UNetOscillatoryNeuron

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

• Loss: 0.5575
• Dice: 0.6904
• Iou: 0.5364
• Precision: 0.9958
• Recall: 0.5374

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: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Dice	Iou	Precision	Recall
0.5468	1.0	27	0.6598	0.4436	0.2873	0.2982	0.9041
0.324	2.0	54	0.5230	0.5408	0.3726	0.6035	0.5030
0.2251	3.0	81	0.6158	0.7171	0.5703	0.9217	0.6041
0.4844	4.0	108	0.6834	0.7842	0.6496	0.8024	0.7784
0.5407	5.0	135	0.5531	0.6705	0.5126	0.9877	0.5157
0.5445	6.0	162	0.6066	0.7390	0.5945	0.9493	0.6140
0.543	7.0	189	0.5387	0.6575	0.4977	0.9919	0.4999
0.5402	8.0	216	0.5975	0.7322	0.5865	0.9599	0.6016
0.5479	9.0	243	0.5710	0.7032	0.5515	0.9845	0.5563
0.5428	10.0	270	0.5723	0.7053	0.5527	0.9849	0.5570
0.5412	11.0	297	0.5991	0.7363	0.5911	0.9633	0.6044
0.5448	12.0	324	0.5792	0.7143	0.5638	0.9817	0.5694
0.5439	13.0	351	0.5896	0.7273	0.5801	0.9764	0.5877
0.5422	14.0	378	0.5863	0.7235	0.5758	0.9789	0.5827
0.5433	15.0	405	0.5615	0.6933	0.5389	0.9925	0.5408
0.5429	16.0	432	0.5913	0.7298	0.5835	0.9752	0.5917
0.5426	17.0	459	0.5643	0.6976	0.5446	0.9920	0.5469
0.541	18.0	486	0.5971	0.7359	0.5911	0.9689	0.6022
0.543	19.0	513	0.6000	0.7390	0.5947	0.9665	0.6064
0.5414	20.0	540	0.5751	0.7117	0.5614	0.9892	0.5646
0.5443	21.0	567	0.5708	0.7066	0.5558	0.9908	0.5585
0.5418	22.0	594	0.5826	0.7203	0.5714	0.9820	0.5770
0.5455	23.0	621	0.5835	0.7219	0.5738	0.9832	0.5790
0.5414	24.0	648	0.5673	0.7025	0.5508	0.9932	0.5528
0.5457	25.0	675	0.5911	0.7307	0.5844	0.9779	0.5917
0.5422	26.0	702	0.5705	0.7066	0.5554	0.9921	0.5576
0.5427	27.0	729	0.5716	0.7079	0.5566	0.9911	0.5591
0.5421	28.0	756	0.5687	0.7041	0.5528	0.9904	0.5557
0.5451	29.0	783	0.5626	0.6968	0.5438	0.9942	0.5453
0.5424	30.0	810	0.5704	0.7068	0.5554	0.9914	0.5578
0.5438	31.0	837	0.5693	0.7054	0.5541	0.9923	0.5563
0.5432	32.0	864	0.5668	0.7022	0.5497	0.9930	0.5516
0.5431	33.0	891	0.5708	0.7075	0.5561	0.9917	0.5584
0.5408	34.0	918	0.5758	0.7137	0.5640	0.9898	0.5671
0.5423	35.0	945	0.5779	0.7161	0.5664	0.9886	0.5698
0.5466	36.0	972	0.5599	0.6936	0.5399	0.9957	0.5410
0.5434	37.0	999	0.5766	0.7149	0.5651	0.9898	0.5681
0.5437	38.0	1026	0.5602	0.6940	0.5405	0.9951	0.5418
0.5432	39.0	1053	0.5690	0.7055	0.5539	0.9923	0.5560
0.5446	40.0	1080	0.5607	0.6944	0.5410	0.9948	0.5423
0.5437	41.0	1107	0.5650	0.7002	0.5475	0.9939	0.5491
0.545	42.0	1134	0.5626	0.6972	0.5443	0.9946	0.5457
0.5457	43.0	1161	0.5565	0.6892	0.5348	0.9961	0.5357
0.5444	44.0	1188	0.5600	0.6938	0.5403	0.9952	0.5415
0.5419	45.0	1215	0.5621	0.6966	0.5435	0.9948	0.5449
0.543	46.0	1242	0.5596	0.6932	0.5394	0.9950	0.5407
0.5436	47.0	1269	0.5587	0.6919	0.5380	0.9959	0.5390
0.5444	48.0	1296	0.5585	0.6917	0.5379	0.9957	0.5390
0.5441	49.0	1323	0.5602	0.6941	0.5406	0.9951	0.5419
0.5425	50.0	1350	0.5575	0.6904	0.5364	0.9958	0.5374

Framework versions

• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0