<!-- 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. -->
BART_pretrained_on_billsum_finetuned_on_small_SCOTUS_extracted_dataset
This model is a fine-tuned version of bheshaj/bart-large-billsum-epochs20 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.6264
- Rouge1: 0.1228
- Rouge2: 0.0388
- Rougel: 0.0995
- Rougelsum: 0.1001
- Gen Len: 19.1037
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
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
No log | 0.98 | 10 | 4.5839 | 0.107 | 0.0259 | 0.0896 | 0.0896 | 20.0 |
No log | 1.98 | 20 | 4.2717 | 0.1329 | 0.0325 | 0.1086 | 0.1086 | 19.7378 |
No log | 2.98 | 30 | 4.1413 | 0.1351 | 0.0333 | 0.1102 | 0.1105 | 19.7134 |
No log | 3.98 | 40 | 4.0582 | 0.1406 | 0.0406 | 0.1151 | 0.115 | 19.5976 |
No log | 4.98 | 50 | 3.9916 | 0.1418 | 0.0407 | 0.1144 | 0.1144 | 19.6829 |
No log | 5.98 | 60 | 3.9373 | 0.1244 | 0.0356 | 0.1 | 0.1001 | 18.7134 |
No log | 6.98 | 70 | 3.8867 | 0.1184 | 0.0322 | 0.0967 | 0.0972 | 18.4146 |
No log | 7.98 | 80 | 3.8363 | 0.122 | 0.0325 | 0.0979 | 0.0981 | 18.4451 |
No log | 8.98 | 90 | 3.7884 | 0.1183 | 0.0324 | 0.096 | 0.0967 | 18.8902 |
No log | 9.98 | 100 | 3.7625 | 0.1153 | 0.0318 | 0.0939 | 0.094 | 18.8049 |
No log | 10.98 | 110 | 3.7269 | 0.1063 | 0.0296 | 0.0882 | 0.0882 | 18.2378 |
No log | 11.98 | 120 | 3.7023 | 0.1194 | 0.0314 | 0.0977 | 0.0981 | 19.1646 |
No log | 12.98 | 130 | 3.6839 | 0.1275 | 0.0371 | 0.1053 | 0.105 | 19.5488 |
No log | 13.98 | 140 | 3.6693 | 0.1237 | 0.0354 | 0.0998 | 0.1004 | 19.2561 |
No log | 14.98 | 150 | 3.6578 | 0.1246 | 0.0371 | 0.1022 | 0.1023 | 19.1951 |
No log | 15.98 | 160 | 3.6447 | 0.1299 | 0.0398 | 0.1062 | 0.1062 | 19.4329 |
No log | 16.98 | 170 | 3.6366 | 0.1247 | 0.0355 | 0.1004 | 0.1007 | 19.1829 |
No log | 17.98 | 180 | 3.6319 | 0.123 | 0.038 | 0.0994 | 0.1 | 19.0122 |
No log | 18.98 | 190 | 3.6279 | 0.1225 | 0.0386 | 0.0994 | 0.1002 | 19.122 |
No log | 19.98 | 200 | 3.6264 | 0.1228 | 0.0388 | 0.0995 | 0.1001 | 19.1037 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2