Gensim是一个Python库，用于主题建模，文档索引和大型语料库的相似性检索

3.8.1, 2019-09-23

🔴 Bug fixes

• Fix usage of base_dir instead of BASE_DIR in _load_info in downloader. (movb, #2605)
• Update the version of smart_open in the setup.py file (AMR-KELEG, #2582)
• Properly handle unicode_errors arg parameter when loading a vocab file (wmtzk, #2570)
• Catch loading older TfidfModels without smartirs (bnomis, #2559)
• Fix bug where a module import set up logging, pin doctools for Py2 (piskvorky, #2552)

📚 Tutorial and doc improvements

• Fix usage example in phrases.py (piskvorky, #2575)

👍 Improvements

• Optimize Poincare model training (koiizukag, #2589)

⚠️ 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

3.8.0, 2019-07-08

⚠️ 3.8.x will be the last Gensim version to support Py2.7. Starting with 4.0.0, Gensim will only support Py3.5 and above

🌟 New Features

• Enable online training of Poincare models (koiizukag, #2505)
• Make BM25 more scalable by adding support for generator inputs (saraswatmks, #2479)
• Allow the Gensim dataset / pre-trained model downloader gensim.downloader to run offline, by introducing a local file cache (mpenkov, #2545)
• Make the gensim.downloader target directory configurable (mpenkov, #2456)
• Support fast kNN document similarity search using NMSLIB (masa3141, #2417)

🔴 Bug fixes

• Fix smart_open deprecation warning globally (itayB, #2530)
• Fix AppVeyor issues with Windows and Py2 (mpenkov, #2546)
• Fix topn=0 versus topn=None bug in most_similar, accept topn of any integer type (Witiko, #2497)
• Fix Python version check (charsyam, #2547)
• Fix typo in FastText documentation (Guitaricet, #2518)
• Fix "Market Matrix" to "Matrix Market" typo. (Shooter23, #2513)
• Fix auto-generated hyperlinks in CHANGELOG.md (mpenkov, #2482)

📚 Tutorial and doc improvements

• Generate documentation for the gensim.similarities.termsim module (Witiko, #2485)
• Simplify the Support section in README (piskvorky, #2542)

👍 Improvements

• Pin sklearn version for Py2, because sklearn dropped py2 support (mpenkov, #2510)

⚠️ 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

3.7.3, 2019-05-06

🔴 Bug fixes

• Fix fasttext model loading from gzip files (mpenkov, #2476)
• Fix misleading Doc2Vec.docvecs comment (gojomo, #2472)
• NMF bugfix (mpenkov, #2466)
• Fix WordEmbeddingsKeyedVectors.most_similar (Witiko, #2461)
• Fix LdaSequence model by updating to num_documents (Bharat123rox, #2410)
• Make termsim matrix positive definite even with negative similarities (Witiko, #2397)
• Fix the off-by-one bug in the TFIDF model. (AMR-KELEG, #2392)
• Update legacy model loading (mpenkov, #2454, #2457)
• Make matutils.unitvec always return float norm when requested (Witiko, #2419)

📚 Tutorial and doc improvements

• Update word2vec.ipynb (asyabo, #2423)

👍 Improvements

• Adding type check for corpus_file argument (saraswatmks, #2469)
• Clean up FastText Cython code, fix division by zero (mpenkov, #2382)

⚠️ 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

3.7.2, 2019-04-06

🌟 New Features

• gensim.models.fasttext.load_facebook_model function: load full model (slower, more CPU/memory intensive, supports training continuation)

  >>> from gensim.test.utils import datapath
  >>>
  >>> cap_path = datapath("crime-and-punishment.bin")
  >>> fb_model = load_facebook_model(cap_path)
  >>>
  >>> 'landlord' in fb_model.wv.vocab  # Word is out of vocabulary
  False
  >>> oov_term = fb_model.wv['landlord']
  >>>
  >>> 'landlady' in fb_model.wv.vocab  # Word is in the vocabulary
  True
  >>> iv_term = fb_model.wv['landlady']
  >>>
  >>> new_sent = [['lord', 'of', 'the', 'rings'], ['lord', 'of', 'the', 'flies']]
  >>> fb_model.build_vocab(new_sent, update=True)
  >>> fb_model.train(sentences=new_sent, total_examples=len(new_sent), epochs=5)  

• gensim.models.fasttext.load_facebook_vectors function: load embeddings only (faster, less CPU/memory usage, does not support training continuation)

