SubwordTokenizer# Constructor# SubwordTokenizer(hash_file[, do_lower_case]) Run CUDA BERT subword tokenizer on cuDF strings column.