metadata
license: mit
base_model: facebook/m2m100_418M
tags:
- generated_from_trainer
model-index:
- name: output
results: []
output
This model is a fine-tuned version of facebook/m2m100_418M on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1847
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.7809 | 0.11 | 100 | 0.3955 |
0.379 | 0.23 | 200 | 0.2623 |
0.2876 | 0.34 | 300 | 0.2431 |
0.2733 | 0.46 | 400 | 0.2327 |
0.2885 | 0.57 | 500 | 0.2230 |
0.2414 | 0.69 | 600 | 0.2162 |
0.2347 | 0.8 | 700 | 0.2119 |
0.2509 | 0.91 | 800 | 0.2056 |
0.2237 | 1.03 | 900 | 0.2042 |
0.1759 | 1.14 | 1000 | 0.2009 |
0.1471 | 1.26 | 1100 | 0.1984 |
0.1759 | 1.37 | 1200 | 0.1959 |
0.1396 | 1.49 | 1300 | 0.1960 |
0.1656 | 1.6 | 1400 | 0.1936 |
0.163 | 1.71 | 1500 | 0.1898 |
0.1463 | 1.83 | 1600 | 0.1879 |
0.1662 | 1.94 | 1700 | 0.1858 |
0.1297 | 2.06 | 1800 | 0.1889 |
0.0841 | 2.17 | 1900 | 0.1886 |
0.1 | 2.29 | 2000 | 0.1880 |
0.0937 | 2.4 | 2100 | 0.1873 |
0.1023 | 2.51 | 2200 | 0.1862 |
0.0918 | 2.63 | 2300 | 0.1864 |
0.0981 | 2.74 | 2400 | 0.1863 |
0.0993 | 2.86 | 2500 | 0.1841 |
0.1086 | 2.97 | 2600 | 0.1834 |
0.083 | 3.09 | 2700 | 0.1869 |
0.07 | 3.2 | 2800 | 0.1873 |
0.0609 | 3.31 | 2900 | 0.1865 |
0.0637 | 3.43 | 3000 | 0.1861 |
0.078 | 3.54 | 3100 | 0.1862 |
0.0605 | 3.66 | 3200 | 0.1844 |
0.0606 | 3.77 | 3300 | 0.1845 |
0.0579 | 3.89 | 3400 | 0.1849 |
0.0687 | 4.0 | 3500 | 0.1847 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2