Sandbox Python project for some text stats. Will be applied to my gallery yaml files.

Using Python 2. Only mining.py requires Anaconda.

Actually, maybe use easy_install? No sudo required.

pip install textstat

pip install nltk

>>> import nltk

>>> nltk.download('punkt')

>>> nltk.download('english')

pip install textblob

pip install pyaml

If Anaconda is installed, the following must run.

conda install -c conda-forge textblob

conda install -c mlgill textstat

This will analyse a local file.

python analyse.py

This will analyse all galleries, assuming they are under the default directory. It will write statistics for each gallery.

python analyse_all.py

Make it verbose to see problems:

python analyse_all.py -v

You can store the output as CSV:

python analyse_all.py -o /tmp/galleries.csv

Use --help for all options.

python analyse_all.py --help

This will do some Pandas data mining on the file generated by the previous script.

This will analyse all galleries as one text corpus.

python analyse_all.py

Unlike other algorithms here, this will make an API call to analyse the difficulty of each word separately.

This will run basic methods against a static text.

python tests.py