Requires a version of the LIWC dictionary in .dic format. Still very much work in process. Results might differ (albeit marginally) from the commercial LIWC software.
For the full documentation look here
pipenv install -e git+https://github.com/dfederschmidt/pyliwc#egg=pyliwc
liwc = LIWC("./LIWC2015.dic")liwc = LIWC("./LIWC2015.dic")
liwc_scores = liwc.process_text("My text to be analyzed")import pandas as pd
liwc = LIWC("./LIWC2015.dic")
df = pd.DataFrame([["all nice people", "mean people"], ["terrible", "awesome"]], columns=list('AB'), index=[12,3])
liwc.process_df(df)
# OR (in case of large series)
liwc.process_df_mp(df)