Li XiaohonginGojek Product + TechIndoNLU — A benchmark for Bahasa Indonesia NLPIntroducing a collection of resources for training, evaluating, and analysing natural language understanding systems in Bahasa Indonesia.Nov 3, 20201Nov 3, 20201
Li XiaohonginGojek Product + TechHow Gojek Uses NLP to Name Pickup Locations at ScaleIntroducing CartoBERT, a Natural Language Processing (NLP) model developed by Gojek’s Cartography Data Science team.May 1, 20202May 1, 20202
Li XiaohonginTowards Data ScienceFine-tune Albert with Pre-training on Custom Corpus — Some notes and hiccups from real application…This is a follow up on my previous post, which illustrated detailed steps of Fine-tune Albert with Pre-training on Custom Corpus using a…Feb 9, 2020Feb 9, 2020
Li XiaohonginTowards Data ScienceFine tune Albert with Pre-training on Custom CorpusPre-train Albert with custom corpus for domain specific texts and further fine-tune the pre-trained model for application tasksDec 29, 20192Dec 29, 20192