CeLLaTe3.0_LLRD_no_vague_adapted_pubmed

This model is a fine-tuned version of Mardiyyah/cellate2.0-tapt_base-LR_5e-05 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2007
• Precision: 0.8141
• Recall: 0.8144
• F1: 0.8143
• Accuracy: 0.9619

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-05
• train_batch_size: 32
• eval_batch_size: 16
• seed: 3407
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
1.1784	1.0417	100	0.3883	0.2271	0.1441	0.1763	0.8872
0.2627	2.0833	200	0.1683	0.6452	0.7211	0.6811	0.9538
0.1123	3.125	300	0.1480	0.8037	0.7503	0.7761	0.9613
0.0693	4.1667	400	0.1629	0.7995	0.8060	0.8027	0.9637
0.0477	5.2083	500	0.1577	0.8110	0.8171	0.8140	0.9652
0.0323	6.25	600	0.2010	0.8141	0.8144	0.8143	0.9619
0.0245	7.2917	700	0.1971	0.7691	0.7584	0.7637	0.9600
0.0185	8.3333	800	0.2117	0.8028	0.7837	0.7931	0.9623
0.0157	9.375	900	0.2137	0.7738	0.8397	0.8054	0.9611

Framework versions

• Transformers 4.48.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.2
• Tokenizers 0.21.0