3.4.0

3.4.0, 2018-03-01

🌟 New features:

• Massive optimizations of gensim.models.LdaModel: much faster training, using Cython. (@arlenk, #1767)

  • Training benchmark 💥

    dataset	old LDA [sec]	optimized LDA [sec]	speed up
    nytimes	3473	1975	1.76x
    enron	774	437	1.77x

  • This change affects all models that depend on LdaModel, such as LdaMulticore, LdaSeqModel, AuthorTopicModel.

• Huge speed-ups to corpus I/O with MmCorpus (Cython) (@arlenk, #1825)

  • File reading benchmark

    dataset	file compressed?	old MmReader [sec]	optimized MmReader [sec]	speed up
    enron	no	22.3	2.6	8.7x
    	yes	37.3	14.4	2.6x
    nytimes	no	419.3	49.2	8.5x
    	yes	686.2	275.1	2.5x
    text8	no	25.4	2.5	10.1x
    	yes	41.9	17.0	2.5x

  • Overall, a 2.5x speedup for compressed .mm.gz input and 8.5x 🔥🔥🔥 for uncompressed plaintext .mm.

• Performance and memory optimization to gensim.models.FastText 🚀 (@jbaiter, #1916)

  • Benchmark (first 500,000 articles from English Wikipedia)

    Metric	old FastText	optimized FastText	improvement
    Training time (1 epoch)	4823.4s (80.38 minutes)	1873.6s (31.22 minutes)	2.57x
    Training time (full)	1h 26min 13s	36min 43s	2.35x
    Training words/sec	72,781	187,366	2.57x
    Training peak memory	5.2 GB	3.7 GB	1.4x

  • Overall, a 2.5x speedup & memory usage reduced by 30%.

• Implemented Soft Cosine Measure (@Witiko, #1827)

  • New method for assessing document similarity, a nice faster alternative to WMD, Word Mover's Distance

  • Benchmark

    Technique	MAP score	Duration
    softcossim	45.99	1.24 sec
    wmd-relax	44.48	12.22 sec
    cossim	44.22	4.39 sec
    wmd-gensim	44.08	98.29 sec

  • Soft Cosine notebook with detailed description, examples & benchmarks

  • Related papers:

    • Soft Similarity and Soft Cosine Measure: Similarity of Features in Vector Space Model
    • SimBow at SemEval-2017 Task 3: Soft-Cosine Semantic Similarity between Questions for Community Question Answering
    • Vector Space Representations in IR

👍 Improvements:

• New method to show the Gensim installation parameters: python -m gensim.scripts.package_info --info. Use this when reporting problems, for easier debugging. Fix #1902 (@sharanry, #1903)
• Added a flag to optionally skip network-related tests, to help maintainers avoid network issues with CI services (@menshikh-iv, #1930)
• Added license field to setup.py, allowing the use of tools like pip-licenses (@nils-werner, #1909)

🔴 Bug fixes:

• Fix Python 3 compatibility for gensim.corpora.UciCorpus.save_corpus (@darindf, #1875)
• Add wv property to KeyedVectors for backward compatibility. Fix #1882 (@manneshiva, #1884)
• Fix deprecation warning from inspect.getargspec. Fix #1878 (@aneesh-joshi, #1887)
• Add LabeledSentence to gensim.models.doc2vec for backward compatibility. Fix #1886 (@manneshiva, #1891)
• Fix empty output bug in Phrases (when using model[tokens] twice). Fix #1401 (@sj29-innovate, #1853)
• Fix type problems for D2VTransformer.fit_transform. Fix #1834 (@Utkarsh-Mishra-CIC, #1845)
• Fix datatype parameter for KeyedVectors.load_word2vec_format. Fix #1682 (@pushpankar, #1819)
• Fix deprecated parameters in doc2vec-lee notebook (@TheFlash10, #1918)
• Fix file-like closing bug in gensim.corpora.MmCorpus. Fix #1869 (@sj29-innovate, #1911)
• Fix precision problem in test_similarities.py, no more FP fails. (@menshikh-iv, #1928)
• Fix encoding in Lee corpus reader. (@menshikh-iv, #1931)
• Fix OOV pairs counter in WordEmbeddingsKeyedVectors.evaluate_word_pairs. (@akutuzov, #1934)

📚 Tutorial and doc improvements:

• Fix example block for gensim.models.Word2Vec (@nzw0301, #1870)
• Fix doc2vec-lee notebook (@numericlee, #1870)
• Store images from README.md directly in repository. Fix #1849 (@ibrahimsharaf, #1861)
• Add windows venv activate command to CONTRIBUTING.md (@aneesh-joshi, #1880)
• Add anaconda-cloud badge. Partial fix #1901 (@sharanry, #1905)
• Fix docstrings for lsi-related code (@steremma, #1892)
• Fix parameter description of sg parameter for gensim.models.word2vec (@mdcclv, #1919)
• Refactor documentation for gensim.similarities.docsim and MmCorpus-related. (@CLearERR & @menshikh-iv, #1910)
• Fix docstrings for gensim.test.utils (@yurkai & @menshikh-iv, #1904)
• Refactor docstrings for gensim.scripts. Partial fix #1665 (@yurkai & @menshikh-iv, #1792)
• Refactor API reference gensim.corpora. Partial fix #1671 (@CLearERR & @menshikh-iv, #1835)
• Fix documentation for gensim.models.wrappers (@kakshay21 & @menshikh-iv, #1859)
• Fix docstrings for gensim.interfaces (@yurkai & @menshikh-iv, #1913)

⚠️ 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