Felix Hamborg

training with custom loss function, or assigning weights to classes

1

Hi! Is it possible to define a custom loss function when training spacy's transformers? Specifically, I'm asking because the classes in my dataset are imbalanced, and I'd like to weight...

filter articles for keywords

7

* Add a new entry to the configuration file * Add a new module in scrapy that discards articles that do not contain values from the newly added config entry

Merge articles spread on multiple pages

2

Example: http://www.zeit.de/2016/18/ttip-barack-obama-hannover-usa-widerstand Under the given URL only the first part of the article is shown. A (human) reader can either click on a link that points to the second page...

doc should either be added to readme.md or maybe we start another page for all functionality that is not directly related to the core functionality - gdeltcrawler - pandas exporter

move nominatim to preprocessing

2

we initially had the nominatim queries directly in the environment extractor, to spare some requests. however, as shown in https://github.com/bkrrr/Giveme5W/blob/master/extractor/extractors/environment_extractor.py we query each phrase that is a `LOCATION` so that...

https://www.ltg.ed.ac.uk/software/geoparser/

Things to improve

1

- semantic similarity in learnweights - tree access is not speed optimized

different LCF-BERT implementations

3

The implementation of LCF-BERT in this repository is different to both the description of LCF-BERT in the paper (https://www.mdpi.com/2076-3417/9/16/3389/htm) as well as different to the original implementation of LCF-BERT by...