Closed evanchenhi closed 8 months ago
我看pycorrector/macbert/train_macbert4csc.yml配置文件里BERT_CKPT是原始 hfl/chinese-macbert-base 模型
微调完保存的也是原始模型,链接如下,并没有保存MacBert4Csc模型结构,只保存了权重参数 bert = BertForMaskedLM.from_pretrained(cfg.MODEL.BERT_CKPT) bert.save_pretrained(cfg.OUTPUT_DIR) https://github.com/shibing624/pycorrector/blob/master/pycorrector/macbert/train.py#L114
bert = BertForMaskedLM.from_pretrained(cfg.MODEL.BERT_CKPT)
bert.save_pretrained(cfg.OUTPUT_DIR)
不知道我看的是否正确
嗯
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.(由于长期不活动,机器人自动关闭此问题,如果需要欢迎提问)
我看pycorrector/macbert/train_macbert4csc.yml配置文件里BERT_CKPT是原始 hfl/chinese-macbert-base 模型
微调完保存的也是原始模型,链接如下,并没有保存MacBert4Csc模型结构,只保存了权重参数
bert = BertForMaskedLM.from_pretrained(cfg.MODEL.BERT_CKPT)
bert.save_pretrained(cfg.OUTPUT_DIR)
https://github.com/shibing624/pycorrector/blob/master/pycorrector/macbert/train.py#L114不知道我看的是否正确