We(nlpp) ranked the third in the public leaderboard.
This is the document that describes the code and its usage for Ebay NER problem.
Major files are shown below.
-
baseline.py: the main function to train the model. -
baseline_gen.py: the main function to generate labels for unlabled data based on the trained model. -
data.py: the designed data loaders for training and testing phrases. -
model.py: the designed model structure. -
baseline_sub.py: the main function to test the model with quiz data by a single trained model. -
baseline_sub_multi.py: the main function to test the model with quiz data by multiple trained models and their performance as weights.