|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: legal-italian-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-italian-roberta-base |
|
|
|
This model was trained from scratch on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4799 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:-------:|:---------------:| |
|
| 1.0248 | 0.05 | 50000 | 0.8033 | |
|
| 0.912 | 0.1 | 100000 | 0.6825 | |
|
| 0.8853 | 1.0 | 150000 | 0.6205 | |
|
| 0.847 | 1.05 | 200000 | 0.5954 | |
|
| 0.8395 | 1.1 | 250000 | 0.5859 | |
|
| 0.7485 | 2.01 | 300000 | 0.5632 | |
|
| 0.7154 | 2.06 | 350000 | 0.5495 | |
|
| 0.6851 | 2.11 | 400000 | 0.5456 | |
|
| 0.6074 | 3.01 | 450000 | 0.5331 | |
|
| 0.6296 | 3.06 | 500000 | 0.5226 | |
|
| 0.6125 | 3.11 | 550000 | 0.5146 | |
|
| 0.5983 | 4.02 | 600000 | 0.5038 | |
|
| 0.6471 | 4.07 | 650000 | 0.4976 | |
|
| 0.633 | 4.12 | 700000 | 0.4982 | |
|
| 0.6917 | 5.02 | 750000 | 0.4906 | |
|
| 0.7178 | 5.07 | 800000 | 0.4833 | |
|
| 0.6988 | 5.12 | 850000 | 0.4754 | |
|
| 0.7135 | 6.02 | 900000 | 0.4734 | |
|
| 0.7269 | 6.07 | 950000 | 0.4826 | |
|
| 0.7085 | 6.12 | 1000000 | 0.4799 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.20.1 |
|
- Pytorch 1.12.0+cu102 |
|
- Datasets 2.8.0 |
|
- Tokenizers 0.12.1 |
|
|
