Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
The split-by-leaf mode (grow_policy="lossguide") is not supported in distributed training,
which makes XGBoost4J on Spark much slower than LightGBM on Spark.
https://towardsdatascience.com/build-xgboost-lightgbm-models-on-large-datasets-what-are-the-possible-solutions-bf882da2c27d
https://xgboost …
Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
SentencePiece is an unsupervised text tokenizer for Neural Network-based text generation.
Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
Classic word representation cannot handle unseen word or rare word well. Character embeddings is one of the solution to overcome out-of-vocabulary (OOV). However, it may be too fine-grained and miss some …
Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
Word Embedding Character Embedding Subword Embeddling Tokenization
General Language Understanding Evaluation (GLUE)
Natural Language Generation (NLG) Natural Language Generation, as defined by Artificial Intelligence: Natural Language Processing Fundamentals, is the “process …
Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
Module.__call__ method register all hooks and call the method Module.forward.
In short,
when you train the model you should use the method forward,
while when you test the …