<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
taNER-1k-V2
This model is a fine-tuned version of livinNector/tabert-1k on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4253
- Precision: 0.7866
- Recall: 0.8029
- F1: 0.7947
- Accuracy: 0.9049
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 256
- eval_batch_size: 512
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.3724 | 0.49 | 1000 | 0.3865 | 0.7280 | 0.7372 | 0.7326 | 0.8758 |
0.3199 | 0.99 | 2000 | 0.3516 | 0.7524 | 0.7561 | 0.7543 | 0.8858 |
0.2911 | 1.48 | 3000 | 0.3436 | 0.7543 | 0.7765 | 0.7653 | 0.8906 |
0.2867 | 1.98 | 4000 | 0.3391 | 0.7522 | 0.7908 | 0.7710 | 0.8909 |
0.2654 | 2.47 | 5000 | 0.3262 | 0.7696 | 0.7845 | 0.7770 | 0.8961 |
0.2616 | 2.96 | 6000 | 0.3294 | 0.7784 | 0.7800 | 0.7792 | 0.8954 |
0.2422 | 3.46 | 7000 | 0.3191 | 0.7779 | 0.7934 | 0.7856 | 0.8999 |
0.2422 | 3.95 | 8000 | 0.3272 | 0.7735 | 0.7962 | 0.7847 | 0.8985 |
0.2208 | 4.44 | 9000 | 0.3252 | 0.7811 | 0.7952 | 0.7881 | 0.9012 |
0.2227 | 4.94 | 10000 | 0.3220 | 0.7789 | 0.7993 | 0.7890 | 0.9026 |
0.204 | 5.43 | 11000 | 0.3413 | 0.7904 | 0.7894 | 0.7899 | 0.9007 |
0.2036 | 5.93 | 12000 | 0.3329 | 0.7810 | 0.7984 | 0.7896 | 0.9009 |
0.1874 | 6.42 | 13000 | 0.3362 | 0.7872 | 0.7986 | 0.7929 | 0.9033 |
0.1877 | 6.91 | 14000 | 0.3414 | 0.7764 | 0.8029 | 0.7894 | 0.9013 |
0.172 | 7.41 | 15000 | 0.3463 | 0.7871 | 0.7997 | 0.7933 | 0.9032 |
0.1729 | 7.9 | 16000 | 0.3441 | 0.7863 | 0.8001 | 0.7931 | 0.9034 |
0.159 | 8.4 | 17000 | 0.3625 | 0.7856 | 0.7970 | 0.7912 | 0.9019 |
0.1585 | 8.89 | 18000 | 0.3575 | 0.7867 | 0.7980 | 0.7923 | 0.9030 |
0.1485 | 9.38 | 19000 | 0.3761 | 0.7850 | 0.7965 | 0.7907 | 0.9029 |
0.1468 | 9.88 | 20000 | 0.3658 | 0.7874 | 0.8019 | 0.7946 | 0.9037 |
0.1378 | 10.37 | 21000 | 0.3835 | 0.7851 | 0.8039 | 0.7944 | 0.9042 |
0.1364 | 10.86 | 22000 | 0.3852 | 0.7861 | 0.8019 | 0.7940 | 0.9043 |
0.1294 | 11.36 | 23000 | 0.3906 | 0.7854 | 0.7973 | 0.7913 | 0.9038 |
0.1277 | 11.85 | 24000 | 0.3947 | 0.7875 | 0.7988 | 0.7931 | 0.9030 |
0.1207 | 12.35 | 25000 | 0.4082 | 0.7841 | 0.7997 | 0.7918 | 0.9035 |
0.1199 | 12.84 | 26000 | 0.4137 | 0.7888 | 0.7993 | 0.7940 | 0.9049 |
0.1144 | 13.33 | 27000 | 0.4155 | 0.7875 | 0.7996 | 0.7935 | 0.9046 |
0.113 | 13.83 | 28000 | 0.4177 | 0.7840 | 0.8053 | 0.7945 | 0.9046 |
0.1103 | 14.32 | 29000 | 0.4280 | 0.7867 | 0.8021 | 0.7943 | 0.9042 |
0.1078 | 14.81 | 30000 | 0.4253 | 0.7866 | 0.8029 | 0.7947 | 0.9049 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3