|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- ccmatrix |
|
|
metrics: |
|
|
- bleu |
|
|
model-index: |
|
|
- name: t5-small-finetuned-en-to-it |
|
|
results: |
|
|
- task: |
|
|
name: Sequence-to-sequence Language Modeling |
|
|
type: text2text-generation |
|
|
dataset: |
|
|
name: ccmatrix |
|
|
type: ccmatrix |
|
|
config: en-it |
|
|
split: train[3000:15000] |
|
|
args: en-it |
|
|
metrics: |
|
|
- name: Bleu |
|
|
type: bleu |
|
|
value: 7.3298 |
|
|
--- |
|
|
|
|
|
<!-- 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. --> |
|
|
|
|
|
# t5-small-finetuned-en-to-it |
|
|
|
|
|
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the ccmatrix dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 2.2698 |
|
|
- Bleu: 7.3298 |
|
|
- Gen Len: 62.3753 |
|
|
|
|
|
## 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: 96 |
|
|
- eval_batch_size: 96 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 40 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |
|
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:| |
|
|
| No log | 1.0 | 125 | 3.0010 | 2.7294 | 56.4513 | |
|
|
| No log | 2.0 | 250 | 2.8999 | 2.3228 | 81.4993 | |
|
|
| No log | 3.0 | 375 | 2.8281 | 2.3065 | 92.3353 | |
|
|
| 3.3202 | 4.0 | 500 | 2.7722 | 2.5982 | 91.8093 | |
|
|
| 3.3202 | 5.0 | 625 | 2.7254 | 2.9279 | 89.0907 | |
|
|
| 3.3202 | 6.0 | 750 | 2.6839 | 3.0747 | 89.2827 | |
|
|
| 3.3202 | 7.0 | 875 | 2.6470 | 3.207 | 87.948 | |
|
|
| 3.0355 | 8.0 | 1000 | 2.6132 | 3.355 | 85.2487 | |
|
|
| 3.0355 | 9.0 | 1125 | 2.5835 | 3.8401 | 80.578 | |
|
|
| 3.0355 | 10.0 | 1250 | 2.5552 | 4.2905 | 75.818 | |
|
|
| 3.0355 | 11.0 | 1375 | 2.5323 | 4.3866 | 75.2433 | |
|
|
| 2.8903 | 12.0 | 1500 | 2.5079 | 4.5687 | 74.906 | |
|
|
| 2.8903 | 13.0 | 1625 | 2.4881 | 4.7844 | 71.5773 | |
|
|
| 2.8903 | 14.0 | 1750 | 2.4668 | 4.876 | 71.68 | |
|
|
| 2.8903 | 15.0 | 1875 | 2.4485 | 5.1292 | 70.118 | |
|
|
| 2.7891 | 16.0 | 2000 | 2.4322 | 5.3297 | 68.894 | |
|
|
| 2.7891 | 17.0 | 2125 | 2.4161 | 5.555 | 68.2293 | |
|
|
| 2.7891 | 18.0 | 2250 | 2.4010 | 5.7113 | 67.2907 | |
|
|
| 2.7891 | 19.0 | 2375 | 2.3892 | 5.9105 | 66.6287 | |
|
|
| 2.713 | 20.0 | 2500 | 2.3756 | 6.0057 | 66.112 | |
|
|
| 2.713 | 21.0 | 2625 | 2.3643 | 6.3118 | 64.6193 | |
|
|
| 2.713 | 22.0 | 2750 | 2.3533 | 6.476 | 64.31 | |
|
|
| 2.713 | 23.0 | 2875 | 2.3432 | 6.7102 | 63.5467 | |
|
|
| 2.6584 | 24.0 | 3000 | 2.3342 | 6.7604 | 63.6567 | |
|
|
| 2.6584 | 25.0 | 3125 | 2.3253 | 6.8418 | 63.6573 | |
|
|
| 2.6584 | 26.0 | 3250 | 2.3180 | 6.9165 | 63.5893 | |
|
|
| 2.6584 | 27.0 | 3375 | 2.3120 | 7.0217 | 63.1033 | |
|
|
| 2.616 | 28.0 | 3500 | 2.3056 | 6.9148 | 63.598 | |
|
|
| 2.616 | 29.0 | 3625 | 2.2987 | 6.9961 | 63.6267 | |
|
|
| 2.616 | 30.0 | 3750 | 2.2935 | 7.2238 | 62.8373 | |
|
|
| 2.616 | 31.0 | 3875 | 2.2892 | 7.1906 | 62.7793 | |
|
|
| 2.587 | 32.0 | 4000 | 2.2849 | 7.2052 | 63.126 | |
|
|
| 2.587 | 33.0 | 4125 | 2.2815 | 7.3272 | 62.526 | |
|
|
| 2.587 | 34.0 | 4250 | 2.2782 | 7.3603 | 62.4313 | |
|
|
| 2.587 | 35.0 | 4375 | 2.2756 | 7.3072 | 62.6307 | |
|
|
| 2.5673 | 36.0 | 4500 | 2.2737 | 7.3586 | 62.1633 | |
|
|
| 2.5673 | 37.0 | 4625 | 2.2718 | 7.3485 | 62.358 | |
|
|
| 2.5673 | 38.0 | 4750 | 2.2707 | 7.3406 | 62.298 | |
|
|
| 2.5673 | 39.0 | 4875 | 2.2700 | 7.3233 | 62.42 | |
|
|
| 2.5591 | 40.0 | 5000 | 2.2698 | 7.3298 | 62.3753 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.22.1 |
|
|
- Pytorch 1.12.1 |
|
|
- Datasets 2.5.1 |
|
|
- Tokenizers 0.11.0 |
|
|
|
