<!-- 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-naamapdam-fine-tuned
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: 0.3466
- Precision: 0.7858
- Recall: 0.8156
- F1: 0.8004
- Accuracy: 0.9134
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 |
---|---|---|---|---|---|---|---|
0.37 | 0.51 | 1000 | 0.3241 | 0.7491 | 0.7745 | 0.7616 | 0.8951 |
0.3118 | 1.03 | 2000 | 0.3120 | 0.7631 | 0.7862 | 0.7745 | 0.8998 |
0.2877 | 1.54 | 3000 | 0.2994 | 0.7679 | 0.7918 | 0.7797 | 0.9027 |
0.284 | 2.06 | 4000 | 0.2977 | 0.7643 | 0.7960 | 0.7798 | 0.9026 |
0.2631 | 2.57 | 5000 | 0.2900 | 0.7716 | 0.8001 | 0.7856 | 0.9040 |
0.259 | 3.08 | 6000 | 0.2958 | 0.7739 | 0.8058 | 0.7895 | 0.9050 |
0.2426 | 3.6 | 7000 | 0.2850 | 0.7795 | 0.8042 | 0.7917 | 0.9072 |
0.2378 | 4.11 | 8000 | 0.2910 | 0.7740 | 0.8069 | 0.7901 | 0.9059 |
0.2231 | 4.63 | 9000 | 0.2878 | 0.7788 | 0.8106 | 0.7944 | 0.9077 |
0.2188 | 5.14 | 10000 | 0.2941 | 0.7752 | 0.8101 | 0.7923 | 0.9087 |
0.2066 | 5.66 | 11000 | 0.2928 | 0.7721 | 0.8147 | 0.7928 | 0.9079 |
0.2008 | 6.17 | 12000 | 0.3048 | 0.7798 | 0.8141 | 0.7966 | 0.9088 |
0.1902 | 6.68 | 13000 | 0.2987 | 0.7834 | 0.8108 | 0.7969 | 0.9099 |
0.1843 | 7.2 | 14000 | 0.3055 | 0.7784 | 0.8137 | 0.7957 | 0.9101 |
0.1775 | 7.71 | 15000 | 0.2991 | 0.7762 | 0.8155 | 0.7953 | 0.9096 |
0.1694 | 8.23 | 16000 | 0.3117 | 0.7876 | 0.8137 | 0.8004 | 0.9120 |
0.1631 | 8.74 | 17000 | 0.3085 | 0.7761 | 0.8210 | 0.7979 | 0.9121 |
0.1585 | 9.25 | 18000 | 0.3144 | 0.7851 | 0.8063 | 0.7955 | 0.9108 |
0.1528 | 9.77 | 19000 | 0.3086 | 0.7834 | 0.8169 | 0.7998 | 0.9124 |
0.1458 | 10.28 | 20000 | 0.3167 | 0.7773 | 0.8208 | 0.7985 | 0.9114 |
0.143 | 10.8 | 21000 | 0.3202 | 0.7822 | 0.8134 | 0.7975 | 0.9123 |
0.1368 | 11.31 | 22000 | 0.3299 | 0.7798 | 0.8176 | 0.7983 | 0.9112 |
0.1337 | 11.83 | 23000 | 0.3369 | 0.7857 | 0.8151 | 0.8002 | 0.9131 |
0.1289 | 12.34 | 24000 | 0.3366 | 0.7855 | 0.8148 | 0.7999 | 0.9128 |
0.1257 | 12.85 | 25000 | 0.3316 | 0.7837 | 0.8172 | 0.8001 | 0.9129 |
0.122 | 13.37 | 26000 | 0.3415 | 0.7880 | 0.8120 | 0.7998 | 0.9136 |
0.1191 | 13.88 | 27000 | 0.3414 | 0.7867 | 0.8182 | 0.8021 | 0.9139 |
0.1157 | 14.4 | 28000 | 0.3481 | 0.7863 | 0.8174 | 0.8016 | 0.9136 |
0.1153 | 14.91 | 29000 | 0.3466 | 0.7858 | 0.8156 | 0.8004 | 0.9134 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3