metadata
base_model: google/mt5-small
datasets:
- syubraj/roman2nepali-transliteration
language:
- ne
- en
library_name: transformers
license: apache-2.0
metrics:
- bleu
tags:
- generated_from_trainer
model-index:
- name: romaneng2nep_v2
results: []
romaneng2nep_v2
This model is a fine-tuned version of google/mt5-small on an syubraj/roman2nepali-transliteration. It achieves the following results on the evaluation set:
- Loss: 2.7225
- Gen Len: 5.2131
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: 2e-05
- train_batch_size: 24
- eval_batch_size: 24
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Gen Len |
|---|---|---|---|---|
| 15.2574 | 0.0626 | 1000 | 5.7371 | 2.0266 |
| 6.1453 | 0.1251 | 2000 | 4.5094 | 4.5514 |
| 5.3182 | 0.1877 | 3000 | 4.0351 | 4.7656 |
| 4.9218 | 0.2503 | 4000 | 3.6947 | 4.9841 |
| 4.6397 | 0.3128 | 5000 | 3.4644 | 5.1216 |
| 4.433 | 0.3754 | 6000 | 3.3009 | 5.2036 |
| 4.2494 | 0.4380 | 7000 | 3.1525 | 5.1748 |
| 4.1467 | 0.5005 | 8000 | 3.0482 | 5.232 |
| 4.0272 | 0.5631 | 9000 | 2.9592 | 5.253 |
| 3.9598 | 0.6257 | 10000 | 2.8917 | 5.1893 |
| 3.9116 | 0.6882 | 11000 | 2.8292 | 5.2252 |
| 3.8435 | 0.7508 | 12000 | 2.7871 | 5.2148 |
| 3.8047 | 0.8134 | 13000 | 2.7574 | 5.2123 |
| 3.7818 | 0.8759 | 14000 | 2.7338 | 5.2409 |
| 3.7764 | 0.9385 | 15000 | 2.7225 | 5.2131 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0