ayshi/undersampling_distil

This model is a fine-tuned version of distilbert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.7701
• Validation Loss: 1.0288
• Train Accuracy: 0.5824
• Epoch: 8

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': None, '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': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 130, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
• training_precision: float32

Training results

Train Loss	Validation Loss	Train Accuracy	Epoch
1.7867	1.7551	0.4176	0
1.7228	1.6637	0.4835	1
1.5961	1.4869	0.5934	2
1.4148	1.3503	0.5934	3
1.2203	1.2274	0.6264	4
1.0720	1.1445	0.5934	5
0.9397	1.0827	0.5824	6
0.8296	1.0548	0.6044	7
0.7701	1.0288	0.5824	8

Framework versions

• Transformers 4.34.0
• TensorFlow 2.13.0
• Datasets 2.14.5
• Tokenizers 0.14.1