``` 涉政 spend time: 157.54228496551514 Predict 0 1 2 3 5 All label 1 22 1350 42 242 80 1736 All 22 1350 42 242 80 1736 Accuracy: 0.7776497695852534 Error amount: 386 / 1736 dataset amount after removing official messages: 5287 dataset amount after removing duplicate messages: 299 C:\Python38\lib\site-packages\transformers\configuration_xlnet.py:205: FutureWarning: This config doesn't use attention memories, a core feature of XLNet. Consider setting `men_len` to a non-zero value, for example `xlnet = XLNetLMHeadModel.from_pretrained('xlnet-base-cased'', mem_len=1024)`, for accurate training performance as well as an order of magnitude faster inference. Starting from version 3.5.0, the default parameter will be 1024, following the implementation in https://arxiv.org/abs/1906.08237 warnings.warn( 辱罵 spend time: 28.961681127548218 Predict 0 1 2 3 4 5 All label 2 10 3 254 1 9 22 299 All 10 3 254 1 9 22 299 Accuracy: 0.8494983277591973 Error amount: 45 / 299 dataset amount after removing official messages: 1183 dataset amount after removing duplicate messages: 1181 C:\Python38\lib\site-packages\transformers\configuration_xlnet.py:205: FutureWarning: This config doesn't use attention memories, a core feature of XLNet. Consider setting `men_len` to a non-zero value, for example `xlnet = XLNetLMHeadModel.from_pretrained('xlnet-base-cased'', mem_len=1024)`, for accurate training performance as well as an order of magnitude faster inference. Starting from version 3.5.0, the default parameter will be 1024, following the implementation in https://arxiv.org/abs/1906.08237 warnings.warn( 違禁 spend time: 111.92304039001465 Predict 0 1 2 3 4 5 All label 3 6 10 30 1052 2 81 1181 All 6 10 30 1052 2 81 1181 Accuracy: 0.890770533446232 Error amount: 129 / 1181 dataset amount after removing official messages: 415 dataset amount after removing duplicate messages: 415 C:\Python38\lib\site-packages\transformers\configuration_xlnet.py:205: FutureWarning: This config doesn't use attention memories, a core feature of XLNet. Consider setting `men_len` to a non-zero value, for example `xlnet = XLNetLMHeadModel.from_pretrained('xlnet-base-cased'', mem_len=1024)`, for accurate training performance as well as an order of magnitude faster inference. Starting from version 3.5.0, the default parameter will be 1024, following the implementation in https://arxiv.org/abs/1906.08237 warnings.warn( 色情 spend time: 46.245901346206665 Predict 0 2 3 4 5 All label 4 6 27 22 336 24 415 All 6 27 22 336 24 415 Accuracy: 0.8096385542168675 Error amount: 79 / 415 dataset amount after removing official messages: 10000 dataset amount after removing duplicate messages: 5786 C:\Python38\lib\site-packages\transformers\configuration_xlnet.py:205: FutureWarning: This config doesn't use attention memories, a core feature of XLNet. Consider setting `men_len` to a non-zero value, for example `xlnet = XLNetLMHeadModel.from_pretrained('xlnet-base-cased'', mem_len=1024)`, for accurate training performance as well as an order of magnitude faster inference. Starting from version 3.5.0, the default parameter will be 1024, following the implementation in https://arxiv.org/abs/1906.08237 warnings.warn( 廣告 spend time: 16.618953704833984 Predict 0 2 4 5 All label 5 23 3 1 5759 5786 All 23 3 1 5759 5786 Accuracy: 0.9953335637746284 Error amount: 27 / 5786 ``` 侯勇 ¹⁵¹⁵⁴⁹⁶⁹⁵⁸⁸ 任益波新号码：180-5142-8688 小ゝ博℡ 纯手工生姜洗护（招代理） 润帆🚗行¹⁸³⁸⁷⁵⁵⁴²²¹ 润帆🚗行¹³⁶²⁸⁸⁷⁷⁶⁶⁹ 郭超℡¹⁸²²⁹⁸²⁰⁵⁰⁰ A隋光耀℡₁₅₅₂₆₆₀₅₃₆₃ 张彦辉¹³³³³⁷³³⁶⁵⁶ 185 2102 1537 西西包包饰品号（招代理） 绽放传媒-招主播（电🚬） 李磊¹³⁷⁷⁴⁴¹²³⁴⁵ 顺丰钟翔℡¹⁵⁷²⁷⁷⁰³¹⁹⁹ YZ₁₈₆₃₇₆₇₇₃₆₇ 毛海一℡¹⁶⁶⁰⁰²⁷²⁶⁶⁶ 胡阳℡¹⁸⁸⁷⁰²⁶⁸³⁴³ A5.🇨🇳 发爵造型公众号 王建一℡¹⁸⁶⁴⁶⁰⁰⁹³⁹⁸ ℡孟₁₈₀₅₄₅₄₅₅₈₈ １３４１４６２８９９９ A谢老师℡¹⁸⁵⁴³⁴²⁷⁷⁷² ``` ```