Merge pull request #139 from stephenhky/readthedocs

pin requirements for ReadTheDocs

.circleci/config.yml

@@ -28,11 +28,6 @@ shared: &shared
 
 
 jobs:
-  py37:
-    <<: *shared
-    docker:
-      - image: cimg/python:3.7
-
   py38:
     <<: *shared
     docker:
@@ -58,7 +53,6 @@ workflows:
   version: 2
   build:
     jobs:
-      - py37
       - py38
       - py39
       - py310

.readthedocs.yml

@@ -9,6 +9,11 @@ version: 2
 sphinx:
   configuration: docs/conf.py
 
+build:
+  os: ubuntu-22.04
+  tools:
+    python: "3.8"
+
 # Build documentation with MkDocs
 #mkdocs:
 #  configuration: mkdocs.yml
@@ -18,11 +23,8 @@ formats: all
 
 # Optionally set the version of Python and requirements required to build your docs
 python:
-  version: 3.7
   install:
-    - requirements: requirements.txt
-    - method: pip
-      path: .
+    - requirements: docs/requirements.txt
 
 # conda environment
 #conda:

README.md

@@ -18,15 +18,7 @@ representation of the texts and documents are needed before they are put into
 any classification algorithm. In this package, it facilitates various types
 of these representations, including topic modeling and word-embedding algorithms.
 
-Since release 1.5.2, it runs on Python 3.9.
-Since release 1.5.0, support for Python 3.6 was decommissioned.
-Since release 1.2.4, it runs on Python 3.8.
-Since release 1.2.3, support for Python 3.5 was decommissioned. 
-Since release 1.1.7, support for Python 2.7 was decommissioned.
-Since release 1.0.8, it runs on Python 3.7 with 'TensorFlow' being the backend for `keras`.
-Since release 1.0.7, it runs on Python 3.7 as well, but the backend for `keras` cannot be `TensorFlow`.
-Since release 1.0.0, `shorttext` runs on Python 2.7, 3.5, and 3.6.
-
+The package `shorttext` runs on Python 3.8, 3.9, 3.10, and 3.11.
 Characteristics:
 
 - example data provided (including subject keywords and NIH RePORT);
@@ -92,6 +84,7 @@ If you would like to contribute, feel free to submit the pull requests. You can
 
 ## News
 
+* 08/26/2023: `shorttext` 1.6.0 released.
 * 06/19/2023: `shorttext` 1.5.9 released.
 * 09/23/2022: `shorttext` 1.5.8 released.
 * 09/22/2022: `shorttext` 1.5.7 released.

docs/conf.py

@@ -56,9 +56,9 @@
 # built documents.
 #
 # The short X.Y version.
-version = u'1.5'
+version = u'1.6'
 # The full version, including alpha/beta/rc tags.
-release = u'1.5.9'
+release = u'1.6.0'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

docs/intro.rst

@@ -8,12 +8,7 @@ representation of the texts and documents are needed before they are put into
 any classification algorithm. In this package, it facilitates various types
 of these representations, including topic modeling and word-embedding algorithms.
 
-The package `shorttext` runs on Python 3.7, 3.8, and 3.9.
-
-Since release 1.0.0, `shorttext` runs on Python 2.7, 3.5, and 3.6. Since release 1.0.7,
-it runs also in Python 3.7. Since release 1.1.7, the support for Python 2.7 was decommissioned.
-Since release 1.2.3, the support for Python 3.5 is decommissioned.
-Since release 1.5.0, the support for Python 3.6 is decommissioned.
+The package `shorttext` runs on Python 3.8, 3.9, 3.10, and 3.11.
 
 Characteristics:
 
@@ -35,7 +30,7 @@ Before release 0.7.2, part of the package was implemented using C, and it is int
 Python using SWIG_ (Simplified Wrapper and Interface Generator). Since 1.0.0, these implementations
 were replaced with Cython_.
 
-Author: Kwan-Yuet Ho (LinkedIn_, ResearchGate_, Twitter_)
+Author: Kwan Yuet Stephen Ho (LinkedIn_, ResearchGate_, Twitter_)
 
 Home: :doc:`index`
 

docs/news.rst

@@ -1,6 +1,7 @@
 News
 ====
 
+* 08/26/2023: `shorttext` 1.6.0 released.
 * 06/19/2023: `shorttext` 1.5.9 released.
 * 09/23/2022: `shorttext` 1.5.8 released.
 * 09/22/2022: `shorttext` 1.5.7 released.
@@ -78,6 +79,14 @@ News
 What's New
 ----------
 
