metadata
tags:
- generated_from_trainer
datasets:
- generator
model-index:
- name: scideberta-cs-tdm-pretrained-finetuned-ner
results: []
scideberta-cs-tdm-pretrained-finetuned-ner
This model was trained from scratch on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.8293
- Overall Precision: 0.6327
- Overall Recall: 0.7460
- Overall F1: 0.6847
- Overall Accuracy: 0.9608
- Datasetname F1: 0.6968
- Hyperparametername F1: 0.6765
- Hyperparametervalue F1: 0.7289
- Methodname F1: 0.7290
- Metricname F1: 0.5269
- Metricvalue F1: 0.8235
- Taskname F1: 0.6099
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 | 131 | 0.4448 | 0.4113 | 0.6147 | 0.4929 | 0.9353 | 0.5312 | 0.3736 | 0.4818 | 0.6256 | 0.4667 | 0.2456 | 0.4526 |
No log | 2.0 | 262 | 0.3527 | 0.4341 | 0.7067 | 0.5378 | 0.9416 | 0.5347 | 0.4549 | 0.5487 | 0.6256 | 0.5026 | 0.72 | 0.4593 |
No log | 3.0 | 393 | 0.4857 | 0.5794 | 0.6491 | 0.6123 | 0.9544 | 0.6420 | 0.5263 | 0.6011 | 0.7030 | 0.5276 | 0.7838 | 0.5385 |
0.3806 | 4.0 | 524 | 0.3789 | 0.4923 | 0.7485 | 0.5940 | 0.9492 | 0.6358 | 0.5418 | 0.6165 | 0.6166 | 0.5227 | 0.7826 | 0.5690 |
0.3806 | 5.0 | 655 | 0.4563 | 0.5736 | 0.7313 | 0.6429 | 0.9568 | 0.6298 | 0.6176 | 0.7143 | 0.6824 | 0.5402 | 0.8090 | 0.5463 |
0.3806 | 6.0 | 786 | 0.4021 | 0.5199 | 0.7215 | 0.6043 | 0.9525 | 0.6581 | 0.5848 | 0.5603 | 0.6431 | 0.4973 | 0.7579 | 0.5738 |
0.3806 | 7.0 | 917 | 0.4851 | 0.5614 | 0.7460 | 0.6407 | 0.9565 | 0.6506 | 0.6199 | 0.6888 | 0.6982 | 0.4787 | 0.7826 | 0.5571 |
0.0724 | 8.0 | 1048 | 0.5002 | 0.5890 | 0.7350 | 0.6539 | 0.9583 | 0.6316 | 0.6150 | 0.7273 | 0.7098 | 0.5357 | 0.8140 | 0.5636 |
0.0724 | 9.0 | 1179 | 0.5948 | 0.6036 | 0.7325 | 0.6619 | 0.9589 | 0.6839 | 0.6408 | 0.6991 | 0.7165 | 0.4918 | 0.7692 | 0.6140 |
0.0724 | 10.0 | 1310 | 0.5070 | 0.5716 | 0.7497 | 0.6486 | 0.9566 | 0.6582 | 0.6164 | 0.6812 | 0.6949 | 0.5371 | 0.7692 | 0.5929 |
0.0724 | 11.0 | 1441 | 0.6557 | 0.6339 | 0.7350 | 0.6807 | 0.9614 | 0.6883 | 0.6650 | 0.7373 | 0.7364 | 0.5143 | 0.8293 | 0.5956 |
0.0285 | 12.0 | 1572 | 0.5910 | 0.5713 | 0.7374 | 0.6438 | 0.9574 | 0.6835 | 0.6150 | 0.6754 | 0.7099 | 0.5114 | 0.6792 | 0.5763 |
0.0285 | 13.0 | 1703 | 0.6679 | 0.6188 | 0.7350 | 0.6719 | 0.9607 | 0.6928 | 0.6539 | 0.7232 | 0.7280 | 0.5 | 0.8333 | 0.5728 |
0.0285 | 14.0 | 1834 | 0.6856 | 0.6246 | 0.7227 | 0.6701 | 0.9612 | 0.6579 | 0.6256 | 0.7123 | 0.7452 | 0.5128 | 0.8148 | 0.6018 |
