metadata
license: mit
base_model: facebook/m2m100_418M
tags:
- generated_from_trainer
model-index:
- name: genre-m2m100_418M
results: []
genre-m2m100_418M
This model is a fine-tuned version of facebook/m2m100_418M on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0132
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: 0.0005
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1500
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.1257 | 0.2 | 1500 | 0.0960 |
| 0.1138 | 0.4 | 3000 | 0.0864 |
| 0.0745 | 0.6 | 4500 | 0.0655 |
| 0.0695 | 0.8 | 6000 | 0.0464 |
| 0.0926 | 1.0 | 7500 | 0.0444 |
| 0.0348 | 1.2 | 9000 | 0.0348 |
| 0.0626 | 1.4 | 10500 | 0.0342 |
| 0.0641 | 1.6 | 12000 | 0.0305 |
| 0.0249 | 1.8 | 13500 | 0.0285 |
| 0.0432 | 2.0 | 15000 | 0.0245 |
| 0.0171 | 2.2 | 16500 | 0.0250 |
| 0.0575 | 2.4 | 18000 | 0.0233 |
| 0.0221 | 2.6 | 19500 | 0.0213 |
| 0.025 | 2.8 | 21000 | 0.0202 |
| 0.0136 | 3.0 | 22500 | 0.0194 |
| 0.0222 | 3.2 | 24000 | 0.0184 |
| 0.0581 | 3.4 | 25500 | 0.0174 |
| 0.0132 | 3.6 | 27000 | 0.0168 |
| 0.0087 | 3.8 | 28500 | 0.0159 |
| 0.0164 | 4.0 | 30000 | 0.0152 |
| 0.0088 | 4.2 | 31500 | 0.0149 |
| 0.0217 | 4.4 | 33000 | 0.0144 |
| 0.0091 | 4.6 | 34500 | 0.0138 |
| 0.0099 | 4.8 | 36000 | 0.0134 |
| 0.0078 | 5.0 | 37500 | 0.0132 |
Framework versions
- Transformers 4.37.1
- Pytorch 2.1.2
- Datasets 2.16.1
- Tokenizers 0.15.1