  >>> fbkv = load_facebook_vectors(cap_path)
  >>>
  >>> 'landlord' in fbkv.vocab  # Word is out of vocabulary
  False
  >>> oov_vector = fbkv['landlord']
  >>>
  >>> 'landlady' in fbkv.vocab  # Word is in the vocabulary
  True
  >>> iv_vector = fbkv['landlady']

🔴 Bug fixes

• Fix unicode error when loading FastText vocabulary (@mpenkov, #2390)
• Avoid division by zero in fasttext_inner.pyx (@mpenkov, #2404)
• Avoid incorrect filename inference when loading model (@mpenkov, #2408)
• Handle invalid unicode when loading native FastText models (@mpenkov, #2411)
• Avoid divide by zero when calculating vectors for terms with no ngrams (@mpenkov, #2411)

📚 Tutorial and doc improvements

• Add link to bindr (rogueleaderr, #2387)

👍 Improvements

• Undo the hash2index optimization (mpenkov, #2370)

⚠️ Changes in FastText behavior

Out-of-vocab word handling

To achieve consistency with the reference implementation from Facebook,
a FastText model will now always report any word, out-of-vocabulary or
not, as being in the model, and always return some vector for any word
looked-up. Specifically:

1. 'any_word' in ft_model will always return True. Previously, it
   returned True only if the full word was in the vocabulary. (To test if a
   full word is in the known vocabulary, you can consult the wv.vocab
   property: 'any_word' in ft_model.wv.vocab will return False if the full
   word wasn't learned during model training.)
2. ft_model['any_word'] will always return a vector. Previously, it
   raised KeyError for OOV words when the model had no vectors
   for any ngrams of the word.
3. If no ngrams from the term are present in the model,
   or when no ngrams could be extracted from the term, a vector pointing
   to the origin will be returned. Previously, a vector of NaN (not a number)
   was returned as a consequence of a divide-by-zero problem.
4. Models may use more more memory, or take longer for word-vector
   lookup, especially after training on smaller corpuses where the previous
   non-compliant behavior discarded some ngrams from consideration.

Loading models in Facebook .bin format

The gensim.models.FastText.load_fasttext_format function (deprecated) now loads the entire model contained in the .bin file, including the shallow neural network that enables training continuation.
Loading this NN requires more CPU and RAM than previously required.

Since this function is deprecated, consider using one of its alternatives (see below).

Furthermore, you must now pass the full path to the file to load, including the file extension.
Previously, if you specified a model path that ends with anything other than .bin, the code automatically appended .bin to the path before loading the model.
This behavior was confusing, so we removed it.

⚠️ 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

3.7.1, 2019-01-31

👍 Improvements

• NMF optimization & documentation (@anotherbugmaster, #2361)
• Optimize FastText.load_fasttext_model (@mpenkov, #2340)
• Add warning when string is used as argument to Doc2Vec.infer_vector (@tobycheese, #2347)
• Fix light linting issues in LdaSeqModel (@horpto, #2360)
• Move out process_result_queue from cycle in LdaMulticore (@horpto, #2358)

🔴 Bug fixes

• Fix infinite diff in LdaModel.do_mstep (@horpto, #2344)
• Fix backward compatibility issue: loading FastTextKeyedVectors using KeyedVectors (missing attribute compatible_hash) (@menshikh-iv, #2349)
• Fix logging issue (conda-forge related) (@menshikh-iv, #2339)
• Fix WordEmbeddingsKeyedVectors.most_similar (@Witiko, #2356)
• Fix issues of flake8==3.7.1 (@horpto, #2365)

📚 Tutorial and doc improvements

• Improve FastText documentation (@mpenkov, #2353)
• Minor corrections and improvements in Any*Vec docstrings (@tobycheese, #2345)
• Fix the example code for SparseTermSimilarityMatrix (@Witiko, #2359)
• Update poincare documentation to indicate the relation format (@AMR-KELEG, #2357)

⚠️ Deprecations (will be removed in the next major release)

• Remove

  • 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

3.7.0, 2019-01-18

🌟 New features

• Fast Online NMF (@anotherbugmaster, #2007)

  • Benchmark wiki-english-20171001

    Model	Perplexity	Coherence	L2 norm	Train time (minutes)
    LDA	4727.07	-2.514	7.372	138
    NMF	975.74	-2.814	7.265	73
    NMF (with regularization)	985.57	-2.436	7.269	441

