rlcc-aroma-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: 1.3566
• Accuracy: 0.7512
• F1 Macro: 0.6770
• Precision Macro: 0.6751
• Recall Macro: 0.6814
• Total Tf: [308, 102, 1128, 102]

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
1.1037	1.0	52	1.0910	0.5805	0.4231	0.5135	0.4909	[238, 172, 1058, 172]
1.0055	2.0	104	1.0163	0.6878	0.5727	0.6418	0.6599	[282, 128, 1102, 128]
0.8528	3.0	156	0.9953	0.6902	0.5688	0.6933	0.6720	[283, 127, 1103, 127]
0.6841	4.0	208	1.0935	0.6634	0.5564	0.5645	0.6144	[272, 138, 1092, 138]
0.644	5.0	260	1.0736	0.7073	0.6202	0.6266	0.6529	[290, 120, 1110, 120]
0.5771	6.0	312	1.1553	0.7146	0.6307	0.6379	0.6619	[293, 117, 1113, 117]
0.4985	7.0	364	1.0993	0.7341	0.6572	0.6578	0.6712	[301, 109, 1121, 109]
0.5024	8.0	416	1.1851	0.7390	0.6641	0.6629	0.6810	[303, 107, 1123, 107]
0.442	9.0	468	1.2099	0.7293	0.6498	0.6490	0.6631	[299, 111, 1119, 111]
0.3703	10.0	520	1.2218	0.7341	0.6551	0.6563	0.6646	[301, 109, 1121, 109]
0.3419	11.0	572	1.1793	0.7463	0.6706	0.6690	0.6764	[306, 104, 1126, 104]
0.2909	12.0	624	1.1914	0.7488	0.6738	0.6724	0.6818	[307, 103, 1127, 103]
0.2524	13.0	676	1.2408	0.7366	0.6577	0.6558	0.6697	[302, 108, 1122, 108]
0.2481	14.0	728	1.2687	0.7366	0.6600	0.6580	0.6676	[302, 108, 1122, 108]
0.2129	15.0	780	1.1961	0.7488	0.6734	0.6731	0.6749	[307, 103, 1127, 103]
0.2016	16.0	832	1.2119	0.7512	0.6741	0.6754	0.6775	[308, 102, 1128, 102]
0.1942	17.0	884	1.2834	0.7341	0.6530	0.6510	0.6597	[301, 109, 1121, 109]
0.1913	18.0	936	1.2919	0.7439	0.6654	0.6630	0.6707	[305, 105, 1125, 105]
0.173	19.0	988	1.2663	0.7512	0.6728	0.6762	0.6734	[308, 102, 1128, 102]
0.1514	20.0	1040	1.3107	0.7415	0.6602	0.6625	0.6635	[304, 106, 1124, 106]
0.1643	21.0	1092	1.3403	0.7390	0.6586	0.6582	0.6630	[303, 107, 1123, 107]
0.1567	22.0	1144	1.3275	0.7439	0.6641	0.6639	0.6675	[305, 105, 1125, 105]
0.1419	23.0	1196	1.3299	0.7488	0.6721	0.6709	0.6756	[307, 103, 1127, 103]
0.1608	24.0	1248	1.3551	0.7512	0.6770	0.6755	0.6814	[308, 102, 1128, 102]
0.1514	25.0	1300	1.3566	0.7512	0.6770	0.6751	0.6814	[308, 102, 1128, 102]

Framework versions

• Transformers 4.48.3
• Pytorch 2.1.0+cu118
• Tokenizers 0.21.0