CeLLaTe3.0_Base_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: 0.1976
• Precision: 0.8264
• Recall: 0.8264
• F1: 0.8264
• Accuracy: 0.9674

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: 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
1.3073	1.0417	100	0.5057	0.0	0.0	0.0	0.8644
0.3496	2.0833	200	0.2117	0.5943	0.5686	0.5812	0.9441
0.1555	3.125	300	0.1366	0.7816	0.8117	0.7963	0.9650
0.0919	4.1667	400	0.1622	0.8154	0.7708	0.7925	0.9631
0.0691	5.2083	500	0.1574	0.7502	0.8023	0.7754	0.9604
0.0497	6.25	600	0.1741	0.8023	0.7921	0.7972	0.9624
0.0373	7.2917	700	0.1761	0.7691	0.8045	0.7864	0.9621
0.03	8.3333	800	0.1903	0.7768	0.8198	0.7977	0.9617
0.0268	9.375	900	0.1791	0.7807	0.8460	0.8120	0.9643
0.0205	10.4167	1000	0.1952	0.7758	0.8132	0.7941	0.9626
0.0142	11.4583	1100	0.1846	0.8152	0.8318	0.8234	0.9672
0.0119	12.5	1200	0.1938	0.8319	0.8057	0.8186	0.9678
0.0105	13.5417	1300	0.2071	0.8196	0.8039	0.8117	0.9658
0.0093	14.5833	1400	0.1989	0.8264	0.8264	0.8264	0.9674
0.0073	15.625	1500	0.2105	0.8337	0.8087	0.8210	0.9669
0.0074	16.6667	1600	0.2177	0.8206	0.7954	0.8078	0.9656
0.006	17.7083	1700	0.2103	0.8273	0.8183	0.8227	0.9679

Framework versions

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