  • Simple to use (same interface as LdaModel)

    from gensim.models.nmf import Nmf
    from gensim.corpora import Dictionary
    import gensim.downloader as api
    
    text8 = api.load('text8')
    
    dictionary = Dictionary(text8)
    dictionary.filter_extremes()
    
    corpus = [
        dictionary.doc2bow(doc) for doc in text8
    ]
    
    nmf = Nmf(
        corpus=corpus,
        num_topics=5,
        id2word=dictionary,
        chunksize=2000,
        passes=5,
        random_state=42,
    )
    
    nmf.show_topics()
    """
    [(0, '0.007*"km" + 0.006*"est" + 0.006*"islands" + 0.004*"league" + 0.004*"rate" + 0.004*"female" + 0.004*"economy" + 0.003*"male" + 0.003*"team" + 0.003*"elections"'),
     (1, '0.006*"actor" + 0.006*"player" + 0.004*"bwv" + 0.004*"writer" + 0.004*"actress" + 0.004*"singer" + 0.003*"emperor" + 0.003*"jewish" + 0.003*"italian" + 0.003*"prize"'),
     (2, '0.036*"college" + 0.007*"institute" + 0.004*"jewish" + 0.004*"universidad" + 0.003*"engineering" + 0.003*"colleges" + 0.003*"connecticut" + 0.003*"technical" + 0.003*"jews" + 0.003*"universities"'),
     (3, '0.016*"import" + 0.008*"insubstantial" + 0.007*"y" + 0.006*"soviet" + 0.004*"energy" + 0.004*"info" + 0.003*"duplicate" + 0.003*"function" + 0.003*"z" + 0.003*"jargon"'),
     (4, '0.005*"software" + 0.004*"games" + 0.004*"windows" + 0.003*"microsoft" + 0.003*"films" + 0.003*"apple" + 0.003*"video" + 0.002*"album" + 0.002*"fiction" + 0.002*"characters"')]
    """

  • See also:

    • NMF tutorial
    • Full NMF Benchmark

• Massive improvement of FastText compatibilities (@mpenkov, #2313)

  from gensim.models import FastText
  
  # 'cc.ru.300.bin' - Russian Facebook FT model trained on Common Crawl
  # Can be downloaded from https://s3-us-west-1.amazonaws.com/fasttext-vectors/word-vectors-v2/cc.ru.300.bin.gz
  
  model = FastText.load_fasttext_format("cc.ru.300.bin")
  
  # Fixed hash-function allow to produce same output as FB FastText & works correctly for non-latin languages (for example, Russian)
  assert "мяу" in m.wv.vocab  # 'мяу' - vocab word
  model.wv.most_similar("мяу")
  """
  [('Мяу', 0.6820122003555298),
   ('МЯУ', 0.6373013257980347),
   ('мяу-мяу', 0.593108594417572),
   ('кис-кис', 0.5899622440338135),
   ('гав', 0.5866007804870605),
   ('Кис-кис', 0.5798211097717285),
   ('Кис-кис-кис', 0.5742273330688477),
   ('Мяу-мяу', 0.5699705481529236),
   ('хрю-хрю', 0.5508339405059814),
   ('ав-ав', 0.5479759573936462)]
  """
  
  assert "котогород" not in m.wv.vocab  # 'котогород' - out-of-vocab word
  model.wv.most_similar("котогород", topn=3)
  """
  [('автогород', 0.5463314652442932),
   ('ТагилНовокузнецкНовомосковскНовороссийскНовосибирскНовотроицкНовочеркасскНовошахтинскНовый',
    0.5423436164855957),
   ('областьНовосибирскБарабинскБердскБолотноеИскитимКарасукКаргатКуйбышевКупиноОбьТатарскТогучинЧерепаново',
    0.5377570390701294)]
  """
  
  # Now we load full model, for this reason, we can continue an training
  
  from gensim.test.utils import datapath
  from smart_open import smart_open
  
  with smart_open(datapath("crime-and-punishment.txt"), encoding="utf-8") as infile:  # russian text
      corpus = [line.strip().split() for line in infile]
  
  model.train(corpus, total_examples=len(corpus), epochs=5)

• Similarity search improvements (@Witiko, #2016)

  • Add similarity search using the Levenshtein distance in gensim.similarities.LevenshteinSimilarityIndex

  • Performance optimizations to gensim.similarities.SoftCosineSimilarity (full benchmark)

    dictionary size	corpus size	speed
    1000	100	1.0×
    1000	1000	53.4×
    1000	100000	156784.8×
    100000	100	3.8×
    100000	1000	405.8×
    100000	100000	66262.0×

  • See updated soft-cosine tutorial for more information and usage examples

• Add python3.7 support (@menshikh-iv, #2211)

