CeLLaTe_V3.3_Reinit_llrd_lr_3e5

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: 0.0776
• Precision: 0.7856
• Recall: 0.6584
• F1: 0.7164
• Accuracy: 0.9799

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: 16
• 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: 0.06
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
1.9644	0.0476	100	1.6693	0.0054	0.0279	0.0091	0.7567
0.8974	0.0951	200	0.2911	0.0	0.0	0.0	0.9453
0.296	0.1427	300	0.1823	0.0	0.0	0.0	0.9453
0.1878	0.1902	400	0.1379	0.2524	0.2375	0.2447	0.9504
0.1571	0.2378	500	0.1167	0.3177	0.3069	0.3122	0.9613
0.128	0.2853	600	0.1117	0.3315	0.3799	0.3541	0.9612
0.1211	0.3329	700	0.0989	0.4715	0.4519	0.4615	0.9717
0.1002	0.3804	800	0.0862	0.5360	0.6222	0.5759	0.9741
0.0753	0.4280	900	0.0751	0.6307	0.7152	0.6703	0.9766
0.0683	0.4755	1000	0.0775	0.7856	0.6584	0.7164	0.9799
0.0684	0.5231	1100	0.0751	0.6488	0.7262	0.6853	0.9771
0.0602	0.5706	1200	0.0681	0.6788	0.7373	0.7068	0.9784
0.0512	0.6182	1300	0.0789	0.6104	0.7583	0.6764	0.9741

Framework versions

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