|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: legal-xlm-roberta-base |
|
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. --> |
|
|
|
# legal-xlm-roberta-base |
|
|
|
This model was trained from scratch on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5484 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- distributed_type: tpu |
|
- num_devices: 8 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 512 |
|
- total_eval_batch_size: 128 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_ratio: 0.05 |
|
- training_steps: 1000000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-------:|:---------------:| |
|
| 1.2285 | 0.05 | 50000 | 0.9298 | |
|
| 1.0417 | 0.1 | 100000 | 0.7723 | |
|
| 0.9525 | 0.15 | 150000 | 0.7258 | |
|
| 0.9668 | 0.2 | 200000 | 0.6884 | |
|
| 0.8949 | 0.25 | 250000 | 0.6714 | |
|
| 0.921 | 0.3 | 300000 | 0.6617 | |
|
| 0.8324 | 0.35 | 350000 | 0.6423 | |
|
| 0.8406 | 0.4 | 400000 | 0.6259 | |
|
| 0.8136 | 0.45 | 450000 | 0.6147 | |
|
| 0.8247 | 0.5 | 500000 | 0.6095 | |
|
| 0.8649 | 0.55 | 550000 | 0.5985 | |
|
| 0.8119 | 0.6 | 600000 | 0.5973 | |
|
| 0.8422 | 0.65 | 650000 | 0.5813 | |
|
| 0.8006 | 0.7 | 700000 | 0.5701 | |
|
| 0.8072 | 0.75 | 750000 | 0.5662 | |
|
| 0.8154 | 0.8 | 800000 | 0.5514 | |
|
| 0.7794 | 0.85 | 850000 | 0.5562 | |
|
| 0.7924 | 0.9 | 900000 | 0.5558 | |
|
| 0.8207 | 0.95 | 950000 | 0.5587 | |
|
| 0.8279 | 1.0 | 1000000 | 0.5484 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.20.1 |
|
- Pytorch 1.12.0+cu102 |
|
- Datasets 2.8.0 |
|
- Tokenizers 0.12.0 |
|
|
