rlcc-palate-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: 1.9359
• Accuracy: 0.8
• F1 Macro: 0.5653
• Precision Macro: 0.5691
• Recall Macro: 0.6086
• Total Tf: [328, 82, 1148, 82]

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: 36
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	Precision Macro	Recall Macro	Total Tf
1.1087	1.0	37	1.0814	0.8	0.4854	0.4886	0.4866	[328, 82, 1148, 82]
1.0809	2.0	74	1.0422	0.8366	0.4819	0.4582	0.5129	[343, 67, 1163, 67]
0.9443	3.0	111	1.0914	0.8195	0.5403	0.5491	0.5776	[336, 74, 1156, 74]
0.7914	4.0	148	1.1077	0.8171	0.5736	0.5725	0.6223	[335, 75, 1155, 75]
0.68	5.0	185	1.2576	0.7585	0.4613	0.4035	0.6144	[311, 99, 1131, 99]
0.5892	6.0	222	1.2538	0.8049	0.5494	0.5447	0.5680	[330, 80, 1150, 80]
0.564	7.0	259	1.2608	0.8098	0.5699	0.5680	0.5977	[332, 78, 1152, 78]
0.5395	8.0	296	1.3882	0.7585	0.4659	0.4086	0.6166	[311, 99, 1131, 99]
0.5745	9.0	333	1.2718	0.8098	0.5878	0.5968	0.6340	[332, 78, 1152, 78]
0.5492	10.0	370	1.3772	0.8122	0.5641	0.5579	0.5838	[333, 77, 1153, 77]
0.4613	11.0	407	1.5313	0.8293	0.5471	0.5552	0.5915	[340, 70, 1160, 70]
0.5209	12.0	444	1.5491	0.8244	0.5125	0.4843	0.5547	[338, 72, 1158, 72]
0.4865	13.0	481	1.6926	0.8073	0.5809	0.5987	0.6386	[331, 79, 1151, 79]
0.4758	14.0	518	1.6802	0.8195	0.6041	0.6071	0.6511	[336, 74, 1156, 74]
0.3792	15.0	555	1.7840	0.8098	0.5873	0.6012	0.6491	[332, 78, 1152, 78]
0.3555	16.0	592	1.8224	0.8122	0.5915	0.5943	0.6423	[333, 77, 1153, 77]
0.2957	17.0	629	1.8715	0.8098	0.5790	0.5777	0.6106	[332, 78, 1152, 78]
0.2693	18.0	666	1.9544	0.8	0.5683	0.5704	0.6157	[328, 82, 1148, 82]
0.248	19.0	703	1.9359	0.8	0.5653	0.5691	0.6086	[328, 82, 1148, 82]

Framework versions

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