metadata
license: mit
base_model: facebook/m2m100_418M
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: my_awesome_opus_books_model
results: []
my_awesome_opus_books_model
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: 0.3013
- Bleu: 18.2865
- Gen Len: 19.1518
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
---|---|---|---|---|---|
0.3015 | 1.0 | 5000 | 0.3255 | 16.0387 | 19.013 |
0.3084 | 2.0 | 10000 | 0.3101 | 17.228 | 19.1362 |
0.2704 | 3.0 | 15000 | 0.3040 | 17.8346 | 18.9994 |
0.2421 | 4.0 | 20000 | 0.3016 | 18.1199 | 19.1469 |
0.2259 | 5.0 | 25000 | 0.3013 | 18.2865 | 19.1518 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1