metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: bert_large_subjqa_model_v4
results: []
bert_large_subjqa_model_v4
This model is a fine-tuned version of bert-large-uncased-whole-word-masking-finetuned-squad on the None dataset. It achieves the following results on the evaluation set:
- Loss: 13.7958
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: 15
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 1.0 | 410 | 11.2497 |
| 0.1447 | 2.0 | 820 | 10.3669 |
| 0.1913 | 3.0 | 1230 | 11.5822 |
| 0.1439 | 4.0 | 1640 | 11.7240 |
| 0.0995 | 5.0 | 2050 | 11.5282 |
| 0.0995 | 6.0 | 2460 | 11.7716 |
| 0.1058 | 7.0 | 2870 | 13.3381 |
| 0.0709 | 8.0 | 3280 | 12.5043 |
| 0.0529 | 9.0 | 3690 | 12.8842 |
| 0.0441 | 10.0 | 4100 | 12.8577 |
| 0.0378 | 11.0 | 4510 | 13.0998 |
| 0.0378 | 12.0 | 4920 | 13.8685 |
| 0.0429 | 13.0 | 5330 | 13.8242 |
| 0.0333 | 14.0 | 5740 | 13.8526 |
| 0.0331 | 15.0 | 6150 | 13.7958 |
Framework versions
- Transformers 4.28.0
- Pytorch 1.13.0a0+d321be6
- Datasets 2.12.0
- Tokenizers 0.13.3