model

This model is a fine-tuned version of Tsubasaz/clinical-pubmed-bert-base-512 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3511
• Precision: 0.6103
• Recall: 0.5640

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: 3e-06
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall
No log	1.0	128	0.4393	0.0	0.0
No log	2.0	256	0.3958	0.5714	0.1706
No log	3.0	384	0.3785	0.5690	0.3128
0.4046	4.0	512	0.3676	0.5789	0.5213
0.4046	5.0	640	0.3606	0.6532	0.3839
0.4046	6.0	768	0.3597	0.6549	0.4408
0.4046	7.0	896	0.3584	0.6376	0.4502
0.3046	8.0	1024	0.3518	0.6310	0.5024
0.3046	9.0	1152	0.3511	0.6133	0.5261
0.3046	10.0	1280	0.3511	0.6103	0.5640

Framework versions

• Transformers 4.35.2
• Pytorch 2.0.0
• Datasets 2.15.0
• Tokenizers 0.15.0