metadata
license: mit
base_model: facebook/m2m100_418M
tags:
- generated_from_trainer
model-index:
- name: fine_tuned_modelsNew
results: []
fine_tuned_modelsNew
This model is a fine-tuned version of facebook/m2m100_418M on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0970
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
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 2780
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5467 | 14.29 | 500 | 0.1427 |
| 0.0527 | 28.57 | 1000 | 0.1012 |
| 0.0272 | 42.86 | 1500 | 0.1052 |
| 0.0163 | 57.14 | 2000 | 0.0984 |
| 0.0106 | 71.43 | 2500 | 0.0943 |
Framework versions
- Transformers 4.35.1
- Pytorch 2.1.0+cu118
- Tokenizers 0.14.1