0.0285 | 15.0 | 1965 | 0.7225 | 0.6238 | 0.7387 | 0.6764 | 0.9606 | 0.6962 | 0.6586 | 0.7117 | 0.7290 | 0.4878 | 0.8095 | 0.6283 |
0.0154 | 16.0 | 2096 | 0.7242 | 0.5980 | 0.7301 | 0.6575 | 0.9591 | 0.6752 | 0.6293 | 0.6987 | 0.7148 | 0.5030 | 0.8193 | 0.5714 |
0.0154 | 17.0 | 2227 | 0.7268 | 0.6282 | 0.7276 | 0.6742 | 0.9606 | 0.7006 | 0.6568 | 0.7059 | 0.7255 | 0.5269 | 0.8148 | 0.5963 |
0.0154 | 18.0 | 2358 | 0.7498 | 0.6233 | 0.7411 | 0.6771 | 0.9606 | 0.6962 | 0.6402 | 0.7321 | 0.7280 | 0.5422 | 0.8434 | 0.5899 |
0.0154 | 19.0 | 2489 | 0.7161 | 0.6202 | 0.7534 | 0.6803 | 0.9595 | 0.7051 | 0.6479 | 0.7085 | 0.7524 | 0.5269 | 0.8148 | 0.5919 |
0.0104 | 20.0 | 2620 | 0.7926 | 0.6315 | 0.7129 | 0.6697 | 0.9615 | 0.6797 | 0.6502 | 0.7027 | 0.7269 | 0.5357 | 0.7949 | 0.5905 |
0.0104 | 21.0 | 2751 | 0.7827 | 0.6464 | 0.7423 | 0.6910 | 0.9626 | 0.7190 | 0.6751 | 0.7123 | 0.7395 | 0.5562 | 0.8205 | 0.6197 |
0.0104 | 22.0 | 2882 | 0.7285 | 0.6300 | 0.7521 | 0.6857 | 0.9599 | 0.7097 | 0.6782 | 0.7207 | 0.7215 | 0.5333 | 0.8333 | 0.6188 |
0.0049 | 23.0 | 3013 | 0.7645 | 0.6413 | 0.7350 | 0.6850 | 0.9620 | 0.6968 | 0.6717 | 0.7182 | 0.7301 | 0.5476 | 0.8395 | 0.6066 |
0.0049 | 24.0 | 3144 | 0.8071 | 0.6466 | 0.7387 | 0.6896 | 0.9616 | 0.7105 | 0.6886 | 0.7189 | 0.7362 | 0.5535 | 0.775 | 0.6019 |
0.0049 | 25.0 | 3275 | 0.8324 | 0.6319 | 0.7350 | 0.6795 | 0.9611 | 0.7059 | 0.6683 | 0.6964 | 0.7280 | 0.5366 | 0.8193 | 0.6063 |
0.0049 | 26.0 | 3406 | 0.8235 | 0.6355 | 0.7337 | 0.6811 | 0.9606 | 0.6928 | 0.6700 | 0.7189 | 0.7328 | 0.5610 | 0.8250 | 0.5674 |
0.004 | 27.0 | 3537 | 0.8106 | 0.6220 | 0.7411 | 0.6764 | 0.9602 | 0.7089 | 0.6536 | 0.7000 | 0.7495 | 0.5089 | 0.85 | 0.5611 |
0.004 | 28.0 | 3668 | 0.8271 | 0.6353 | 0.7460 | 0.6862 | 0.9611 | 0.7013 | 0.6634 | 0.7054 | 0.7457 | 0.5644 | 0.8293 | 0.5936 |
0.004 | 29.0 | 3799 | 0.8630 | 0.6400 | 0.7374 | 0.6853 | 0.9613 | 0.6923 | 0.6634 | 0.7189 | 0.7348 | 0.5783 | 0.8537 | 0.5888 |
0.004 | 30.0 | 3930 | 0.8055 | 0.6163 | 0.7411 | 0.6730 | 0.9598 | 0.7226 | 0.6522 | 0.7074 | 0.7063 | 0.5176 | 0.8537 | 0.6161 |
0.0029 | 31.0 | 4061 | 0.8293 | 0.6327 | 0.7460 | 0.6847 | 0.9608 | 0.6968 | 0.6765 | 0.7289 | 0.7290 | 0.5269 | 0.8235 | 0.6099 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu102
- Datasets 2.6.1
- Tokenizers 0.13.1