2545))
* Make the `gensim.downloader` target directory configurable ([mpenkov](https://github.com/mpenkov), [#2456](https://github.com/RaRe-Technologies/gensim/pull/2456))
* Add `nmslib` indexer ([masa3141](https://github.com/masa3141), [#2417](https://github.com/RaRe-Technologies/gensim/pull/2417))
:red_circle: Bug fixes
* Fix `smart_open` deprecation warning globally ([itayB](https://github.com/itayB), [#2530](https://github.com/RaRe-Technologies/gensim/pull/2530))
* Fix AppVeyor issues with Windows and Py2 ([mpenkov](https://github.com/mpenkov), [#2546](https://github.com/RaRe-Technologies/gensim/pull/2546))
* Fix `topn=0` versus `topn=None` bug in `most_similar`, accept `topn` of any integer type ([Witiko](https://github.com/Witiko), [#2497](https://github.com/RaRe-Technologies/gensim/pull/2497))
* Fix Python version check ([charsyam](https://github.com/charsyam), [#2547](https://github.com/RaRe-Technologies/gensim/pull/2547))
* Fix typo in FastText documentation ([Guitaricet](https://github.com/Guitaricet), [#2518](https://github.com/RaRe-Technologies/gensim/pull/2518))
* Fix "Market Matrix" to "Matrix Market" typo. ([Shooter23](https://github.com/Shooter23), [#2513](https://github.com/RaRe-Technologies/gensim/pull/2513))
* Fix auto-generated hyperlinks in `CHANGELOG.md` ([mpenkov](https://github.com/mpenkov), [#2482](https://github.com/RaRe-Technologies/gensim/pull/2482))
:books: Tutorial and doc improvements
* Generate documentation for the `gensim.similarities.termsim` module ([Witiko](https://github.com/Witiko), [#2485](https://github.com/RaRe-Technologies/gensim/pull/2485))
* Simplify the `Support` section in README ([piskvorky](https://github.com/piskvorky), [#2542](https://github.com/RaRe-Technologies/gensim/pull/2542))
:+1: Improvements
* Pin sklearn version for Py2, because sklearn dropped py2 support ([mpenkov](https://github.com/mpenkov), [#2510](https://github.com/RaRe-Technologies/gensim/pull/2510))
:warning: Deprecations (will be removed in the next major release)
* Remove
- `gensim.models.FastText.load_fasttext_format`: use load_facebook_vectors to load embeddings only (faster, less CPU/memory usage, does not support training continuation) and load_facebook_model to load full model (slower, more CPU/memory intensive, supports training continuation)
- `gensim.models.wrappers.fasttext` (obsoleted by the new native `gensim.models.fasttext` implementation)
- `gensim.examples`
- `gensim.nosy`
- `gensim.scripts.word2vec_standalone`
- `gensim.scripts.make_wiki_lemma`
- `gensim.scripts.make_wiki_online`
- `gensim.scripts.make_wiki_online_lemma`
- `gensim.scripts.make_wiki_online_nodebug`
- `gensim.scripts.make_wiki` (all of these obsoleted by the new native `gensim.scripts.segment_wiki` implementation)
- "deprecated" functions and attributes
* Move
- `gensim.scripts.make_wikicorpus` ➡ `gensim.scripts.make_wiki.py`
- `gensim.summarization` ➡ `gensim.models.summarization`
- `gensim.topic_coherence` ➡ `gensim.models._coherence`
- `gensim.utils` ➡ `gensim.utils.utils` (old imports will continue to work)
- `gensim.parsing.*` ➡ `gensim.utils.text_utils`