mt5_model.h5

This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 6.8154
• Train Accuracy: 0.0937
• Validation Loss: 1.5181
• Validation Accuracy: 0.4921
• Epoch: 9

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': 'AdamWeightDecay', 'learning_rate': 3e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.99, 'epsilon': 0.1, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Train Loss	Train Accuracy	Validation Loss	Validation Accuracy	Epoch
20.8851	0.0146	7.8326	0.1620	0
16.2075	0.0207	6.0446	0.2008	1
13.9846	0.0253	5.3711	0.2087	2
12.4570	0.0299	4.1290	0.2562	3
10.9899	0.0390	3.0786	0.3184	4
9.9004	0.0466	2.3305	0.3880	5
9.0396	0.0548	1.9971	0.4272	6
8.1132	0.0670	1.7134	0.4631	7
7.3820	0.0827	1.6130	0.4775	8
6.8154	0.0937	1.5181	0.4921	9

Framework versions

• Transformers 4.47.0
• TensorFlow 2.17.1
• Datasets 3.2.0
• Tokenizers 0.21.0