我在运行test_pro.py 时，报错，缺少文件

[songsh]

No such file or directory: './output/best_ner.bin', 从哪找

[guangze233]

No such file or directory: './output/best_ner.bin', 从哪找

还目录是程序新建的。运行NER_main.py，会训练NER模型并保存在新建的/output目录下

[997261095]

生成的，生成的模型很大啊300多M，而且是两个。我知道训练的时候有一个错误 key error 我估计是依赖的版本不同，把以前的key改了，这其实是一个小问题，自己把key打印出来看看改成什么了，就可以了

…

------------------&nbsp;原始邮件&nbsp;------------------ 发件人:&nbsp;"guangzezhu"<notifications@github.com&gt;; 发送时间:&nbsp;2020年5月10日(星期天) 下午2:29 收件人:&nbsp;"997261095/bert-kbqa"<bert-kbqa@noreply.github.com&gt;; 抄送:&nbsp;"Subscribed"<subscribed@noreply.github.com&gt;; 主题:&nbsp;Re: [997261095/bert-kbqa] 我在运行test_pro.py 时，报错，缺少文件 (#3) No such file or directory: './output/best_ner.bin', 从哪找 还目录是程序新建的。运行NER_main.py，会训练NER模型并保存在新建的/output目录下 — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[songsh]

感谢，只是跑这个best_ner.bin 花了很长时间，best_sim.bin 跑了天都没跑完

[songsh]

{'0': {'precision': 0.972353061642433, 'recall': 0.8919572619466949, 'f1-score': 0.9304216830579677, 'support': 425
85}, '1': {'precision': 0.7982297324481995, 'recall': 0.8659973810563073, 'f1-score': 0.8307338009002407, 'support'
: 4582}, '2': {'precision': 0.8254122574131548, 'recall': 0.934903223073991, 'f1-score': 0.8767525916511101, 'suppo
rt': 23611}, 'accuracy': 0.9046031252649128, 'macro avg': {'precision': 0.8653316838345958, 'recall': 0.89761928869
23311, 'f1-score': 0.8793026918697727, 'support': 70778}, 'weighted avg': {'precision': 0.9120624003772734, 'recall
': 0.9046031252649128, 'f1-score': 0.906064526974806, 'support': 70778}}

这是打印出来的ret 值， 是把micro avg 改成 weighted avg ?

[Annielina]

{'0': {'precision': 0.972353061642433, 'recall': 0.8919572619466949, 'f1-score': 0.9304216830579677, 'support': 425
85}, '1': {'precision': 0.7982297324481995, 'recall': 0.8659973810563073, 'f1-score': 0.8307338009002407, 'support'
: 4582}, '2': {'precision': 0.8254122574131548, 'recall': 0.934903223073991, 'f1-score': 0.8767525916511101, 'suppo
rt': 23611}, 'accuracy': 0.9046031252649128, 'macro avg': {'precision': 0.8653316838345958, 'recall': 0.89761928869
23311, 'f1-score': 0.8793026918697727, 'support': 70778}, 'weighted avg': {'precision': 0.9120624003772734, 'recall
': 0.9046031252649128, 'f1-score': 0.906064526974806, 'support': 70778}}

这是打印出来的ret 值， 是把micro avg 改成 weighted avg ?

您好，您训练完了吗，是不是把micro avg 改成 weighted avg

[wuhq1]

No such file or directory: './output/best_ner.bin', 从哪找

还目录是程序新建的。运行NER_main.py，会训练NER模型并保存在新建的/output目录下

error: the following arguments are required: --data_dir, --vob_file, 运行NER_main.py文件出现这样的错误。怎么处理？

[wangjitoan]

兄弟姐妹们，有训练出来的 可以发给我吗