hello,After running your code, the f value is only 0.4,sorry,Is there anything I should pay attention to when running

[Biubiulity]

please tell me why?

[jiesutd]

Please give more details, then can I give you advice. And please do not repeatly close-reopen issue, it will cause spam in email system.

[Biubiulity]

Please give more details, then can I give you advice. And please do not repeatly close-reopen issue, it will cause spam in email system.

I'm really sorry.Because this is the first time I used Github's comment,so I don't know use it others can see it.Please forgive me. And I want to know:

1. I want to know whether the data set in data in your code is the data set of weibo. 2.Does your code run other models as well, for example word_baseline ?If yes ,Could you tell me how to modify it？ 3.when I run the dataset in your code ,the F_score is 0.4177,so I want to know whether the data set I am running is the data set of weibo. The log file is: D:\Anaconda3\python.exe G:/experiment/main.py CuDNN: True GPU available: False Status: train Seg: True Train file: ./data/demo.train.char Dev file: ./data/demo.dev.char Test file: ./data/demo.test.char Raw file: ../data/onto4ner.cn/demo.test.char Char emb: data/gigaword_chn.all.a2b.uni.ite50.vec Bichar emb: None Gaz file: data/ctb.50d.vec Model saved to: data/demo Load gaz file: data/ctb.50d.vec total size: 704368 gaz alphabet size: 10130 gaz alphabet size: 10966 gaz alphabet size: 13652 build word pretrain emb... Embedding: pretrain word:11327, prefect match:2551, case_match:0, oov:26, oov%:0.01008533747090768 build biword pretrain emb... Embedding: pretrain word:0, prefect match:0, case_match:0, oov:31748, oov%:0.9999685029449746 build gaz pretrain emb... Embedding: pretrain word:704368, prefect match:13650, case_match:0, oov:1, oov%:7.32493407559332e-05 Training model... DATA SUMMARY START: Tag scheme: BMES MAX SENTENCE LENGTH: 250 MAX WORD LENGTH: -1 Number normalized: True Use bigram: False Word alphabet size: 2578 Biword alphabet size: 31749 Char alphabet size: 2578 Gaz alphabet size: 13652 Label alphabet size: 18 Word embedding size: 50 Biword embedding size: 50 Char embedding size: 30 Gaz embedding size: 50 Norm word emb: True Norm biword emb: True Norm gaz emb: False Norm gaz dropout: 0.5 Train instance number: 1147 Dev instance number: 113 Test instance number: 316 Raw instance number: 0 Hyperpara iteration: 100 Hyperpara batch size: 1 Hyperpara lr: 0.015 Hyperpara lr_decay: 0.05 Hyperpara HP_clip: 5.0 Hyperpara momentum: 0 Hyperpara hidden_dim: 200 Hyperpara dropout: 0.5 Hyperpara lstm_layer: 1 Hyperpara bilstm: True Hyperpara GPU: False Hyperpara use_gaz: True Hyperpara fix gaz emb: False Hyperpara use_char: False DATA SUMMARY END. Data setting saved to file: data/demo.dset build batched lstmcrf... build batched bilstm... build LatticeLSTM... forward , Fix emb: False gaz drop: 0.5 load pretrain word emb... (13652, 50) build LatticeLSTM... backward , Fix emb: False gaz drop: 0.5 load pretrain word emb... (13652, 50) build batched crf... finished built model. Epoch: 0/100 Learning rate is setted as: 0.015 Instance: 500; Time: 140.08s; loss: 9362.1726; acc: 18641/21743=0.8573 Instance: 1000; Time: 123.59s; loss: 5484.8606; acc: 37797/43593=0.8670 Instance: 1147; Time: 33.99s; loss: 1021.0708; acc: 43146/49471=0.8721 Epoch: 0 training finished. Time: 297.66s, speed: 3.85st/s, total loss: 15868.103970527649 gold_num = 215 pred_num = 102 right_num = 69 Dev: time: 3.70s, speed: 30.60st/s; acc: 0.9162, p: 0.6765, r: 0.3209, f: 0.4353 Exceed previous best f score: -1 gold_num = 340 pred_num = 109 right_num = 55 Test: time: 9.94s, speed: 31.95st/s; acc: 0.9317, p: 0.5046, r: 0.1618, f: 0.2450 Epoch: 1/100 Learning rate is setted as: 0.014249999999999999 Instance: 500; Time: 121.06s; loss: 3302.7600; acc: 19684/21682=0.9078 Instance: 1000; Time: 113.08s; loss: 3065.0727; acc: 39064/42920=0.9102 Instance: 1147; Time: 35.27s; loss: 912.5710; acc: 45036/49471=0.9104 Epoch: 1 training finished. Time: 269.42s, speed: 4.26st/s, total loss: 7280.403723716736 gold_num = 215 pred_num = 149 right_num = 105 Dev: time: 3.43s, speed: 33.00st/s; acc: 0.9267, p: 0.7047, r: 0.4884, f: 0.5769 Exceed previous best f score: 0.43533123028391163 gold_num = 340 pred_num = 186 right_num = 103 Test: time: 8.56s, speed: 37.14st/s; acc: 0.9410, p: 0.5538, r: 0.3029, f: 0.3916 ... ... Epoch: 99/100 Learning rate is setted as: 9.348204032106312e-05 Instance: 500; Time: 115.19s; loss: 281.3181; acc: 21674/21864=0.9913 Instance: 1000; Time: 109.61s; loss: 273.0064; acc: 42363/42772=0.9904 Instance: 1147; Time: 34.63s; loss: 64.7753; acc: 49009/49471=0.9907 Epoch: 99 training finished. Time: 259.43s, speed: 4.42st/s, total loss: 619.0996932983398 gold_num = 215 pred_num = 144 right_num = 117 Dev: time: 3.36s, speed: 33.64st/s; acc: 0.9382, p: 0.8125, r: 0.5442, f: 0.6518 gold_num = 340 pred_num = 201 right_num = 113 Test: time: 8.18s, speed: 38.68st/s; acc: 0.9433, p: 0.5622, r: 0.3324, f: 0.4177

Process finished with exit code 0 Since this is my first time to touch pytorch, read codes is difficult to me .So I don't know what'wrong with them . If you have time, please find time to answer my questions.Thank you very much !

[jiesutd]

Train file: ./data/demo.train.char Dev file: ./data/demo.dev.char Test file: ./data/demo.test.char

You used the wrong data, this is the demo data which is to test if you can run the code. Please us the correct data as the input.

[Biubiulity]

OK, I know ,Thank you for your patient answer