|
--- |
|
license: apache-2.0 |
|
base_model: google/long-t5-tglobal-base |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- arrow |
|
model-index: |
|
- name: RoBERTa_LongT5_dependent_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. --> |
|
|
|
# RoBERTa_LongT5_dependent_V1 |
|
|
|
This model is a fine-tuned version of [google/long-t5-tglobal-base](https://huggingface.co/google/long-t5-tglobal-base) on the arrow dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.5152 |
|
|
|
## 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 | |
|
|:-------------:|:-------:|:----:|:---------------:| |
|
| 3.4528 | 0.9963 | 68 | 1.9964 | |
|
| 2.8835 | 1.9927 | 136 | 1.8580 | |
|
| 2.5241 | 2.9890 | 204 | 1.7485 | |
|
| 2.2845 | 4.0 | 273 | 1.6858 | |
|
| 2.1993 | 4.9963 | 341 | 1.6422 | |
|
| 2.1193 | 5.9927 | 409 | 1.6253 | |
|
| 2.067 | 6.9890 | 477 | 1.6027 | |
|
| 1.9859 | 8.0 | 546 | 1.5902 | |
|
| 1.9823 | 8.9963 | 614 | 1.5784 | |
|
| 1.9528 | 9.9927 | 682 | 1.5714 | |
|
| 1.9304 | 10.9890 | 750 | 1.5636 | |
|
| 1.8756 | 12.0 | 819 | 1.5591 | |
|
| 1.891 | 12.9963 | 887 | 1.5537 | |
|
| 1.8688 | 13.9927 | 955 | 1.5506 | |
|
| 1.8497 | 14.9890 | 1023 | 1.5423 | |
|
| 1.8089 | 16.0 | 1092 | 1.5425 | |
|
| 1.8222 | 16.9963 | 1160 | 1.5369 | |
|
| 1.8087 | 17.9927 | 1228 | 1.5376 | |
|
| 1.7963 | 18.9890 | 1296 | 1.5328 | |
|
| 1.7618 | 20.0 | 1365 | 1.5321 | |
|
| 1.7753 | 20.9963 | 1433 | 1.5267 | |
|
| 1.7671 | 21.9927 | 1501 | 1.5280 | |
|
| 1.7578 | 22.9890 | 1569 | 1.5248 | |
|
| 1.7261 | 24.0 | 1638 | 1.5268 | |
|
| 1.7427 | 24.9963 | 1706 | 1.5265 | |
|
| 1.7338 | 25.9927 | 1774 | 1.5221 | |
|
| 1.7303 | 26.9890 | 1842 | 1.5214 | |
|
| 1.6963 | 28.0 | 1911 | 1.5201 | |
|
| 1.7173 | 28.9963 | 1979 | 1.5178 | |
|
| 1.7132 | 29.9927 | 2047 | 1.5180 | |
|
| 1.7088 | 30.9890 | 2115 | 1.5167 | |
|
| 1.6809 | 32.0 | 2184 | 1.5155 | |
|
| 1.7037 | 32.9963 | 2252 | 1.5162 | |
|
| 1.699 | 33.9927 | 2320 | 1.5161 | |
|
| 1.6964 | 34.9890 | 2388 | 1.5152 | |
|
| 1.6718 | 36.0 | 2457 | 1.5154 | |
|
| 1.6944 | 36.9963 | 2525 | 1.5151 | |
|
| 1.6909 | 37.9927 | 2593 | 1.5154 | |
|
| 1.6884 | 38.9890 | 2661 | 1.5152 | |
|
| 1.6631 | 39.8535 | 2720 | 1.5152 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.1 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.17.1 |
|
- Tokenizers 0.19.1 |
|
|
