Merge with develop (#1)

* Add wordnet mammal train file for Poincare notebook (piskvorky#1781)

* Adds wordnet mammal train file

* Adds link to data file in notebook

* Fix tox.ini/setup.cfg configuration (piskvorky#1815)

* update according to new pytest_benchmark version

* update wheel-storage url

* use only twine

* Fix docstrings for `gensim.utils` (piskvorky#1797)

* Add docstrings in numpy-style fromat

* fix PEP8

* remove outdated "hack" (smart_open is core dependency right now)

* fix docstrings[1]

* remove unused internal class

* fix docstrings[2]

* fix docstrings[3]

* fix docstrings[4]

* fix docstrings[5]

* fix docstrings[6]

* fix docstrings[7]

* fix docstrings[8]

* add missing `pattern` to doc dependencies

* fix docstrings[9]

* fix docstrings[10]

* Fix docstrings for `gensim.models.rpmodel` (piskvorky#1802)

* first attempt to convert few lines into numpy-style doc

* added parameters in documentation

* more documentation

* few corrections

* show inheritance and undoc members

* show special members

* example is executable now

* link to the paper added, named parameters

* fixed doc

* fixed doc

* fixed whitespaces

* fix docstrings & PEP8

* fix docstrings

* fix typo

* Fix docstrings for `gensim.models.translation_matrix` (piskvorky#1806)

* convert Space class doc to numpy style

* fix docstrings[1]

* fix docstrings[2]

* remove useless load

* fix docstrings[3]

* add missing import

* fix docstrings[4]

* Add CircleCI for build documentation. Fix piskvorky#1807 (piskvorky#1822)

* init config for circle

* change

* rm cache, install tox distinctly

* fix indentation & command

* update venv

* add pip-cache

* add apt packages for latex

* rename

* enable latex rendering

* remove doc building from Travis

* store new doc version

* Refactor API reference `gensim.topic_coherence`. Fix piskvorky#1669 (piskvorky#1714)

* Refactored aggregation

* Micro-Fix for aggregation.py, partially refactored direct_confirmation.py

* Partially refactored indirect_confirmation_measure

* Some additions

* Math attempts

* add math extension for sphinx

* Minor refactoring

* Some refactoring for probability_estimation

* Beta-strings

* Different additions

* Minor changes

* text_analysis left

* Added example for ContextVectorComputer class

* probability_estimation 0.9

* beta_version

* Added some examples for text_analysis

* text_analysis: corrected example for class UsesDictionary

* Final additions for text_analysis.py

* fix cross-reference problem

* fix pep8

* fix aggregation

* fix direct_confirmation_measure

* fix types in direct_confirmation_measure

* partial fix indirect_confirmation_measure

* HotFix for probability_estimation and segmentation

* Refactoring for probability_estimation

* Changes for indirect_confirmation_measure

* Fixed segmentation, partly fixed text_analysis

* Add Notes for text_analysis

* fix di/ind

* fix doc examples in probability_estimation

* fix probability_estimation

* fix segmentation

* fix docstring in probability_estimation

* partial fix test_analysis

* add latex stuff for docs build

* doc fix[1]

* doc fix[2]

* remove apt install from travis (now doc build in circle)

* Fix docstrings for `gensim.models.normmodel` (piskvorky#1805)

* First edits

* changed bow

* Added examples

* Final commit of the night

* Still struggling with docs

* Removed examples but still struggling with documentation

* fix docstring

* fix docstring[2]

* Fix docstrings for `gensim.models.logentropy_model` (piskvorky#1803)

* improve and correct documentation of models/logentropy_model

* include fixes according to comments

* implement fixes suggested

* associate methods with examples.

* fix minor typos

* doc fix

* Fix docstrings for `gensim.matutils` (piskvorky#1804)

* numpy style documentation on matutils.py

* doc fix[1]

* doc fix[2]

* doc fix[3]

* doc fix[4]

* doc fix[5]

* doc fix[6]

* Fix formula in `gensim.summarization.bm25`. Fix piskvorky#1828 (piskvorky#1833)

* bm25 scoring function updated

* Fixes piskvorky#1828

* Fixes piskvorky#1828

* Fixes piskvorky#1828

* Fixes piskvorky#1828

* Fixes piskvorky#1828

* Fixes piskvorky#1828 , Tests added

* Fixes piskvorky#1828 , Tests added

* Fixes piskvorky#1828 , Tests Added

* Fixes piskvorky#1828 , Tests Added

* Fixes piskvorky#1828 , Tests Added

* Fixes piskvorky#1828 , Tests Added

* Fixes piskvorky#1828

* Refactor tests for `gensim.corpora.WikiCorpus`(piskvorky#1821)

* minor style refactoring and comment fixes in accordance to PEP8

* Created test data in legitimate compressed XML format (.xml.bz2) for the WikiCorpus class.

