rlcc-aroma-upsample_replacement-absa-max

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

• Loss: 1.8514
• Accuracy: 0.7537
• F1 Macro: 0.6792
• Precision Macro: 0.6786
• Recall Macro: 0.6826
• Total Tf: [309, 101, 1129, 101]

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
1.1002	1.0	52	1.1086	0.5902	0.4404	0.4878	0.5264	[242, 168, 1062, 168]
1.0065	2.0	104	1.0822	0.6756	0.5729	0.5874	0.5768	[277, 133, 1097, 133]
0.8195	3.0	156	1.0753	0.6927	0.5628	0.5938	0.5755	[284, 126, 1104, 126]
0.6647	4.0	208	1.1600	0.7049	0.6180	0.6253	0.6558	[289, 121, 1109, 121]
0.6082	5.0	260	1.2364	0.7073	0.6175	0.6343	0.6636	[290, 120, 1110, 120]
0.5552	6.0	312	1.3171	0.7049	0.6131	0.6370	0.6659	[289, 121, 1109, 121]
0.4456	7.0	364	1.3888	0.7171	0.6325	0.6393	0.6523	[294, 116, 1114, 116]
0.3962	8.0	416	1.3679	0.7366	0.6596	0.6673	0.6849	[302, 108, 1122, 108]
0.3641	9.0	468	1.4010	0.7366	0.6588	0.6672	0.6787	[302, 108, 1122, 108]
0.3076	10.0	520	1.4285	0.7415	0.6671	0.6676	0.6861	[304, 106, 1124, 106]
0.2978	11.0	572	1.4946	0.7317	0.6536	0.6689	0.6803	[300, 110, 1120, 110]
0.2732	12.0	624	1.5010	0.7585	0.6868	0.6838	0.6986	[311, 99, 1131, 99]
0.2439	13.0	676	1.5810	0.7415	0.6685	0.6672	0.6895	[304, 106, 1124, 106]
0.2007	14.0	728	1.5708	0.7634	0.6926	0.6916	0.6949	[313, 97, 1133, 97]
0.1443	15.0	780	1.6609	0.7537	0.6795	0.6782	0.6923	[309, 101, 1129, 101]
0.1362	16.0	832	1.7380	0.7512	0.6754	0.6744	0.6852	[308, 102, 1128, 102]
0.1375	17.0	884	1.8404	0.7537	0.6818	0.6791	0.6944	[309, 101, 1129, 101]
0.1237	18.0	936	1.8327	0.7439	0.6693	0.6683	0.6835	[305, 105, 1125, 105]
0.0975	19.0	988	1.8514	0.7537	0.6792	0.6786	0.6826	[309, 101, 1129, 101]

Framework versions

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