  • Wheels for Window, OSX and Linux platforms (@menshikh-iv, MacPython/gensim-wheels/#12)
  • Faster installation

👍 Improvements

Optimizations

• Reduce Phraser memory usage (drop frequencies) (@jenishah, #2208)
• Reduce memory consumption of summarizer (@horpto, #2298)
• Replace inline slow equivalent of mean_absolute_difference with fast (@horpto, #2284)
• Reuse precalculated updated prior in ldamodel.update_dir_prior (@horpto, #2274)
• Improve KeyedVector.wmdistance (@horpto, #2326)
• Optimize remove_unreachable_nodes in gensim.summarization (@horpto, #2263)
• Optimize mz_entropy from gensim.summarization (@horpto, #2267)
• Improve filter_extremes methods in Dictionary and HashDictionary (@horpto, #2303)

Additions

• Add KeyedVectors.relative_cosine_similarity (@rsdel2007, #2307)
• Add random_seed to LdaMallet (@Zohaggie & @menshikh-iv, #2153)
• Add common_terms parameter to sklearn_api.PhrasesTransformer (@pmlk, #2074)
• Add method for patch corpora.Dictionary based on special tokens (@Froskekongen, #2200)

Cleanup

• Improve six usage (xrange, map, zip) (@horpto, #2264)
• Refactor line2doc methods of LowCorpus and MalletCorpus (@horpto, #2269)
• Get rid most of warnings in testing (@menshikh-iv, #2191)
• Fix non-deterministic test failures (pin PYTHONHASHSEED) (@menshikh-iv, #2196)
• Fix "aliasing chunkize to chunkize_serial" warning on Windows (@aquatiko, #2202)
• Remove __getitem__ code duplication in gensim.models.phrases (@jenishah, #2206)
• Add flake8-rst for docstring code examples (@kataev, #2192)
• Get rid py26 stuff (@menshikh-iv, #2214)
• Use itertools.chain instead of sum to concatenate lists (@Stigjb, #2212)
• Fix flake8 warnings W605, W504 (@horpto, #2256)
• Remove unnecessary creations of lists at all (@horpto, #2261)
• Fix extra list creation in utils.get_max_id (@horpto, #2254)
• Fix deprecation warning np.sum(generator) (@rsdel2007, #2296)
• Refactor BM25 (@horpto, #2275)
• Fix pyemd import (@ramprakash-94, #2240)
• Set metadata=True for make_wikicorpus script by default (@Xinyi2016, #2245)
• Remove unimportant warning from Phrases (@rsdel2007, #2331)
• Replace open() by smart_open() in gensim.models.fasttext._load_fasttext_format (@rsdel2007, #2335)

🔴 Bug fixes

• Fix overflow error for *Vec corpusfile-based training (@bm371613, #2239)
• Fix malletmodel2ldamodel conversion (@horpto, #2288)
• Replace custom epsilons with numpy equivalent in LdaModel (@horpto, #2308)
• Add missing content to tarball (@menshikh-iv, #2194)
• Fixes divided by zero when w_star_count==0 (@allenyllee, #2259)
• Fix check for callbacks (@allenyllee, #2251)
• Fix SvmLightCorpus.serialize if labels instance of numpy.ndarray (@aquatiko, #2243)
• Fix poincate viz incompatibility with plotly>=3.0.0 (@jenishah, #2226)
• Fix keep_n behavior for Dictionary.filter_extremes (@johann-petrak, #2232)
• Fix for sphinx==1.8.1 (last r (@menshikh-iv, #None)
• Fix np.issubdtype warnings (@marioyc, #2210)
• Drop wrong key -c from gensim.downloader description (@horpto, #2262)
• Fix gensim build (docs & pyemd issues) (@menshikh-iv, #2318)
• Limit visdom version (avoid py2 issue from the latest visdom release) (@menshikh-iv, #2334)
• Fix visdom integration (using viz.line() instead of viz.updatetrace()) (@allenyllee, #2252)

📚 Tutorial and doc improvements

• Add gensim-data repo to gensim.downloader & fix rendering of code examples (@menshikh-iv, #2327)
• Fix typos in gensim.models (@rsdel2007, #2323)
• Fixed typos in notebooks (@rsdel2007, #2322)
• Update Doc2Vec documentation: how tags are assigned in corpus_file mode (@persiyanov, #2320)
• Fix typos in gensim/models/keyedvectors.py (@rsdel2007, #2290)
• Add documentation about ranges to scoring functions for Phrases (@jenishah, #2242)
• Update return sections for KeyedVectors.evaluate_word_* (@Stigjb, #2205)
• Fix return type in KeyedVector.evaluate_word_analogies (@Stigjb, #2207)
• Fix WmdSimilarity documentation (@jagmoreira, #2217)
• Replace fify -> fifty in gensim.parsing.preprocessing.STOPWORDS (@coderwassananmol, #2220)
• Remove alpha="auto" from LdaMulticore (not supported yet) (@johann-petrak, #2225)
• Update Adopters in README (@piskvorky, #2234)
• Fix broken link in tutorials.md (@rsdel2007, #2302)

