rlcc-appearance-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: 1.5289
• Accuracy: 0.6829
• F1 Macro: 0.6528
• Precision Macro: 0.6573
• Recall Macro: 0.6497
• Total Tf: [280, 130, 1100, 130]

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: 65
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	Precision Macro	Recall Macro	Total Tf
1.104	1.0	66	1.0942	0.5707	0.4639	0.4375	0.5035	[234, 176, 1054, 176]
0.9421	2.0	132	0.9844	0.6122	0.5511	0.5757	0.6227	[251, 159, 1071, 159]
0.7491	3.0	198	1.0425	0.6268	0.5706	0.6097	0.6372	[257, 153, 1077, 153]
0.6547	4.0	264	1.1165	0.6317	0.5931	0.6007	0.6266	[259, 151, 1079, 151]
0.5895	5.0	330	1.2059	0.6341	0.5947	0.6001	0.6282	[260, 150, 1080, 150]
0.5299	6.0	396	1.2221	0.6488	0.6078	0.6120	0.6197	[266, 144, 1086, 144]
0.4922	7.0	462	1.1999	0.6512	0.6145	0.6147	0.6191	[267, 143, 1087, 143]
0.413	8.0	528	1.3816	0.6439	0.6047	0.6067	0.6174	[264, 146, 1084, 146]
0.4016	9.0	594	1.3556	0.6439	0.6112	0.6085	0.6169	[264, 146, 1084, 146]
0.3321	10.0	660	1.3395	0.6561	0.6233	0.6204	0.6316	[269, 141, 1089, 141]
0.3126	11.0	726	1.4235	0.6683	0.6368	0.6363	0.6445	[274, 136, 1094, 136]
0.2674	12.0	792	1.4367	0.6707	0.6372	0.6358	0.6443	[275, 135, 1095, 135]
0.257	13.0	858	1.3366	0.6902	0.6595	0.6616	0.6585	[283, 127, 1103, 127]
0.2149	14.0	924	1.4133	0.6683	0.6346	0.6391	0.6370	[274, 136, 1094, 136]
0.1996	15.0	990	1.3019	0.6927	0.6610	0.6746	0.6538	[284, 126, 1104, 126]
0.1883	16.0	1056	1.4445	0.6585	0.6254	0.6285	0.6261	[270, 140, 1090, 140]
0.1642	17.0	1122	1.4636	0.6707	0.6400	0.6434	0.6403	[275, 135, 1095, 135]
0.166	18.0	1188	1.4318	0.6927	0.6622	0.6661	0.6592	[284, 126, 1104, 126]
0.1452	19.0	1254	1.4605	0.6927	0.6626	0.6662	0.6605	[284, 126, 1104, 126]
0.1569	20.0	1320	1.4895	0.6805	0.6495	0.6543	0.6463	[279, 131, 1099, 131]
0.1372	21.0	1386	1.5280	0.6780	0.6488	0.6504	0.6478	[278, 132, 1098, 132]
0.1436	22.0	1452	1.5309	0.6878	0.6584	0.6608	0.6569	[282, 128, 1102, 128]
0.1251	23.0	1518	1.5289	0.6829	0.6528	0.6573	0.6497	[280, 130, 1100, 130]

Framework versions

• Transformers 4.48.3
• Pytorch 2.1.0+cu118
• Tokenizers 0.21.0