input_processor.json

{"py/object": "input_processor.InputProcessor", "py/state": {"model_cls": "deberta", "bert_model": "microsoft/deberta-base", "word_embeddings_file": "./fcm/wordEmbeddings/glove.6B.300d.txt", "word_embeddings_format": "text", "train_files": ["./training/train.bucket23.txt", "./training/train.bucket17.txt", "./training/train.bucket2.txt", "./training/train.bucket1.txt", "./training/train.bucket9.txt", "./training/train.bucket8.txt", "./training/train.bucket10.txt", "./training/train.bucket24.txt", "./training/train.bucket12.txt", "./training/train.bucket22.txt", "./training/train.bucket20.txt", "./training/train.bucket21.txt", "./training/train.bucket0.txt", "./training/train.bucket18.txt", "./training/train.bucket15.txt", "./training/train.bucket5.txt", "./training/train.bucket19.txt", "./training/train.bucket11.txt", "./training/train.bucket7.txt", "./training/train.bucket13.txt", "./training/train.bucket6.txt", "./training/train.bucket16.txt", "./training/train.bucket4.txt", "./training/train.bucket14.txt", "./training/train.bucket3.txt"], "vocab_file": "./outputs/DeBERTa_fused/input_processor.json.vocab", "vector_size": 768, "ngram_threshold": 4, "nmin": 3, "nmax": 5, "ngram_dropout": 0.1, "min_word_count": 100, "max_copies": 5, "smin": 4, "smax": 32, "max_seq_length": 96, "mode": null, "form_only": false, "sep_symbol": ":", "train_file_idx": 0}}