petermr / docanalysis

Semantic analysis of text documents including sentence and paragraph splitting
Apache License 2.0
12 stars 3 forks source link

relation between `pygetpapers` output and `docanalysis` project and redundant args #11

Open petermr opened 2 years ago

petermr commented 2 years ago

It would be nice, but maybe impossible) if argument names are reasonably consistent between pygetpapers, docanalysis and py4ami. (This is a problem in UNIX tools where there is little consistency

-q is passed to pygetpapers with the same meaning as standalone pygetpapers. I assumed (wrongly) that -o was the same , but it has a different role (output CSV).

Some parameters imply or depend on others:

docanalysis 
[...]
optional arguments:
  -h, --help            show this help message and exit
  --run_pygetpapers     queries EuropePMC via pygetpapers
  --run_sectioning      make sections
  -q QUERY, --query QUERY
                        query to pygetpapers
  --project_name PROJECT_NAME
                        name of CProject folder
  -d DICTIONARY, --dictionary DICTIONARY
                        Ami Dictionary to tag sentences and support supervised entity extraction
  -o OUTPUT, --output OUTPUT
                        Output CSV file [default=entities.csv]
  --make_ami_dict MAKE_AMI_DICT
                        if provided will make ami dict with given title
  -l LOGLEVEL, --loglevel LOGLEVEL
                        [All] Provide logging level. Example --log warning
                        <<info,warning,debug,error,critical>>, default='info'
  -f LOGFILE, --logfile LOGFILE
                        [All] save log to specified file in output directory as well as printing to
                        terminal
  --section [SECTION [SECTION ...]]
                        Which section to get
  --entities [ENTITIES [ENTITIES ...]]
                        Which entities to get. Default(ALL)
  --spacy_model SPACY_MODEL
                        Optional. (spacy, scispacy). Default(spacy)
  --html HTML           Saves output in html format to given path