metadata
library_name: transformers
base_model: danasone/bart-small-ru-en
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: bart_eng_hin_mt
results: []
bart_eng_hin_mt
This model is a fine-tuned version of danasone/bart-small-ru-en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3398
- Bleu: 10.015
- Gen Len: 123.5141
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: 0.0008
- train_batch_size: 300
- eval_batch_size: 20
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 2400
- total_eval_batch_size: 160
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 15.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 0.5522 | 1.0 | 689 | 0.5343 | 3.4797 | 123.3969 |
| 0.3988 | 2.0 | 1378 | 0.4020 | 7.5644 | 123.3578 |
| 0.3496 | 3.0 | 2067 | 0.3601 | 9.3506 | 123.4641 |
| 0.3191 | 4.0 | 2756 | 0.3398 | 10.015 | 123.5141 |
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1