Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- squad
|
| 7 |
+
model-index:
|
| 8 |
+
- name: greek-nllb-5ep-384
|
| 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 |
+
# greek-nllb-4ep-384
|
| 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 squad dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.2837
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 0.0001
|
| 39 |
+
- train_batch_size: 2
|
| 40 |
+
- eval_batch_size: 2
|
| 41 |
+
- seed: 42
|
| 42 |
+
- gradient_accumulation_steps: 16
|
| 43 |
+
- total_train_batch_size: 32
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: linear
|
| 46 |
+
- num_epochs: 5
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
+
| 2.2156 | 0.17 | 100 | 1.5542 |
|
| 53 |
+
| 1.7147 | 0.34 | 200 | 1.4823 |
|
| 54 |
+
| 1.5438 | 0.51 | 300 | 1.4377 |
|
| 55 |
+
| 1.5006 | 0.67 | 400 | 1.3997 |
|
| 56 |
+
| 1.4573 | 0.84 | 500 | 1.3753 |
|
| 57 |
+
| 1.4484 | 1.01 | 600 | 1.3596 |
|
| 58 |
+
| 1.3177 | 1.18 | 700 | 1.3491 |
|
| 59 |
+
| 1.2981 | 1.35 | 800 | 1.3432 |
|
| 60 |
+
| 1.2977 | 1.52 | 900 | 1.3346 |
|
| 61 |
+
| 1.2898 | 1.69 | 1000 | 1.3183 |
|
| 62 |
+
| 1.2964 | 1.86 | 1100 | 1.3114 |
|
| 63 |
+
| 1.2480 | 2.03 | 1200 | 1.3176 |
|
| 64 |
+
| 1.1766 | 2.2 | 1300 | 1.3019 |
|
| 65 |
+
| 1.1779 | 2.36 | 1400 | 1.3076 |
|
| 66 |
+
| 1.1770 | 2.53 | 1500 | 1.2968 |
|
| 67 |
+
| 1.1793 | 2.7 | 1600 | 1.2967 |
|
| 68 |
+
| 1.1721 | 2.87 | 1700 | 1.2879 |
|
| 69 |
+
| 1.1593 | 3.04 | 1800 | 1.2964 |
|
| 70 |
+
| 1.0904 | 3.21 | 1900 | 1.2948 |
|
| 71 |
+
| 1.1110 | 3.38 | 2000 | 1.2885 |
|
| 72 |
+
| 1.0947 | 3.55 | 2100 | 1.2885 |
|
| 73 |
+
| 1.0936 | 3.72 | 2200 | 1.2872 |
|
| 74 |
+
| 1.0875 | 4.05 | 2300 | 1.2823 |
|
| 75 |
+
| 1.0875 | 4.22 | 2400 | 1.2904 |
|
| 76 |
+
| 1.0875 | 4.39 | 2500 | 1.2875 |
|
| 77 |
+
| 1.0875 | 4.56 | 2600 | 1.2888 |
|
| 78 |
+
| 1.0875 | 4.73 | 2700 | 1.2865 |
|
| 79 |
+
| 1.0875 | 4.89 | 2800 | 1.2861 |
|
| 80 |
+
| 1.0875 | 5.00 | 2900 | 1.2837 |
|
| 81 |
+
|
| 82 |
+
|
| 83 |
+
### Framework versions
|
| 84 |
+
|
| 85 |
+
- Transformers 4.27.0.dev0
|
| 86 |
+
- Pytorch 1.13.0
|
| 87 |
+
- Datasets 2.1.0
|
| 88 |
+
- Tokenizers 0.13.2
|