BeerTai
commented
5 years ago 找到了，top_k设置过大，原先代码设置是100，我的是96
1760083838
邮箱：1760083838@qq.com
Signature is customized by Netease Mail MasterOn 10/30/2019 00:13, lkluo wrote: @SJIANAN 找到解决方法了吗？
—You are receiving this because you were mentioned.Reply to this email directly, view it on GitHub, or unsubscribe. [ { "@context": "http://schema.org", "@type": "EmailMessage", "potentialAction": { "@type": "ViewAction", "target": "https://github.com/Tencent/NeuralNLP-NeuralClassifier/issues/31?email_source=notifications\u0026email_token=AHC6LY6FBMJA3QTNRBK3DXLQRBOLHA5CNFSM4JFLNBWKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOECRDZSA#issuecomment-547503304", "url": "https://github.com/Tencent/NeuralNLP-NeuralClassifier/issues/31?email_source=notifications\u0026email_token=AHC6LY6FBMJA3QTNRBK3DXLQRBOLHA5CNFSM4JFLNBWKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOECRDZSA#issuecomment-547503304", "name": "View Issue" }, "description": "View this Issue on GitHub", "publisher": { "@type": "Organization", "name": "GitHub", "url": "https://github.com" } } ]
你好，我在自己的数据上跑时，总是显示题目中的错误： File "/home/jsun/NeuralNLP-NeuralClassifier-master/evaluate/classification_evaluate.py", line 270, in evaluate if prob_np[predict_label_idx[j]] > threshold: 然后这是我的数据格式： {"doc_label": ["label_06"], "doc_token": ["右", "侧", "甲", "状", "腺", "叶", "切", "除", "术"], "doc_keyword": [], "doc_topic": []} 在rcv1中跑却没错。