rlcc-taste-upsample_replacement-absa-min

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

• Loss: 2.0451
• Accuracy: 0.6463
• F1 Macro: 0.6925
• Precision Macro: 0.6941
• Recall Macro: 0.6925
• 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
1.0951	1.0	91	1.1126	0.4366	0.4534	0.6658	0.5044	[179, 231, 999, 231]
0.9371	2.0	182	0.9998	0.5756	0.5422	0.5071	0.5904	[236, 174, 1056, 174]
0.7416	3.0	273	1.0900	0.5488	0.5805	0.5892	0.6078	[225, 185, 1045, 185]
0.6429	4.0	364	1.1685	0.5634	0.5517	0.4949	0.6483	[231, 179, 1051, 179]
0.5992	5.0	455	1.0657	0.6366	0.6595	0.6879	0.6604	[261, 149, 1081, 149]
0.5504	6.0	546	1.1434	0.6463	0.6839	0.7002	0.6790	[265, 145, 1085, 145]
0.4743	7.0	637	1.1307	0.6415	0.6852	0.6817	0.6951	[263, 147, 1083, 147]
0.3508	8.0	728	1.2377	0.6244	0.6679	0.6637	0.6810	[256, 154, 1076, 154]
0.3027	9.0	819	1.3096	0.6366	0.6821	0.6824	0.6818	[261, 149, 1081, 149]
0.2456	10.0	910	1.3739	0.6537	0.6986	0.6975	0.7006	[268, 142, 1088, 142]
0.2111	11.0	1001	1.4215	0.6341	0.6809	0.6801	0.6821	[260, 150, 1080, 150]
0.1452	12.0	1092	1.5324	0.6366	0.6822	0.6832	0.6814	[261, 149, 1081, 149]
0.1426	13.0	1183	1.6104	0.6415	0.6871	0.6918	0.6845	[263, 147, 1083, 147]
0.142	14.0	1274	1.6417	0.6390	0.6852	0.6846	0.6867	[262, 148, 1082, 148]
0.1004	15.0	1365	1.7112	0.6439	0.6904	0.6945	0.6896	[264, 146, 1084, 146]
0.1292	16.0	1456	1.7041	0.6463	0.6932	0.6974	0.6921	[265, 145, 1085, 145]
0.0998	17.0	1547	1.7698	0.6512	0.6956	0.6951	0.6964	[267, 143, 1087, 143]
0.073	18.0	1638	1.8860	0.6488	0.6948	0.7013	0.6919	[266, 144, 1086, 144]
0.0736	19.0	1729	1.9039	0.6390	0.6859	0.6875	0.6854	[262, 148, 1082, 148]
0.0548	20.0	1820	2.0032	0.6366	0.6844	0.6855	0.6864	[261, 149, 1081, 149]
0.0554	21.0	1911	2.0158	0.6317	0.6804	0.6843	0.6807	[259, 151, 1079, 151]
0.0583	22.0	2002	2.0387	0.6439	0.6908	0.6949	0.6903	[264, 146, 1084, 146]
0.0427	23.0	2093	2.0383	0.6512	0.6965	0.6992	0.6962	[267, 143, 1087, 143]
0.0453	24.0	2184	2.0345	0.6463	0.6925	0.6941	0.6925	[265, 145, 1085, 145]
0.0466	25.0	2275	2.0451	0.6463	0.6925	0.6941	0.6925	[265, 145, 1085, 145]

Framework versions

• Transformers 4.48.3
• Pytorch 2.1.0+cu118
• Tokenizers 0.21.0