⚠️ Deprecations (will be removed in the next major release)

• Remove

  • 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

3.6.0, 2018-09-20

🌟 New features

• File-based training for *2Vec models (@persiyanov, #2127 & #2078 & #2048)

  Blog post / Jupyter tutorial.

  New training mode for *2Vec models (word2vec, doc2vec, fasttext) that allows model training to scale linearly with the number of cores (full GIL elimination). The result of our Google Summer of Code 2018 project by Dmitry Persiyanov.

  Benchmark on the full English Wikipedia, Intel(R) Xeon(R) CPU @ 2.30GHz 32 cores (GCE cloud), MKL BLAS:

  Model	Queue-based version [sec]	File-based version [sec]	speed up	Accuracy (queue-based)	Accuracy (file-based)
  Word2Vec	9230	2437	3.79x	0.754 (± 0.003)	0.750 (± 0.001)
  Doc2Vec	18264	2889	6.32x	0.721 (± 0.002)	0.683 (± 0.003)
  FastText	16361	10625	1.54x	0.642 (± 0.002)	0.660 (± 0.001)

  Usage:

  import gensim.downloader as api
  from multiprocessing import cpu_count
  from gensim.utils import save_as_line_sentence
  from gensim.test.utils import get_tmpfile
  from gensim.models import Word2Vec, Doc2Vec, FastText
  
  
  # Convert any corpus to the needed format: 1 document per line, words delimited by " "
  corpus = api.load("text8")
  corpus_fname = get_tmpfile("text8-file-sentence.txt")
  save_as_line_sentence(corpus, corpus_fname)
  
  # Choose num of cores that you want to use (let's use all, models scale linearly now!)
  num_cores = cpu_count()
  
  # Train models using all cores
  w2v_model = Word2Vec(corpus_file=corpus_fname, workers=num_cores)
  d2v_model = Doc2Vec(corpus_file=corpus_fname, workers=num_cores)
  ft_model = FastText(corpus_file=corpus_fname, workers=num_cores)
  

  Read notebook tutorial with full description.

👍 Improvements

• Add scikit-learn wrapper for FastText (@mcemilg, #2178)
• Add multiprocessing support for BM25 (@Shiki-H, #2146)
• Add name_only option for downloader api (@aneesh-joshi, #2143)
• Make word2vec2tensor script compatible with python3 (@vsocrates, #2147)
• Add custom filter for Wikicorpus (@mattilyra, #2089)
• Make similarity_matrix support non-contiguous dictionaries (@Witiko, #2047)

🔴 Bug fixes

• Fix memory consumption in AuthorTopicModel (@philipphager, #2122)
• Correctly process empty documents in AuthorTopicModel (@probinso, #2133)
• Fix ZeroDivisionError keywords issue with short input (@LShostenko, #2154)
• Fix min_count handling in phrases detection using npmi_scorer (@lopusz, #2072)
• Remove duplicate count from Phraser log message (@robguinness, #2151)
• Replace np.integer -> np.int in AuthorTopicModel (@menshikh-iv, #2145)

📚 Tutorial and doc improvements

• Update docstring with new analogy evaluation method (@akutuzov, #2130)
• Improve prune_at parameter description for gensim.corpora.Dictionary (@yxonic, #2128)
• Fix default -> auto prior parameter in documentation for lda-related models (@Laubeee, #2156)
• Use heading instead of bold style in gensim.models.translation_matrix (@nzw0301, #2164)
• Fix quote of vocabulary from gensim.models.Word2Vec (@nzw0301, #2161)
• Replace deprecated parameters with new in docstring of gensim.models.Doc2Vec (@xuhdev, #2165)
• Fix formula in Mallet documentation (@Laubeee, #2186)
• Fix minor semantic issue in docs for Phrases (@RunHorst, #2148)
• Fix typo in documentation (@KenjiOhtsuka, #2157)
• Additional documentation fixes (@piskvorky, #2121)

⚠️ Deprecations (will be removed in the next major release)

• Remove

  • 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