pip install 装上了pyLDAvis
pipinstallpyLDAvis-ihttp://pypi.douban.com/simple--trusted-host
(base) C:\Users\Administrator>pipinstallpyLDAvis-ihttp://pypi.douban.com/simple--trusted-hostpypi.douban.comLookinginindexes: http://pypi.douban.com/simpleCollectingpyLDAvisDownloadinghttp://pypi.doubanio.com/packages/03/a5/15a0da6b0150b8b68610cc78af80364a80a9a4c8b6dd5ee549b8989d4b60/pyLDAvis-3.3.1.tar.gz (1.7MB) |████████████████████████████████|1.7MB6.8MB/sInstallingbuilddependencies ... doneGettingrequirementstobuildwheel ... doneInstallingbackenddependencies ... donePreparingwheelmetadata ... doneRequirementalreadysatisfied: pandas>=1.2.0ind:\python\anaconda3\lib\site-packages (frompyLDAvis) (1.3.4) Requirementalreadysatisfied: futureind:\python\anaconda3\lib\site-packages (frompyLDAvis) (0.18.2) Requirementalreadysatisfied: numpy>=1.20.0ind:\python\anaconda3\lib\site-packages (frompyLDAvis) (1.20.3) Requirementalreadysatisfied: scipyinc:\users\administrator\appdata\roaming\python\python38\site-packages (frompyLDAvis) (1.6.0) Requirementalreadysatisfied: joblibind:\python\anaconda3\lib\site-packages (frompyLDAvis) (1.0.1) Requirementalreadysatisfied: gensimind:\python\anaconda3\lib\site-packages (frompyLDAvis) (3.8.3) Requirementalreadysatisfied: scikit-learnind:\python\anaconda3\lib\site-packages (frompyLDAvis) (0.23.1) Requirementalreadysatisfied: jinja2ind:\python\anaconda3\lib\site-packages (frompyLDAvis) (2.7.2) Requirementalreadysatisfied: setuptoolsind:\python\anaconda3\lib\site-packages (frompyLDAvis) (49.2.0.post20200714) CollectingsklearnDownloadinghttp://pypi.doubanio.com/packages/1e/7a/dbb3be0ce9bd5c8b7e3d87328e79063f8b263b2b1bfa4774cb1147bfcd3f/sklearn-0.0.tar.gz (1.1kB) Requirementalreadysatisfied: numexprind:\python\anaconda3\lib\site-packages (frompyLDAvis) (2.7.1) Requirementalreadysatisfied: funcyind:\python\anaconda3\lib\site-packages (frompyLDAvis) (1.17) Requirementalreadysatisfied: pytz>=2017.3ind:\python\anaconda3\lib\site-packages (frompandas>=1.2.0->pyLDAvis) (2020.1) Requirementalreadysatisfied: python-dateutil>=2.7.3ind:\python\anaconda3\lib\site-packages (frompandas>=1.2.0->pyLDAvis) (2.8.1) Requirementalreadysatisfied: six>=1.5.0ind:\python\anaconda3\lib\site-packages (fromgensim->pyLDAvis) (1.15.0) Requirementalreadysatisfied: smart-open>=1.8.1ind:\python\anaconda3\lib\site-packages (fromgensim->pyLDAvis) (4.1.2) Requirementalreadysatisfied: Cython==0.29.14ind:\python\anaconda3\lib\site-packages (fromgensim->pyLDAvis) (0.29.14) Requirementalreadysatisfied: threadpoolctl>=2.0.0ind:\python\anaconda3\lib\site-packages (fromscikit-learn->pyLDAvis) (2.1.0) Requirementalreadysatisfied: markupsafeind:\python\anaconda3\lib\site-packages (fromjinja2->pyLDAvis) (2.0.1) Buildingwheelsforcollectedpackages: pyLDAvis, sklearnBuildingwheelforpyLDAvis (PEP517) ... doneCreatedwheelforpyLDAvis: filename=pyLDAvis-3.3.1-py2.py3-none-any.whlsize=136900sha256=e23c194c0ef03967d487011208dede4141f5f6a163a77111283455f230c697faStoredindirectory: c:\users\administrator\appdata\local\pip\cache\wheels\b5\ee\b2\29f82d7103ba90942d31cdeb29372b27fb74dbe7ff535cc081Buildingwheelforsklearn (setup.py) ... doneCreatedwheelforsklearn: filename=sklearn-0.0-py2.py3-none-any.whlsize=1320sha256=9a20c412366931bdd7ca5bad4a82cdac502d9414a32a5320641b1898e633cd6eStoredindirectory: c:\users\administrator\appdata\local\pip\cache\wheels\c5\c2\a1\e36638731a4ac05326b1bf08abc0d79c19ba07700cf6b5d648SuccessfullybuiltpyLDAvissklearnInstallingcollectedpackages: sklearn, pyLDAvisSuccessfullyinstalledpyLDAvis-3.3.1sklearn-0.0
本以为开开心心地就能用了，然而一运行代码报错：
卸载刚装的 pyLDAvis==3.3.1
安装指定版本 pyLDAvis==2.1.2。
pipinstallpyLDAvis==2.1.2-ihttp://pypi.douban.com/simple--trusted-hostpypi.douban.com
或者不用安装旧版本，用新版本的方法：
importpyLDAvis.gensim_modelsasgensimvisfrompyLDAvisimportgensimpyLDAvis.enable_notebook() '''lda: 计算好的话题模型corpus: 文档词频矩阵dictionary: 词语空间'''d=gensim.prepare(lda,corpus,dictionary) d=gensimvis.prepare(lda,corpus,dictionary) pyLDAvis.display(d) # 展示在浏览器# pyLDAvis.save_html(p, 'lda.html')
利用 pyLDAvis.save_html(p, ‘lda.html’) 方法可以将可视化结果保存为单独的 HTML 文件。
参考了：
关于python包安装错误解决办法（以gensim、pyLDAvis为例）
