---
language:
- ko
- ja
base_model: facebook/mbart-large-50-many-to-many-mmt
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: koja_reverse_13p_run1
  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. -->

# koja_reverse_13p_run1

This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.3532
- Bleu: 5.7574
- Gen Len: 20.4109

## 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
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- num_epochs: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | Bleu   | Gen Len |
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|
| 1.9693        | 0.42  | 800  | 1.5445          | 4.3096 | 20.2019 |
| 1.4914        | 0.83  | 1600 | 1.4233          | 4.307  | 20.1204 |
| 1.4272        | 1.25  | 2400 | 1.3761          | 4.8111 | 22.8555 |
| 1.1571        | 1.66  | 3200 | 1.3532          | 5.7574 | 21.151  |
| 1.0833        | 2.08  | 4000 | 1.3449          | 5.652  | 21.2158 |
| 0.9394        | 2.49  | 4800 | 1.3905          | 5.45   | 19.9005 |
| 0.8633        | 2.91  | 5600 | 1.3719          | 4.9973 | 19.7844 |


### Framework versions

- Transformers 4.34.1
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1