Move models.term_similarity, and levenshtein to similarities

gensim/matutils.py

@@ -778,7 +778,7 @@ def cossim(vec1, vec2):
 
 @deprecated(
     "Function will be removed in 4.0.0, use " +
-    "gensim.models.term_similarity.SparseTermSimilarityMatrix.inner_product instead")
+    "gensim.similarities.termsim.SparseTermSimilarityMatrix.inner_product instead")
 def softcossim(vec1, vec2, similarity_matrix):
     """Get Soft Cosine Measure between two vectors given a term similarity matrix.
 
@@ -812,7 +812,7 @@ def softcossim(vec1, vec2, similarity_matrix):
     --------
     :meth:`gensim.models.keyedvectors.WordEmbeddingsKeyedVectors.similarity_matrix`
         A term similarity matrix produced from term embeddings.
-    :func:`gensim.models.levenshtein.similarity_matrix`
+    :func:`gensim.similarities.levenshtein.similarity_matrix`
         A term similarity matrix produced from Levenshtein distances.
     :class:`gensim.similarities.docsim.SoftCosineSimilarity`
         A class for performing corpus-based similarity queries with Soft Cosine Measure.

gensim/models/__init__.py

@@ -21,10 +21,7 @@
 from .ldaseqmodel import LdaSeqModel  # noqa:F401
 from .fasttext import FastText  # noqa:F401
 from .translation_matrix import TranslationMatrix, BackMappingTranslationMatrix  # noqa:F401
-from .term_similarity import TermSimilarityIndex, UniformTermSimilarityIndex, SparseTermSimilarityMatrix  # noqa:F401
-from .levenshtein import LevenshteinSimilarityIndex  # noqa:F401
 
-from . import levenshtein  # noqa:F401
 from . import wrappers  # noqa:F401
 from . import deprecated  # noqa:F401
 

gensim/models/keyedvectors.py

@@ -84,7 +84,7 @@
 from scipy import stats
 from gensim.utils import deprecated
 from gensim.models.utils_any2vec import _save_word2vec_format, _load_word2vec_format, _compute_ngrams, _ft_hash
-from gensim.models.term_similarity import TermSimilarityIndex, SparseTermSimilarityMatrix
+from gensim.similarities.termsim import TermSimilarityIndex, SparseTermSimilarityMatrix
 
 logger = logging.getLogger(__name__)
 
@@ -538,7 +538,7 @@ def similarity_matrix(self, dictionary, tfidf=None, threshold=0.0, exponent=2.0,
             The Soft Cosine Measure.
         :class:`gensim.similarities.docsim.SoftCosineSimilarity`
             A class for performing corpus-based similarity queries with Soft Cosine Measure.
-        :func:`gensim.models.levenshtein.similarity_matrix`
+        :func:`gensim.similarities.levenshtein.similarity_matrix`
             A term similarity matrix produced from Levenshtein distances.
 
 
@@ -1079,7 +1079,7 @@ class WordEmbeddingSimilarityIndex(TermSimilarityIndex):
 
     See Also
     --------
-    :class:`~gensim.models.term_similarity.SparseTermSimilarityMatrix`
+    :class:`~gensim.similarities.termsim.SparseTermSimilarityMatrix`
         Build a term similarity matrix and compute the Soft Cosine Measure.
 
     """

gensim/similarities/__init__.py

@@ -4,3 +4,7 @@
 
 # bring classes directly into package namespace, to save some typing
 from .docsim import Similarity, MatrixSimilarity, SparseMatrixSimilarity, SoftCosineSimilarity, WmdSimilarity  # noqa:F401
+from .termsim import TermSimilarityIndex, UniformTermSimilarityIndex, SparseTermSimilarityMatrix  # noqa:F401
+from .levenshtein import LevenshteinSimilarityIndex  # noqa:F401
+
+from . import levenshtein  # noqa:F401

gensim/similarities/docsim.py

@@ -875,7 +875,7 @@ class SoftCosineSimilarity(interfaces.SimilarityABC):
     """
     @deprecated(
         "Method will be removed in 4.0.0, use " +
-        "gensim.models.term_similarity.SparseTermSimilarityMatrix.inner_product instead")
+        "gensim.similarities.termsim.SparseTermSimilarityMatrix.inner_product instead")
     def __init__(self, corpus, similarity_matrix, num_best=None, chunksize=256):
         """
 
@@ -895,7 +895,7 @@ def __init__(self, corpus, similarity_matrix, num_best=None, chunksize=256):
         --------
         :meth:`gensim.models.keyedvectors.WordEmbeddingsKeyedVectors.similarity_matrix`
             A term similarity matrix produced from term embeddings.
-        :func:`gensim.models.levenshtein.similarity_matrix`
+        :func:`gensim.similarities.levenshtein.similarity_matrix`
             A term similarity matrix produced from Levenshtein distances.
         :func:`gensim.matutils.softcossim`
             The Soft Cosine Measure.
@@ -916,13 +916,13 @@ def __init__(self, corpus, similarity_matrix, num_best=None, chunksize=256):
 
     @deprecated(
         "Method will be removed in 4.0.0, use " +
-        "gensim.models.term_similarity.SparseTermSimilarityMatrix.inner_product instead")
+        "gensim.similarities.termsim.SparseTermSimilarityMatrix.inner_product instead")
     def __len__(self):
         return len(self.corpus)
 
     @deprecated(
         "Method will be removed in 4.0.0, use " +
-        "gensim.models.term_similarity.SparseTermSimilarityMatrix.inner_product instead")
+        "gensim.similarities.termsim.SparseTermSimilarityMatrix.inner_product instead")
     def get_similarities(self, query):
         """Get similarity between `query` and current index instance.
 
