|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: kaz_legal_bert_5 |
|
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. --> |
|
|
|
# kaz_legal_bert_5 |
|
|
|
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 4.8262 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 8.1161 | 0.08 | 1000 | 7.7116 | |
|
| 7.6258 | 0.17 | 2000 | 7.4606 | |
|
| 7.4268 | 0.25 | 3000 | 7.3184 | |
|
| 7.2837 | 0.34 | 4000 | 7.2020 | |
|
| 7.1969 | 0.42 | 5000 | 7.1236 | |
|
| 7.1201 | 0.5 | 6000 | 7.0599 | |
|
| 7.0683 | 0.59 | 7000 | 6.9990 | |
|
| 6.9956 | 0.67 | 8000 | 6.9369 | |
|
| 6.9392 | 0.76 | 9000 | 6.8828 | |
|
| 6.8949 | 0.84 | 10000 | 6.8263 | |
|
| 6.8437 | 0.92 | 11000 | 6.7913 | |
|
| 6.8027 | 1.01 | 12000 | 6.7392 | |
|
| 6.7539 | 1.09 | 13000 | 6.7010 | |
|
| 6.7316 | 1.18 | 14000 | 6.6663 | |
|
| 6.6853 | 1.26 | 15000 | 6.6338 | |
|
| 6.6449 | 1.34 | 16000 | 6.6004 | |
|
| 6.6188 | 1.43 | 17000 | 6.5463 | |
|
| 6.5831 | 1.51 | 18000 | 6.5042 | |
|
| 6.5498 | 1.6 | 19000 | 6.4581 | |
|
| 6.5116 | 1.68 | 20000 | 6.4205 | |
|
| 6.4579 | 1.77 | 21000 | 6.3473 | |
|
| 6.3996 | 1.85 | 22000 | 6.2794 | |
|
| 6.3358 | 1.93 | 23000 | 6.2082 | |
|
| 6.2827 | 2.02 | 24000 | 6.1448 | |
|
| 6.2381 | 2.1 | 25000 | 6.0923 | |
|
| 6.1947 | 2.19 | 26000 | 6.0460 | |
|
| 6.1479 | 2.27 | 27000 | 6.0002 | |
|
| 6.1095 | 2.35 | 28000 | 5.9537 | |
|
| 6.0669 | 2.44 | 29000 | 5.9139 | |
|
| 6.0411 | 2.52 | 30000 | 5.8827 | |
|
| 6.0081 | 2.61 | 31000 | 5.8454 | |
|
| 5.9939 | 2.69 | 32000 | 5.8276 | |
|
| 5.9714 | 2.77 | 33000 | 5.8060 | |
|
| 5.9524 | 2.86 | 34000 | 5.7878 | |
|
| 5.9357 | 2.94 | 35000 | 5.7772 | |
|
| 5.9705 | 3.03 | 36000 | 5.7964 | |
|
| 5.9276 | 3.11 | 37000 | 5.7410 | |
|
| 5.8802 | 3.19 | 38000 | 5.6813 | |
|
| 5.8342 | 3.28 | 39000 | 5.6268 | |
|
| 5.786 | 3.36 | 40000 | 5.5729 | |
|
| 5.7328 | 3.45 | 41000 | 5.5030 | |
|
| 5.6604 | 3.53 | 42000 | 5.4495 | |
|
| 5.6102 | 3.61 | 43000 | 5.3746 | |
|
| 5.5296 | 3.7 | 44000 | 5.3149 | |
|
| 5.4876 | 3.78 | 45000 | 5.2536 | |
|
| 5.417 | 3.87 | 46000 | 5.2004 | |
|
| 5.3665 | 3.95 | 47000 | 5.1488 | |
|
| 5.3131 | 4.03 | 48000 | 5.0948 | |
|
| 5.2697 | 4.12 | 49000 | 5.0538 | |
|
| 5.2307 | 4.2 | 50000 | 5.0139 | |
|
| 5.1975 | 4.29 | 51000 | 4.9757 | |
|
| 5.1506 | 4.37 | 52000 | 4.9439 | |
|
| 5.1285 | 4.45 | 53000 | 4.9238 | |
|
| 5.1009 | 4.54 | 54000 | 4.8900 | |
|
| 5.072 | 4.62 | 55000 | 4.8735 | |
|
| 5.0533 | 4.71 | 56000 | 4.8548 | |
|
| 5.0412 | 4.79 | 57000 | 4.8421 | |
|
| 5.0319 | 4.88 | 58000 | 4.8326 | |
|
| 5.0119 | 4.96 | 59000 | 4.8262 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.28.1 |
|
- Pytorch 2.0.0+cu118 |
|
- Datasets 2.11.0 |
|
- Tokenizers 0.13.3 |
|
|
