<!-- 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. -->
neuroscience-to-dev-bio-4
This model is a fine-tuned version of facebook/bart-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0211
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 128
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 5000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
18.5056 | 0.97 | 8 | 18.2694 |
15.7993 | 1.97 | 16 | 14.5706 |
12.8347 | 2.97 | 24 | 12.0677 |
11.5971 | 3.97 | 32 | 10.8629 |
10.463 | 4.97 | 40 | 9.3275 |
8.9798 | 5.97 | 48 | 7.0959 |
7.2515 | 6.97 | 56 | 5.9271 |
6.2773 | 7.97 | 64 | 5.3001 |
5.636 | 8.97 | 72 | 4.7396 |
5.0218 | 9.97 | 80 | 4.1504 |
4.3526 | 10.97 | 88 | 3.4576 |
3.5813 | 11.97 | 96 | 2.6589 |
2.7243 | 12.97 | 104 | 1.7789 |
1.7997 | 13.97 | 112 | 0.9672 |
0.995 | 14.97 | 120 | 0.4184 |
0.4459 | 15.97 | 128 | 0.1611 |
0.1844 | 16.97 | 136 | 0.0645 |
0.077 | 17.97 | 144 | 0.0292 |
0.0332 | 18.97 | 152 | 0.0212 |
0.0197 | 19.97 | 160 | 0.0187 |
0.0151 | 20.97 | 168 | 0.0169 |
0.0245 | 21.97 | 176 | 0.0160 |
0.0099 | 22.97 | 184 | 0.0206 |
0.0094 | 23.97 | 192 | 0.0158 |
0.0082 | 24.97 | 200 | 0.0170 |
0.0063 | 25.97 | 208 | 0.0159 |
0.0075 | 26.97 | 216 | 0.0169 |
0.0059 | 27.97 | 224 | 0.0154 |
0.0047 | 28.97 | 232 | 0.0164 |
0.0045 | 29.97 | 240 | 0.0181 |
0.0037 | 30.97 | 248 | 0.0192 |
0.0038 | 31.97 | 256 | 0.0160 |
0.0045 | 32.97 | 264 | 0.0162 |
0.0056 | 33.97 | 272 | 0.0150 |
0.0043 | 34.97 | 280 | 0.0149 |
0.0036 | 35.97 | 288 | 0.0155 |
0.0032 | 36.97 | 296 | 0.0183 |
0.0032 | 37.97 | 304 | 0.0158 |
0.0028 | 38.97 | 312 | 0.0155 |
0.0032 | 39.97 | 320 | 0.0160 |
0.0027 | 40.97 | 328 | 0.0180 |
0.0033 | 41.97 | 336 | 0.0164 |
0.0035 | 42.97 | 344 | 0.0211 |
Framework versions
- Transformers 4.21.3
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1