|
--- |
|
license: apache-2.0 |
|
base_model: google/long-t5-tglobal-base |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- eur-lex-sum |
|
model-index: |
|
- name: LongT5_no_extraction_V1 |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# LongT5_no_extraction_V1 |
|
|
|
This model is a fine-tuned version of [google/long-t5-tglobal-base](https://huggingface.co/google/long-t5-tglobal-base) on the eur-lex-sum dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.3639 |
|
|
|
## 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 |
|
- distributed_type: multi-GPU |
|
- num_devices: 4 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 16 |
|
- total_eval_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 40 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-------:|:----:|:---------------:| |
|
| 3.2571 | 0.9963 | 68 | 1.8571 | |
|
| 2.6516 | 1.9927 | 136 | 1.7238 | |
|
| 2.2687 | 2.9890 | 204 | 1.6153 | |
|
| 2.0466 | 4.0 | 273 | 1.5414 | |
|
| 1.9659 | 4.9963 | 341 | 1.4955 | |
|
| 1.8813 | 5.9927 | 409 | 1.4752 | |
|
| 1.8277 | 6.9890 | 477 | 1.4571 | |
|
| 1.7626 | 8.0 | 546 | 1.4437 | |
|
| 1.7528 | 8.9963 | 614 | 1.4315 | |
|
| 1.7249 | 9.9927 | 682 | 1.4229 | |
|
| 1.6981 | 10.9890 | 750 | 1.4126 | |
|
| 1.6559 | 12.0 | 819 | 1.4061 | |
|
| 1.6599 | 12.9963 | 887 | 1.3983 | |
|
| 1.6465 | 13.9927 | 955 | 1.3994 | |
|
| 1.6282 | 14.9890 | 1023 | 1.3923 | |
|
| 1.5906 | 16.0 | 1092 | 1.3873 | |
|
| 1.6035 | 16.9963 | 1160 | 1.3878 | |
|
| 1.5909 | 17.9927 | 1228 | 1.3851 | |
|
| 1.5802 | 18.9890 | 1296 | 1.3799 | |
|
| 1.5481 | 20.0 | 1365 | 1.3860 | |
|
| 1.5607 | 20.9963 | 1433 | 1.3745 | |
|
| 1.5517 | 21.9927 | 1501 | 1.3736 | |
|
| 1.5436 | 22.9890 | 1569 | 1.3735 | |
|
| 1.5126 | 24.0 | 1638 | 1.3728 | |
|
| 1.5289 | 24.9963 | 1706 | 1.3739 | |
|
| 1.5234 | 25.9927 | 1774 | 1.3706 | |
|
| 1.5179 | 26.9890 | 1842 | 1.3671 | |
|
| 1.4908 | 28.0 | 1911 | 1.3680 | |
|
| 1.5057 | 28.9963 | 1979 | 1.3688 | |
|
| 1.5026 | 29.9927 | 2047 | 1.3649 | |
|
| 1.498 | 30.9890 | 2115 | 1.3662 | |
|
| 1.4866 | 32.0 | 2184 | 1.3655 | |
|
| 1.493 | 32.9963 | 2252 | 1.3644 | |
|
| 1.4877 | 33.9927 | 2320 | 1.3669 | |
|
| 1.4858 | 34.9890 | 2388 | 1.3650 | |
|
| 1.465 | 36.0 | 2457 | 1.3649 | |
|
| 1.4822 | 36.9963 | 2525 | 1.3647 | |
|
| 1.4797 | 37.9927 | 2593 | 1.3644 | |
|
| 1.4803 | 38.9890 | 2661 | 1.3640 | |
|
| 1.4548 | 39.8535 | 2720 | 1.3639 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.1 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.17.1 |
|
- Tokenizers 0.19.1 |
|
|
