metadata
tags:
- generated_from_trainer
datasets:
- xtreme_s
metrics:
- bleu
model-index:
- name: ''
results: []
This model was trained from scratch on the xtreme_s dataset. It achieves the following results on the evaluation set:
- Loss: 1.6425
- Bleu: 0.0000
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: 0.00012335092351490598
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu |
---|---|---|---|---|
2.8303 | 0.15 | 500 | 4.9238 | 0.0 |
2.4062 | 0.31 | 1000 | 4.4017 | 0.0 |
1.9171 | 0.46 | 1500 | 3.6431 | 0.0000 |
1.4558 | 0.62 | 2000 | 2.8292 | 0.0000 |
1.2393 | 0.77 | 2500 | 2.3704 | 0.0000 |
1.3315 | 0.93 | 3000 | 2.3101 | 0.0000 |
1.8476 | 1.08 | 3500 | 1.9936 | 0.0000 |
1.683 | 1.23 | 4000 | 1.9308 | 0.0000 |
1.8298 | 1.39 | 4500 | 1.8817 | 0.0000 |
1.5955 | 1.54 | 5000 | 1.8171 | 0.0000 |
1.6288 | 1.7 | 5500 | 1.7821 | 0.0000 |
1.4107 | 1.85 | 6000 | 1.7170 | 0.0000 |
1.0363 | 2.01 | 6500 | 1.7419 | 0.0000 |
0.9667 | 2.16 | 7000 | 1.7309 | 0.0000 |
0.9147 | 2.31 | 7500 | 1.7244 | 0.0000 |
1.1975 | 2.47 | 8000 | 1.6716 | 0.0000 |
0.9071 | 2.62 | 8500 | 1.6886 | 0.0000 |
0.9735 | 2.78 | 9000 | 1.6609 | 0.0000 |
0.908 | 2.93 | 9500 | 1.6425 | 0.0000 |
Framework versions
- Transformers 4.19.0.dev0
- Pytorch 1.10.2+cu113
- Datasets 2.1.1.dev0
- Tokenizers 0.11.0