This package automatically caches the results of all queries made to external resources. This includes the results of:
- SPARQL queries (via the
endpoint_wrapper() <kgextension.sparql_helper.endpoint_wrapper()>
method). - URI querying (via the
uri_querier() <kgextension.uri_helper.uri_querier()>
method). - DBpedia Lookup Linker requests (via the
dll_query_resolver() <kgextension.linking_helper.dll_query_resolver()>
method). - DBpedia Spotlight requests (via the
spotlight_uri_extractor() <kgextension.linking_helper.spotlight_uri_extractor()>
method). - Checks for URL availability (via the
url_exists() < kgextension.utilities_helper.url_exists()>
method).´
Note
As of now the size of the cache is and can not be limited. If you run into issues with regards of memory usage, try to clear the cache <clearing_cache>
.
Caching can be manually disabling on a "per call" level by setting the caching
parameter of a cached method to False, as shown in the example below.
df_linked = label_linker(df, column='author', caching=False)
At the moment it is not possible to easily disable caching for all methods at once.
To check the status of the cache, call the show_cache_info() <kgextension.caching_helper.show_cache_info()>
method, as shown in the example below.
from kgextension.caching_helper import show_cache_info
show_cache_info()
To clear the cache, call the clear_cache() <kgextension.caching_helper.clear_cache()>
method, as shown in the example below.
from kgextension.caching_helper import clear_cache
clear_cache()