The cleanNLP package does not supply the raw java files provided by the Stanford NLP Group as they are quite large. This function downloads the libraries automatically, by default into the directory where the package was installed. These are not required for using the spaCy Python implementation.
download_core_nlp(type = c("default", "base", "en", "fr", "de", "es"),
output_loc, url = NULL, url_core = TRUE)type of files to download. The base backage is always required. Other jars include model files for French, German, and Spanish. These can be installed in addition to the base package. By default, the function downloads the base package and English model files.
a string showing where the files are to be downloaded. If missing, will try to download files into the directory where the package was original installed.
the url to try to download components from. Setting to NULL
uses the default location on the Stanford NLP server, but
you can set this manually by using this option. It also allows
for local files, but note that the path must include the prefix
file://. For details, see download.file.
if url is not null, this flag indicates whehter the path
given to url points to the core nlp files (which are zipped) or
to model files (which are unzipped).
# NOT RUN {
download_core_nlp()
download_core_nlp(type="spanish")
# }
Run the code above in your browser using DataLab