Sim
commited on
Commit
•
ef8b319
1
Parent(s):
e117888
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- bleu
|
| 7 |
+
model-index:
|
| 8 |
+
- name: NLLB-600m-nlg_Latn-to-eng_Latn
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# NLLB-600m-nlg_Latn-to-eng_Latn
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/nllb-200-distilled-600M](https://huggingface.co/facebook/nllb-200-distilled-600M) on the None dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.9402
|
| 20 |
+
- Bleu: 45.9717
|
| 21 |
+
- Gen Len: 42.476
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 2e-05
|
| 41 |
+
- train_batch_size: 2
|
| 42 |
+
- eval_batch_size: 2
|
| 43 |
+
- seed: 42
|
| 44 |
+
- gradient_accumulation_steps: 6
|
| 45 |
+
- total_train_batch_size: 12
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: linear
|
| 48 |
+
- training_steps: 10000
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
| 54 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
|
| 55 |
+
| 2.5032 | 0.49 | 500 | 1.7451 | 24.369 | 42.66 |
|
| 56 |
+
| 1.732 | 0.98 | 1000 | 1.3896 | 31.9939 | 42.304 |
|
| 57 |
+
| 1.4344 | 1.47 | 1500 | 1.2333 | 36.4344 | 42.384 |
|
| 58 |
+
| 1.3141 | 1.96 | 2000 | 1.1442 | 38.5023 | 41.96 |
|
| 59 |
+
| 1.1877 | 2.45 | 2500 | 1.0936 | 41.3292 | 42.668 |
|
| 60 |
+
| 1.1355 | 2.94 | 3000 | 1.0460 | 43.1357 | 43.22 |
|
| 61 |
+
| 1.0623 | 3.43 | 3500 | 1.0197 | 43.2339 | 42.692 |
|
| 62 |
+
| 1.0353 | 3.93 | 4000 | 1.0010 | 43.8863 | 43.012 |
|
| 63 |
+
| 0.9786 | 4.42 | 4500 | 0.9899 | 44.2478 | 43.012 |
|
| 64 |
+
| 0.9682 | 4.91 | 5000 | 0.9731 | 44.9191 | 42.816 |
|
| 65 |
+
| 0.9184 | 5.4 | 5500 | 0.9690 | 44.908 | 42.496 |
|
| 66 |
+
| 0.9208 | 5.89 | 6000 | 0.9558 | 45.5488 | 42.772 |
|
| 67 |
+
| 0.8854 | 6.38 | 6500 | 0.9561 | 45.7261 | 42.844 |
|
| 68 |
+
| 0.8815 | 6.87 | 7000 | 0.9495 | 45.1231 | 42.38 |
|
| 69 |
+
| 0.8543 | 7.36 | 7500 | 0.9475 | 45.6717 | 42.56 |
|
| 70 |
+
| 0.8462 | 7.85 | 8000 | 0.9442 | 45.9782 | 42.652 |
|
| 71 |
+
| 0.8422 | 8.34 | 8500 | 0.9436 | 45.9353 | 42.628 |
|
| 72 |
+
| 0.8323 | 8.83 | 9000 | 0.9407 | 45.7945 | 42.492 |
|
| 73 |
+
| 0.8218 | 9.32 | 9500 | 0.9405 | 46.0215 | 42.472 |
|
| 74 |
+
| 0.8226 | 9.81 | 10000 | 0.9402 | 45.9717 | 42.476 |
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
### Framework versions
|
| 78 |
+
|
| 79 |
+
- Transformers 4.21.3
|
| 80 |
+
- Pytorch 1.12.1+cu113
|
| 81 |
+
- Datasets 2.4.0
|
| 82 |
+
- Tokenizers 0.12.1
|