BERT / run_ner 运行报错

[aeoling]

trafficstars

模型do_train阶段没有问题，但是运行do_eval会报错。 报错如下： CHECK failed: (byte_size_before_serialization) == (byte_size_after_serialization): tensorflow.GraphDef was modified concurrently during serialization. terminate called after throwing an instance of 'google::protobuf3::FatalException' what(): CHECK failed: (byte_size_before_serialization) == (byte_size_after_serialization): tensorflow.GraphDef was modified concurrently during serialization. 训练：13000样本 测试集：1100样本

[DukeEnglish]

你看里面 没有写 do eval的flag。发自我的iPhone------------------ 原始邮件 ------------------发件人: aeo.lin [email protected]发送时间: 2020年5月11日 16:33收件人: CLUEbenchmark/CLUE [email protected]抄送: Subscribed [email protected]主题: 回复：[CLUEbenchmark/CLUE] BERT / run_ner 运行报错 (#88) 模型do_train阶段没有问题，但是运行do_eval会报错。 报错如下： CHECK failed: (byte_size_before_serialization) == (byte_size_after_serialization): tensorflow.GraphDef was modified concurrently during serialization. terminate called after throwing an instance of 'google::protobuf3::FatalException' what(): CHECK failed: (byte_size_before_serialization) == (byte_size_after_serialization): tensorflow.GraphDef was modified concurrently during serialization. 训练：13000样本 测试集：1100样本

—You are receiving this because you are subscribed to this thread.Reply to this email directly, view it on GitHub, or unsubscribe. [ { "@context": "http://schema.org", "@type": "EmailMessage", "potentialAction": { "@type": "ViewAction", "target": "https://github.com/CLUEbenchmark/CLUE/issues/88", "url": "https://github.com/CLUEbenchmark/CLUE/issues/88", "name": "View Issue" }, "description": "View this Issue on GitHub", "publisher": { "@type": "Organization", "name": "GitHub", "url": "https://github.com" } } ]