####command + pythonrun.py-data_dir=data-vocab_file=pretrain/chinese/vocab.txt-bert_config_file=pretrain/chinese/bert_config.json-init_checkpoint=pretrain/chinese/bert_model.ckpt-max_seq_length=128-learning_rate=2e-5-num_train_epochs=3.0-output_dir=output/

bert-base-ner-train\ -data_dir=/media/filesNew/szq/BERT-BiLSTM-CRF-NER-master/data\ -output_dir=/media/filesNew/szq/BERT-BiLSTM-CRF-NER-master/output/\ -init_checkpoint=/media/filesNew/szq/BERT-BiLSTM-CRF-NER-master/pretrain/chinese/bert_model.ckpt\ -bert_config_file=/media/filesNew/szq/BERT-BiLSTM-CRF-NER-master/pretrain/chinese/bert_config.json\ -vocab_file=/media/filesNew/szq/BERT-BiLSTM-CRF-NER-master/pretrain/chinese/vocab.txt -device_map=[2,3]

bert-base-ner-train[-h][-data_dirDATA_DIR] [-bert_config_fileBERT_CONFIG_FILE] [-output_dirOUTPUT_DIR] [-init_checkpointINIT_CHECKPOINT] [-vocab_fileVOCAB_FILE] [-max_seq_lengthMAX_SEQ_LENGTH] [-do_trainDO_TRAIN][-do_evalDO_EVAL] [-do_predictDO_PREDICT][-batch_sizeBATCH_SIZE] [-learning_rateLEARNING_RATE] [-num_train_epochsNUM_TRAIN_EPOCHS] [-dropout_rateDROPOUT_RATE][-clipCLIP] [-warmup_proportionWARMUP_PROPORTION] [-lstm_sizeLSTM_SIZE][-num_layersNUM_LAYERS] [-cellCELL] [-save_checkpoints_stepsSAVE_CHECKPOINTS_STEPS] [-save_summary_stepsSAVE_SUMMARY_STEPS] [-filter_adam_varFILTER_ADAM_VAR] [-do_lower_caseDO_LOWER_CASE][-cleanCLEAN] [-device_mapDEVICE_MAP][-label_listLABEL_LIST] [-verbose][-nerNER][-version]