|
--- |
|
license: apache-2.0 |
|
base_model: t5-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
- wer |
|
model-index: |
|
- name: 10_randomization_model |
|
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. --> |
|
|
|
# 10_randomization_model |
|
|
|
This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1399 |
|
- Bleu: 0.0001 |
|
- Wer: 0.9311 |
|
- Rougel: 0.1663 |
|
- Gen Len: 18.9987 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 2 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Wer | Rougel | Gen Len | |
|
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:-------:| |
|
| 0.716 | 0.16 | 1000 | 0.2572 | 0.0001 | 0.932 | 0.1648 | 18.9987 | |
|
| 0.2981 | 0.32 | 2000 | 0.2055 | 0.0001 | 0.9317 | 0.1655 | 18.9987 | |
|
| 0.2596 | 0.48 | 3000 | 0.1836 | 0.0001 | 0.9315 | 0.1658 | 18.9987 | |
|
| 0.2371 | 0.64 | 4000 | 0.1685 | 0.0001 | 0.9314 | 0.1659 | 18.9987 | |
|
| 0.2266 | 0.8 | 5000 | 0.1616 | 0.0001 | 0.9313 | 0.1661 | 18.9987 | |
|
| 0.2134 | 0.96 | 6000 | 0.1531 | 0.0001 | 0.9313 | 0.1662 | 18.9987 | |
|
| 0.2035 | 1.12 | 7000 | 0.1505 | 0.0001 | 0.9312 | 0.1662 | 18.9987 | |
|
| 0.1973 | 1.28 | 8000 | 0.1466 | 0.0001 | 0.9312 | 0.1663 | 18.9987 | |
|
| 0.1942 | 1.44 | 9000 | 0.1430 | 0.0001 | 0.9312 | 0.1663 | 18.9987 | |
|
| 0.1905 | 1.6 | 10000 | 0.1416 | 0.0001 | 0.9312 | 0.1663 | 18.9987 | |
|
| 0.1892 | 1.76 | 11000 | 0.1402 | 0.0001 | 0.9312 | 0.1663 | 18.9987 | |
|
| 0.1867 | 1.92 | 12000 | 0.1399 | 0.0001 | 0.9311 | 0.1663 | 18.9987 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.37.1 |
|
- Pytorch 2.3.0.dev20240122+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.1 |
|
|
