<!-- 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-500-V2
This model is a fine-tuned version of livinNector/tabert-500 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4057
- Precision: 0.7870
- Recall: 0.8040
- F1: 0.7954
- Accuracy: 0.9056
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 |
---|---|---|---|---|---|---|---|
0.3751 | 0.49 | 1000 | 0.3876 | 0.7294 | 0.7381 | 0.7337 | 0.8758 |
0.3211 | 0.99 | 2000 | 0.3530 | 0.7603 | 0.7427 | 0.7514 | 0.8851 |
0.2932 | 1.48 | 3000 | 0.3443 | 0.7501 | 0.7757 | 0.7627 | 0.8882 |
0.2884 | 1.98 | 4000 | 0.3404 | 0.7553 | 0.7878 | 0.7712 | 0.8907 |
0.268 | 2.47 | 5000 | 0.3241 | 0.7705 | 0.7888 | 0.7795 | 0.8959 |
0.2638 | 2.96 | 6000 | 0.3246 | 0.7823 | 0.7850 | 0.7836 | 0.8954 |
0.246 | 3.46 | 7000 | 0.3175 | 0.7769 | 0.7989 | 0.7878 | 0.8999 |
0.2457 | 3.95 | 8000 | 0.3216 | 0.7732 | 0.7934 | 0.7832 | 0.8999 |
0.2253 | 4.44 | 9000 | 0.3180 | 0.7792 | 0.7983 | 0.7887 | 0.8995 |
0.2271 | 4.94 | 10000 | 0.3250 | 0.7868 | 0.7895 | 0.7882 | 0.8996 |
0.2085 | 5.43 | 11000 | 0.3435 | 0.7838 | 0.7967 | 0.7902 | 0.8995 |
0.2091 | 5.93 | 12000 | 0.3300 | 0.7855 | 0.7958 | 0.7906 | 0.9009 |
0.1927 | 6.42 | 13000 | 0.3272 | 0.7771 | 0.7983 | 0.7876 | 0.9017 |
0.1932 | 6.91 | 14000 | 0.3310 | 0.7836 | 0.8060 | 0.7946 | 0.9047 |
0.1777 | 7.41 | 15000 | 0.3377 | 0.7882 | 0.8045 | 0.7963 | 0.9052 |
0.1785 | 7.9 | 16000 | 0.3406 | 0.7812 | 0.8042 | 0.7925 | 0.9036 |
0.1658 | 8.4 | 17000 | 0.3528 | 0.7892 | 0.7992 | 0.7942 | 0.9043 |
0.1651 | 8.89 | 18000 | 0.3419 | 0.7914 | 0.8072 | 0.7992 | 0.9068 |
0.1549 | 9.38 | 19000 | 0.3600 | 0.7931 | 0.7964 | 0.7948 | 0.9045 |
0.1539 | 9.88 | 20000 | 0.3525 | 0.7851 | 0.8091 | 0.7970 | 0.9052 |
0.1449 | 10.37 | 21000 | 0.3634 | 0.7881 | 0.7998 | 0.7939 | 0.9046 |
0.1436 | 10.86 | 22000 | 0.3736 | 0.7916 | 0.8058 | 0.7986 | 0.9069 |
0.1368 | 11.36 | 23000 | 0.3771 | 0.7892 | 0.8020 | 0.7955 | 0.9053 |
0.1347 | 11.85 | 24000 | 0.3800 | 0.7861 | 0.8060 | 0.7959 | 0.9045 |
0.1281 | 12.35 | 25000 | 0.3911 | 0.7852 | 0.8055 | 0.7952 | 0.9059 |
0.1272 | 12.84 | 26000 | 0.3919 | 0.7880 | 0.8005 | 0.7942 | 0.9052 |
0.1217 | 13.33 | 27000 | 0.4021 | 0.7887 | 0.7981 | 0.7934 | 0.9050 |
0.1202 | 13.83 | 28000 | 0.3959 | 0.7845 | 0.8057 | 0.7950 | 0.9056 |
0.1175 | 14.32 | 29000 | 0.4066 | 0.7864 | 0.8031 | 0.7947 | 0.9052 |
0.115 | 14.81 | 30000 | 0.4057 | 0.7870 | 0.8040 | 0.7954 | 0.9056 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3