+Released 1.6.0 (August 26, 2023)
+--------------------------------
+
+* Pinned requirements for ReadTheDocs documentation;
+* Fixed bugs in word-embedding model mean pooling classifiers;
+* Updated package requirements.
+
+
 Release 1.5.9 (June 19, 2023)
 -----------------------------
 

docs/requirements.txt

@@ -0,0 +1,14 @@
+Cython==3.0.0
+numpy==1.23.3
+scipy==1.10.1
+joblib==1.3.0
+scikit-learn==1.2.0
+tensorflow==2.13.0
+keras==2.13.1
+gensim==4.0.0
+pandas==1.2.4
+snowballstemmer==2.1.0
+transformers==4.32.0
+torch==2.0.1
+python-Levenshtein==0.21.1
+numba==0.57.1

pyproject.toml

@@ -1,2 +1,2 @@
 [build-system]
-requires = ["setuptools", "wheel", "Cython>=0.29", "numpy >= 1.16"]
+requires = ["setuptools", "wheel", "Cython>=3.0.0", "numpy >= 1.23.3"]

requirements.txt

@@ -1,15 +1,14 @@
-Cython>=0.29.0
-numpy>=1.16.0
-scipy>=1.6.0
-joblib>=0.14
-scikit-learn>=0.22.0
-tensorflow>=2.5.0
-keras>=2.4.0
+Cython>=3.0.0
+numpy>=1.23.3
+scipy>=1.10.0
+joblib>=1.3.0
+scikit-learn>=1.2.0
+tensorflow>=2.13.0
+keras>=2.13.0
 gensim>=4.0.0
-pandas>=1.0.0
-Flask>=2.0.0
+pandas>=1.2.0
 snowballstemmer>=2.0.0
-transformers>=4.1.0
-torch>=1.5.0
-python-Levenshtein>=0.12.0
-numba>=0.52.0
+transformers>=4.32.0
+torch>=2.0.0
+python-Levenshtein>=0.21.0
+numba>=0.57.0

setup.py

@@ -1,11 +1,17 @@
 
 from setuptools import setup
 import numpy as np
-from Cython.Build import cythonize
 
-
-ext_modules = cythonize(['shorttext/metrics/dynprog/dldist.pyx',
-                         'shorttext/metrics/dynprog/lcp.pyx'])
+try:
+    from Cython.Build import cythonize
+    ext_modules = cythonize(['shorttext/metrics/dynprog/dldist.pyx',
+                             'shorttext/metrics/dynprog/lcp.pyx'])
+except ImportError:
+    from setuptools import Extension
+    ext_modules = [
+        Extension('shorttext.metrics.dynprog.dldist', ['shorttext/metrics/dynprog/dldist.c']),
+        Extension('shorttext.metrics.dynprog.lcp', ['shorttext/metrics/dynprog/lcp.c'])
+    ]
 
 
 def package_description():
@@ -28,7 +34,7 @@ def test_requirements():
 
 
 setup(name='shorttext',
-      version='1.5.9',
+      version='1.6.0',
       description="Short Text Mining",
       long_description=package_description(),
       long_description_content_type='text/markdown',
@@ -37,10 +43,10 @@ def test_requirements():
           "Topic :: Scientific/Engineering :: Mathematics",
           "Topic :: Text Processing :: Linguistic",
           "Topic :: Software Development :: Libraries :: Python Modules",
-          "Programming Language :: Python :: 3.7",
           "Programming Language :: Python :: 3.8",
           "Programming Language :: Python :: 3.9",
           "Programming Language :: Python :: 3.10",
+          "Programming Language :: Python :: 3.11",
           "Programming Language :: Cython",
           "Programming Language :: C",
           "Natural Language :: English",
@@ -52,7 +58,7 @@ def test_requirements():
       ],
       keywords="shorttext natural language processing text mining",
       url="https://github.com/stephenhky/PyShortTextCategorization",
-      author="Kwan-Yuet Ho",
+      author="Kwan Yuet Stephen Ho",
       author_email="[email protected]",
       license='MIT',
       ext_modules=ext_modules,

shorttext/classifiers/embed/nnlib/VarNNEmbedVecClassification.py

@@ -78,7 +78,7 @@ def convert_trainingdata_matrix(self, classdict):
         for i in range(len(phrases)):
             for j in range(min(self.maxlen, len(phrases[i]))):
                 train_embedvec[i, j] = self.word_to_embedvec(phrases[i][j])
-        indices = np.array(indices, dtype=np.int)
+        indices = np.array(indices, dtype=np.int_)
 
         return classlabels, train_embedvec, indices