* Used the same raw data found for other sources (9 articles).

* Added Various wiki markup to test the parsing regural expressions

* Added test class for the WikiCorpus source.

* Following the same inheritance schema as in the source TestWikiCorpus > TestTextCorpus > CorpusTestCase.

* Testing methods are overriden where necessary to reflect logic changes.

* All existing functionality is tested (account for markup handling, minimum article length etc)

* Fix python 3 compatibility for generator next method

* code review corrections

* Moved WikiCorpus tests from test/test_wikicorpus.py into its class within the test_corpora.py file.

* Adapted all old tests to the new class

* Current Test class schema ensures that WikiCorpus also passes tests defined in parents

* Deleted test_wikicorpus.py since it is now redundant

* Discarded the empty input test for the WikiCorpus since an empty file is not legitimate XML

* Added 2 more tests

* Fix parameter setting for `FastText.train`. Fix piskvorky#1818 (piskvorky#1837)

* bm25 scoring function updated

* Fixes piskvorky#1828

* Fixes piskvorky#1828

* Fixes piskvorky#1828

* Fixes piskvorky#1828

* Fixes piskvorky#1828

* Fixes piskvorky#1828 , Tests added

* Fixes piskvorky#1828 , Tests added

* Fixes piskvorky#1828 , Tests Added

* Fixes piskvorky#1828 , Tests Added

* Fixes piskvorky#1828 , Tests Added

* Fixes piskvorky#1828 , Tests Added

* Fixes piskvorky#1828

* Function Parameters corrected , Fixes piskvorky#1818

* add missing params + add supercall

* Fix positional params used for `gensim.models.CoherenceModel` in `gensim.models.callbacks` (piskvorky#1823)

* add keyword params for call to gensim.models.CoherenceModel as positional arguments for coherence and topn were incorrect due to skipping param for keyed_vectors

* Fix PEP8

.circleci/config.yml

@@ -0,0 +1,44 @@
+version: 2
+jobs:
+  build:
+    docker:
+      - image: circleci/python:2.7
+
+    working_directory: ~/gensim
+
+    steps:
+      - checkout
+
+      - restore_cache:
+          key: pip-cache
+
+      - run:
+          name: Apt install (for latex render)
+          command: |
+            sudo apt-get -yq update
+            sudo apt-get -yq remove texlive-binaries --purge
+            sudo apt-get -yq --no-install-suggests --no-install-recommends --force-yes install dvipng texlive-latex-base texlive-latex-extra texlive-latex-recommended texlive-latex-extra texlive-fonts-recommended latexmk
+
+      - run:
+          name: Basic installation (tox)
+          command: |
+            python -m virtualenv venv
+            source venv/bin/activate
+            pip install tox
+
+      - run:
+          name: Build documentation
+          command: |
+            source venv/bin/activate
+            tox -e docs -vv
+
+      - store_artifacts:
+          path:  docs/src/_build
+          destination: documentation
+
+      - save_cache:
+          key: pip-cache
+          paths:
+            - "~/.cache/pip"
+            - "~/.ccache"
+            - "~/.pip-cache"

.gitignore

@@ -72,4 +72,5 @@ data
 *_out.txt
 *.html
 *.inv
-*.js
+*.js
+docs/_images/

.travis.yml

@@ -13,7 +13,7 @@ language: python
 matrix:
   include:
     - python: '2.7'
-      env: TOXENV="flake8, docs"
+      env: TOXENV="flake8"
 
     - python: '2.7'
       env: TOXENV="py27-linux"

docs/notebooks/Poincare Evaluation.ipynb

@@ -289,6 +289,9 @@
    "outputs": [],
    "source": [
     "# Prepare the WordNet data\n",
+    "# Can also be downloaded directly from -\n",
+    "# https://github.com/jayantj/gensim/raw/wordnet_data/docs/notebooks/poincare/data/wordnet_noun_hypernyms.tsv\n",
+    "\n",
     "wordnet_file = os.path.join(data_directory, 'wordnet_noun_hypernyms.tsv')\n",
     "if not os.path.exists(wordnet_file):\n",
     "    ! python {parent_directory}/{cpp_repo_name}/scripts/create_wordnet_noun_hierarchy.py {wordnet_file}"