CDial-GPT
CDial-GPT copied to clipboard
thu-coai
A Large-scale Chinese Short-Text Conversation Dataset and Chinese pre-training dialog models
您好，我在运行train()时报错 TypeError: unsupported operand type(s) for /: 'str' and 'int' (lm_loss), *_ = model(input_ids, labels=lm_labels, token_type_ids=token_type_ids) loss = lm_loss / args.gradient_accumulation_steps debug发现lm_loss 是str类型，请问是哪里出错了。 用的预训练thu-coai/CDial-GPT_LCCC-base模型，"--data_path"是"data/STC.json" 谢谢！
请问原文中用的GPT2-chitchat是直接使用的https://github.com/yangjianxin1/GPT2-chitchat 中提供的训练好的权重model_epoch40_50w吗？
INFO:transformers.tokenization_utils:Model name '/dfs/data/ckpt/CDial_GPT/CDial-GPT_LCCC-large' not found in model shortcut name list (bert-base-uncased, bert-large-uncased, bert-base-cased, bert-large-cased, bert-base-multilingual-uncased, bert-base-multilingual-cased, bert-base-chinese, bert-base-german-cased, bert-large-uncased-whole-word-masking, bert-large-cased-whole-word-masking, bert-large-uncased-whole-word-masking-finetuned-squad, bert-large-cased-whole-word-masking-finetuned-squad, bert-base-cased-finetuned-mrpc, bert-base-german-dbmdz-cased, bert-base-german-dbmdz-uncased). Assuming '/dfs/data/ckpt/CDial_GPT/CDial-GPT_LCCC-large' is a path...
使用“thu-coai/CDial-GPT_LCCC-base”提一些问题，如 “秦朝如何灭亡的？” “程序员需要具备哪些基本技能？” 回答分别是 “不 知 道 ， 我 也 不 知 道” “我 也 想 知 道” 这种有点像Blandness Problem的回复，有没有在特定知识或者场景下fine-tuned 的预训练模型？
