n3wtou/mt5-small-finedtuned-4-swahili

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

• Train Loss: 2.4419
• Validation Loss: 2.4809
• 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': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 0.0003, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0003, 'decay_steps': 19900, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 100, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.001}
• training_precision: mixed_float16

Training results

Train Loss	Validation Loss	Epoch
5.6636	2.9818	0
3.7789	2.7822	1
3.3841	2.6840	2
3.1496	2.6238	3
2.9656	2.5816	4
2.8134	2.5522	5
2.6914	2.5315	6
2.5935	2.4980	7
2.5056	2.4764	8
2.4419	2.4809	9

Framework versions

• Transformers 4.30.2
• TensorFlow 2.12.0
• Datasets 2.12.0
• Tokenizers 0.13.3