|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
model-index: |
|
- name: opus-mt-ru-en-finetuned |
|
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. --> |
|
|
|
# opus-mt-ru-en-finetuned |
|
|
|
This model is a fine-tuned version of [kazandaev/opus-mt-ru-en-finetuned](https://huggingface.co/kazandaev/opus-mt-ru-en-finetuned) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0678 |
|
- Bleu: 41.9575 |
|
- Gen Len: 26.1221 |
|
|
|
## 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: 49 |
|
- eval_batch_size: 24 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |
|
|:-------------:|:-----:|:------:|:---------------:|:-------:|:-------:| |
|
| 0.9398 | 1.0 | 35147 | 1.1344 | 39.1234 | 26.0997 | |
|
| 0.9278 | 2.0 | 70294 | 1.1184 | 39.6701 | 26.2065 | |
|
| 0.9076 | 3.0 | 105441 | 1.1048 | 39.7062 | 26.0862 | |
|
| 0.8972 | 4.0 | 140588 | 1.0987 | 40.106 | 26.105 | |
|
| 0.8859 | 5.0 | 175735 | 1.0959 | 40.6067 | 26.1026 | |
|
| 0.8788 | 6.0 | 210882 | 1.0849 | 41.1633 | 26.0607 | |
|
| 0.8767 | 7.0 | 246029 | 1.0835 | 41.3413 | 26.0259 | |
|
| 0.8681 | 8.0 | 281176 | 1.0742 | 41.4849 | 26.1068 | |
|
| 0.8615 | 9.0 | 316323 | 1.0714 | 42.0031 | 26.1483 | |
|
| 0.87 | 10.0 | 351470 | 1.0678 | 41.9575 | 26.1221 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.16.2 |
|
- Pytorch 1.10.0+cu111 |
|
- Datasets 1.18.3 |
|
- Tokenizers 0.11.0 |
|
|