@@ -969,7 +969,7 @@ def get_similarities(self, query):
 
     @deprecated(
         "Method will be removed in 4.0.0, use " +
-        "gensim.models.term_similarity.SparseTermSimilarityMatrix.inner_product instead")
+        "gensim.similarities.termsim.SparseTermSimilarityMatrix.inner_product instead")
     def __str__(self):
         return "%s<%i docs, %i features>" % (self.__class__.__name__, len(self), self.similarity_matrix.shape[0])
 

gensim/models/levenshtein.py → gensim/similarities/levenshtein.py

@@ -15,7 +15,7 @@
 from Levenshtein import distance
 from numpy import float32 as REAL
 
-from gensim.models.term_similarity import TermSimilarityIndex, SparseTermSimilarityMatrix
+from gensim.similarities.termsim import TermSimilarityIndex, SparseTermSimilarityMatrix
 from gensim.utils import deprecated
 
 logger = logging.getLogger(__name__)
@@ -80,9 +80,9 @@ class LevenshteinSimilarityIndex(TermSimilarityIndex):
 
     See Also
     --------
-    :func:`gensim.models.levenshtein.levsim`
+    :func:`gensim.similarities.levenshtein.levsim`
         The Levenshtein similarity.
-    :class:`~gensim.models.term_similarity.SparseTermSimilarityMatrix`
+    :class:`~gensim.similarities.termsim.SparseTermSimilarityMatrix`
         Build a term similarity matrix and compute the Soft Cosine Measure.
 
     """
@@ -120,7 +120,7 @@ def most_similar(self, t1, topn=10):
 
 @deprecated(
     "Function will be deprecated in 4.0.0, use " +
-    "gensim.models.levenshtein.LevenshteinSimilarityIndex instead")
+    "gensim.similarities.levenshtein.LevenshteinSimilarityIndex instead")
 def similarity_matrix(dictionary, tfidf=None, threshold=0.0, alpha=1.8, beta=5.0,
                       nonzero_limit=100, workers=1, dtype=REAL):
     """Constructs a term similarity matrix for computing Soft Cosine Measure.

gensim/models/term_similarity.py → gensim/similarities/termsim.py

@@ -27,7 +27,7 @@ class TermSimilarityIndex(SaveLoad):
 
     See Also
     --------
-    :class:`~gensim.models.term_similarity.SparseTermSimilarityMatrix`
+    :class:`~gensim.similarities.termsim.SparseTermSimilarityMatrix`
         Build a term similarity matrix and compute the Soft Cosine Measure.
 
     """
@@ -67,7 +67,7 @@ class UniformTermSimilarityIndex(TermSimilarityIndex):
 
     See Also
     --------
-    :class:`~gensim.models.term_similarity.SparseTermSimilarityMatrix`
+    :class:`~gensim.similarities.termsim.SparseTermSimilarityMatrix`
         Build a term similarity matrix and compute the Soft Cosine Measure.
 
     Notes
@@ -93,7 +93,7 @@ class SparseTermSimilarityMatrix(SaveLoad):
 
     Parameters
     ----------
-    source : :class:`~gensim.models.term_similarity.TermSimilarityIndex` or :class:`scipy.sparse.spmatrix`
+    source : :class:`~gensim.similarities.termsim.TermSimilarityIndex` or :class:`scipy.sparse.spmatrix`
         The source of the term similarity. Either a term similarity index that will be used for
         building the term similarity matrix, or an existing sparse term similarity matrix that will
         be encapsulated and stored in the matrix attribute.

gensim/test/test_levenshtein.py

@@ -6,7 +6,7 @@
 # Licensed under the GNU LGPL v2.1 - http://www.gnu.org/licenses/lgpl.html
 
 """
-Automated tests for checking the gensim.models.levenshtein module.
+Automated tests for checking the gensim.similarities.levenshtein module.
 """
 
 import logging

gensim/test/test_term_similarity.py

@@ -6,15 +6,16 @@
 # Licensed under the GNU LGPL v2.1 - http://www.gnu.org/licenses/lgpl.html
 
 """
-Automated tests for checking the gensim.models.term_similarity module.
+Automated tests for checking the gensim.similarities.termsim module.
 """
 
 import logging
 from math import sqrt
 import unittest
 
 from gensim.corpora import Dictionary
-from gensim.models import UniformTermSimilarityIndex, SparseTermSimilarityMatrix, TfidfModel
+from gensim.models import TfidfModel
+from gensim.similarities import UniformTermSimilarityIndex, SparseTermSimilarityMatrix
 
 import numpy as np
 from scipy.sparse import csc_matrix, csr_matrix