<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->
Prgrg/ja-en-dataset-v3.0-subset-v2.0
This model is a fine-tuned version of Prgrg/ja-en-dataset-v3.0-subset-v1.0 on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 1.1008
- Validation Loss: 1.1490
- Epoch: 7
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:
- optimizer: {'name': 'Adam', 'weight_decay': 0.0, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'LinearWarmupCosineSchedule', 'config': {'initial_learning_rate': 0.0005, 'num_warmup_steps': 37575, 'num_training_steps': 375752}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
- training_precision: mixed_float16
Training results
Train Loss | Validation Loss | Epoch |
---|---|---|
1.6417 | 1.6251 | 0 |
1.6668 | 1.5313 | 1 |
1.5536 | 1.4377 | 2 |
1.4515 | 1.3447 | 3 |
1.3305 | 1.2609 | 4 |
1.2264 | 1.1908 | 5 |
1.1451 | 1.1540 | 6 |
1.1008 | 1.1490 | 7 |
Framework versions
- Transformers 4.27.3
- TensorFlow 2.11.0
- Datasets 2.10.1
- Tokenizers 0.13.2