File size: 3,596 Bytes
4a15ec8 |
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 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 |
---
language:
- ko
- en
base_model: facebook/mbart-large-50-many-to-many-mmt
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: koen_mbartLarge_64p_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. -->
# koen_mbartLarge_64p_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: 1.0989
- Bleu: 33.8958
- Gen Len: 18.5033
## 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: 500
- num_epochs: 10
### Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
| 1.5543 | 0.13 | 2500 | 1.5068 | 25.2368 | 18.5097 |
| 1.4399 | 0.26 | 5000 | 1.3972 | 27.0554 | 18.5539 |
| 1.3448 | 0.39 | 7500 | 1.3132 | 28.8579 | 18.6315 |
| 1.3205 | 0.52 | 10000 | 1.2873 | 29.5611 | 18.7781 |
| 1.2786 | 0.65 | 12500 | 1.2399 | 30.3042 | 18.5644 |
| 1.2561 | 0.78 | 15000 | 1.2173 | 30.5801 | 19.0186 |
| 1.2479 | 0.91 | 17500 | 1.2125 | 30.8896 | 18.7636 |
| 1.1891 | 1.04 | 20000 | 1.1776 | 31.9834 | 18.7002 |
| 1.1943 | 1.17 | 22500 | 1.1651 | 32.0205 | 18.7054 |
| 1.1375 | 1.3 | 25000 | 1.1492 | 32.3658 | 18.6287 |
| 1.1351 | 1.43 | 27500 | 1.1460 | 32.339 | 18.7655 |
| 1.0859 | 1.56 | 30000 | 1.1623 | 31.5418 | 19.016 |
| 1.0373 | 1.69 | 32500 | 1.1383 | 32.672 | 18.7224 |
| 1.0824 | 1.82 | 35000 | 1.1232 | 33.2231 | 18.6697 |
| 1.0242 | 1.95 | 37500 | 1.1313 | 32.813 | 18.2553 |
| 1.0649 | 2.08 | 40000 | 1.1182 | 33.2021 | 18.7216 |
| 1.054 | 2.21 | 42500 | 1.1329 | 33.0588 | 18.4992 |
| 1.0143 | 2.34 | 45000 | 1.1187 | 33.2176 | 18.7156 |
| 1.0037 | 2.47 | 47500 | 1.1162 | 33.3754 | 18.6443 |
| 0.9928 | 2.61 | 50000 | 1.1306 | 33.0727 | 18.6361 |
| 0.9497 | 2.74 | 52500 | 1.1170 | 33.227 | 18.7638 |
| 1.0157 | 2.87 | 55000 | 1.1072 | 33.685 | 18.5847 |
| 0.9876 | 3.0 | 57500 | 1.1035 | 33.6971 | 18.6873 |
| 0.9665 | 3.13 | 60000 | 1.0989 | 33.8919 | 18.5258 |
| 0.9197 | 3.26 | 62500 | 1.1060 | 33.7036 | 18.5407 |
| 0.9427 | 3.39 | 65000 | 1.0995 | 33.7642 | 18.7 |
| 0.8993 | 3.52 | 67500 | 1.1364 | 33.1757 | 18.646 |
| 0.8957 | 3.65 | 70000 | 1.1251 | 33.0954 | 18.3129 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1
|