File size: 2,013 Bytes
c4317cc 7fe66d2 c4317cc 4e90018 c4317cc 4e90018 ebac85d 7fe66d2 c4317cc ebac85d c4317cc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 |
---
language:
- es
- maq
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: byt5-base-es_maq
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. -->
# byt5-base-es_maq
This model is a fine-tuned version of [google/byt5-base](https://huggingface.co/google/byt5-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7257
- Bleu: 9.3405
- Gen Len: 99.8552
## 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: 16
- eval_batch_size: 16
- seed: 65
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|
| No log | 1.0 | 398 | 0.9986 | 0.0463 | 19.0 |
| 1.43 | 2.0 | 796 | 0.8809 | 0.5462 | 19.0 |
| 0.9934 | 3.0 | 1194 | 0.8291 | 0.4119 | 19.0 |
| 0.8998 | 4.0 | 1592 | 0.7965 | 0.4437 | 19.0 |
| 0.8998 | 5.0 | 1990 | 0.7747 | 0.3303 | 19.0 |
| 0.851 | 6.0 | 2388 | 0.7562 | 0.3764 | 19.0 |
| 0.8139 | 7.0 | 2786 | 0.7415 | 0.3244 | 19.0 |
| 0.7865 | 8.0 | 3184 | 0.7341 | 0.4334 | 19.0 |
| 0.7767 | 9.0 | 3582 | 0.7294 | 0.3918 | 19.0 |
| 0.7767 | 10.0 | 3980 | 0.7257 | 0.3979 | 19.0 |
### Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
|