The file "FB15k_Entity_Type_train.txt" contains 151818 entities, which is the same as the file with 10% noise "FB15k_Entity_Type_train10.txt". But the table.1 in the paper shows that the original train dataset is supposed to contain 136618 entities.
Is the "with 10% noise" dataset file uploaded as the original dataset file by mistake, or I misunderstand something?
The file "FB15k_Entity_Type_train.txt" contains 151818 entities, which is the same as the file with 10% noise "FB15k_Entity_Type_train10.txt". But the table.1 in the paper shows that the original train dataset is supposed to contain 136618 entities. Is the "with 10% noise" dataset file uploaded as the original dataset file by mistake, or I misunderstand something?