rlcc-aroma-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: 1.5902
• Accuracy: 0.7707
• F1 Macro: 0.7015
• Precision Macro: 0.7008
• Recall Macro: 0.7091
• Total Tf: [316, 94, 1136, 94]

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: 51
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	Precision Macro	Recall Macro	Total Tf
1.1003	1.0	52	1.0973	0.6073	0.4395	0.4182	0.4880	[249, 161, 1069, 161]
0.9934	2.0	104	1.1124	0.6707	0.5499	0.6186	0.5663	[275, 135, 1095, 135]
0.8138	3.0	156	1.1006	0.7073	0.6159	0.6228	0.6121	[290, 120, 1110, 120]
0.6792	4.0	208	1.2207	0.6927	0.5975	0.6140	0.6475	[284, 126, 1104, 126]
0.6213	5.0	260	1.2312	0.7049	0.6159	0.6401	0.6586	[289, 121, 1109, 121]
0.5836	6.0	312	1.3172	0.7024	0.6092	0.6290	0.6611	[288, 122, 1108, 122]
0.4774	7.0	364	1.3690	0.7146	0.6333	0.6364	0.6577	[293, 117, 1113, 117]
0.4054	8.0	416	1.3341	0.7341	0.6562	0.6671	0.6826	[301, 109, 1121, 109]
0.348	9.0	468	1.3494	0.7610	0.6872	0.6883	0.7002	[312, 98, 1132, 98]
0.3111	10.0	520	1.3309	0.7756	0.7080	0.7062	0.7178	[318, 92, 1138, 92]
0.2831	11.0	572	1.4596	0.7439	0.6692	0.6790	0.6922	[305, 105, 1125, 105]
0.2606	12.0	624	1.4924	0.7610	0.6884	0.6872	0.6978	[312, 98, 1132, 98]
0.2118	13.0	676	1.5774	0.7463	0.6739	0.6763	0.6941	[306, 104, 1126, 104]
0.1923	14.0	728	1.5830	0.7683	0.6989	0.6984	0.7096	[315, 95, 1135, 95]
0.1543	15.0	780	1.5902	0.7707	0.7015	0.7008	0.7091	[316, 94, 1136, 94]

Framework versions

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