rlcc-taste-upsample_replacement-absa-avg

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.4031
• Accuracy: 0.6463
• F1 Macro: 0.6925
• Precision Macro: 0.6953
• Recall Macro: 0.6916
• Total Tf: [265, 145, 1085, 145]

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use 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_steps: 90
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	Precision Macro	Recall Macro	Total Tf
1.09	1.0	91	1.0912	0.4341	0.4600	0.4475	0.5420	[178, 232, 998, 232]
0.8861	2.0	182	1.0079	0.5634	0.5653	0.5666	0.5870	[231, 179, 1051, 179]
0.7236	3.0	273	1.1235	0.5634	0.5535	0.6075	0.5892	[231, 179, 1051, 179]
0.6335	4.0	364	1.1511	0.5976	0.6183	0.6662	0.6729	[245, 165, 1065, 165]
0.5932	5.0	455	1.1886	0.5805	0.5472	0.5175	0.5931	[238, 172, 1058, 172]
0.5596	6.0	546	1.2477	0.6683	0.7069	0.7132	0.7039	[274, 136, 1094, 136]
0.5334	7.0	637	1.2524	0.6366	0.6810	0.6773	0.6922	[261, 149, 1081, 149]
0.4517	8.0	728	1.2644	0.6488	0.6942	0.6937	0.7036	[266, 144, 1086, 144]
0.3672	9.0	819	1.2428	0.6488	0.6940	0.6937	0.6944	[266, 144, 1086, 144]
0.2938	10.0	910	1.3009	0.6488	0.6925	0.6895	0.7012	[266, 144, 1086, 144]
0.2446	11.0	1001	1.4031	0.6463	0.6925	0.6953	0.6916	[265, 145, 1085, 145]

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0