nlp_te_class_biobert

This model is a fine-tuned version of AmedeoBonatti/nlp_te_mlm_biobert on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7355
• Accuracy: 0.8884

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	57	0.3001	0.875
No log	2.0	114	0.3770	0.8571
No log	3.0	171	0.8971	0.7812
No log	4.0	228	0.5314	0.875
No log	5.0	285	0.6840	0.8393
No log	6.0	342	0.5531	0.8795
No log	7.0	399	0.6778	0.875
No log	8.0	456	0.7355	0.8884
0.1178	9.0	513	0.7552	0.8839
0.1178	10.0	570	0.7898	0.8839
0.1178	11.0	627	0.8090	0.8884
0.1178	12.0	684	0.8381	0.8884
0.1178	13.0	741	0.8491	0.8884

Framework versions

• Transformers 4.41.2
• Pytorch 2.2.1
• Datasets 2.19.2
• Tokenizers 0.19.1