en-multinerd-masked-ner-more-training
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0371
- Precision: 0.9386
- Recall: 0.9456
- F1: 0.9421
- Accuracy: 0.9920
- Per-precision: 0.9960
- Per-recall: 0.9961
- Per-f1: 0.9961
- Org-precision: 0.9364
- Org-recall: 0.9419
- Org-f1: 0.9392
- Loc-precision: 0.9719
- Loc-recall: 0.9752
- Loc-f1: 0.9735
- Dis-precision: 0.7059
- Dis-recall: 0.7363
- Dis-f1: 0.7208
- Anim-precision: 0.7033
- Anim-recall: 0.7275
- Anim-f1: 0.7152
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | Per-precision | Per-recall | Per-f1 | Org-precision | Org-recall | Org-f1 | Loc-precision | Loc-recall | Loc-f1 | Dis-precision | Dis-recall | Dis-f1 | Anim-precision | Anim-recall | Anim-f1 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.0243 | 1.0 | 8205 | 0.0271 | 0.9184 | 0.9458 | 0.9319 | 0.9906 | 0.9927 | 0.9963 | 0.9945 | 0.8838 | 0.9499 | 0.9157 | 0.9586 | 0.9707 | 0.9646 | 0.6719 | 0.7274 | 0.6986 | 0.6659 | 0.7557 | 0.7080 |
0.0159 | 2.0 | 16410 | 0.0263 | 0.9324 | 0.9448 | 0.9386 | 0.9917 | 0.9913 | 0.9965 | 0.9939 | 0.9237 | 0.9507 | 0.9370 | 0.9613 | 0.9739 | 0.9675 | 0.7128 | 0.7233 | 0.7180 | 0.6990 | 0.7187 | 0.7087 |
0.0101 | 3.0 | 24615 | 0.0295 | 0.9396 | 0.9426 | 0.9411 | 0.9919 | 0.9935 | 0.9963 | 0.9949 | 0.9352 | 0.9386 | 0.9369 | 0.9704 | 0.9732 | 0.9718 | 0.7168 | 0.7286 | 0.7226 | 0.7143 | 0.7055 | 0.7098 |
0.0057 | 4.0 | 32820 | 0.0335 | 0.9375 | 0.9465 | 0.9420 | 0.9919 | 0.9967 | 0.9955 | 0.9961 | 0.9341 | 0.9470 | 0.9405 | 0.9694 | 0.9739 | 0.9717 | 0.7065 | 0.7499 | 0.7275 | 0.7052 | 0.7257 | 0.7153 |
0.004 | 5.0 | 41025 | 0.0371 | 0.9386 | 0.9456 | 0.9421 | 0.9920 | 0.9960 | 0.9961 | 0.9961 | 0.9364 | 0.9419 | 0.9392 | 0.9719 | 0.9752 | 0.9735 | 0.7059 | 0.7363 | 0.7208 | 0.7033 | 0.7275 | 0.7152 |
Framework versions
- Transformers 4.36.1
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0
- Downloads last month
- 0