<!-- 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. -->
scideberta-cs-tdm-pretrained-finetuned-ner
This model is a fine-tuned version of sohamtiwari3120/scideberta-cs-tdm-pretrained on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.6836
- Overall Precision: 0.5912
- Overall Recall: 0.6850
- Overall F1: 0.6347
- Overall Accuracy: 0.9609
- Datasetname F1: 0.5882
- Hyperparametername F1: 0.6897
- Hyperparametervalue F1: 0.7619
- Methodname F1: 0.6525
- Metricname F1: 0.7500
- Metricvalue F1: 0.6452
- Taskname F1: 0.5370
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 100
Training results
| Training Loss | Epoch | Step | Validation Loss | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | Datasetname F1 | Hyperparametername F1 | Hyperparametervalue F1 | Methodname F1 | Metricname F1 | Metricvalue F1 | Taskname F1 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 132 | 0.3507 | 0.3972 | 0.6870 | 0.5034 | 0.9410 | 0.4370 | 0.5441 | 0.5814 | 0.6124 | 0.5604 | 0.6207 | 0.3724 |
| No log | 2.0 | 264 | 0.3079 | 0.4066 | 0.7520 | 0.5278 | 0.9430 | 0.4138 | 0.5380 | 0.6222 | 0.5895 | 0.625 | 0.7273 | 0.4340 |
| No log | 3.0 | 396 | 0.3740 | 0.5007 | 0.7195 | 0.5905 | 0.9535 | 0.4882 | 0.6777 | 0.7500 | 0.6254 | 0.6747 | 0.7097 | 0.4962 |
| 0.4014 | 4.0 | 528 | 0.4072 | 0.5161 | 0.7154 | 0.5997 | 0.9540 | 0.5167 | 0.6612 | 0.6374 | 0.6337 | 0.6753 | 0.6061 | 0.5341 |
| 0.4014 | 5.0 | 660 | 0.4088 | 0.5590 | 0.7317 | 0.6338 | 0.9582 | 0.5660 | 0.6667 | 0.7397 | 0.6250 | 0.7226 | 0.75 | 0.5794 |
| 0.4014 | 6.0 | 792 | 0.4810 | 0.5201 | 0.7093 | 0.6002 | 0.9550 | 0.4874 | 0.5970 | 0.6506 | 0.6207 | 0.6708 | 0.6250 | 0.5756 |
| 0.4014 | 7.0 | 924 | 0.5288 | 0.5403 | 0.6809 | 0.6025 | 0.9576 | 0.4915 | 0.6500 | 0.6133 | 0.6255 | 0.7006 | 0.7879 | 0.5389 |
| 0.0912 | 8.0 | 1056 | 0.5281 | 0.5468 | 0.6890 | 0.6097 | 0.9574 | 0.5370 | 0.7143 | 0.6866 | 0.5854 | 0.6939 | 0.7742 | 0.5491 |
| 0.0912 | 9.0 | 1188 | 0.4744 | 0.5371 | 0.7358 | 0.6209 | 0.9560 | 0.5370 | 0.6341 | 0.6753 | 0.6554 | 0.6795 | 0.7059 | 0.5699 |
| 0.0912 | 10.0 | 1320 | 0.5498 | 0.5686 | 0.7073 | 0.6304 | 0.9586 | 0.5370 | 0.6349 | 0.7500 | 0.6553 | 0.7152 | 0.7742 | 0.5573 |
| 0.0912 | 11.0 | 1452 | 0.6424 | 0.5857 | 0.7012 | 0.6383 | 0.9597 | 0.56 | 0.6789 | 0.7246 | 0.6667 | 0.6974 | 0.6875 | 0.5757 |
| 0.0354 | 12.0 | 1584 | 0.5867 | 0.5641 | 0.6890 | 0.6203 | 0.9585 | 0.5185 | 0.6496 | 0.7213 | 0.6619 | 0.7152 | 0.7333 | 0.5402 |
| 0.0354 | 13.0 | 1716 | 0.5500 | 0.5667 | 0.6992 | 0.6260 | 0.9592 | 0.5524 | 0.6829 | 0.7222 | 0.6621 | 0.6466 | 0.7333 | 0.5607 |
| 0.0354 | 14.0 | 1848 | 0.5743 | 0.5780 | 0.7154 | 0.6394 | 0.9596 | 0.5283 | 0.6833 | 0.7222 | 0.6644 | 0.6716 | 0.7742 | 0.5960 |
| 0.0354 | 15.0 | 1980 | 0.6836 | 0.5912 | 0.6850 | 0.6347 | 0.9609 | 0.5882 | 0.6897 | 0.7619 | 0.6525 | 0.7500 | 0.6452 | 0.5370 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu102
- Datasets 2.6.1
- Tokenizers 0.13.1
