swin-large-mask2former-finetuned-Mitoliked-Mito-Innertongue

This model is a fine-tuned version of Dnq2025/swin-large-mask2former-finetuned-ER-Mito-LD8 on the Dnq2025/Mask2former_Finetune dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 0.0004
train_batch_size: 4
eval_batch_size: 4
seed: 1337
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: polynomial
training_steps: 3225

Training Loss	Epoch	Step	Mean Iou	Validation Loss
No log	16.6667	100	0.4737	34.6305
20.404	33.3333	200	0.4715	42.0013
10.0702	50.0	300	0.4774	46.1805
10.0702	66.6667	400	0.4821	50.9460
9.0859	83.3333	500	0.4920	48.5783
8.6094	100.0	600	0.4749	52.5121
8.6094	116.6667	700	0.4865	49.8120
8.4671	133.3333	800	0.4812	56.1730
8.3066	150.0	900	0.4934	60.8687
8.3066	166.6667	1000	0.4916	59.8411
8.1972	183.3333	1100	0.4916	63.6835
8.0926	200.0	1200	0.4905	60.2863
8.0926	216.6667	1300	0.4886	63.5495
8.0059	233.3333	1400	0.4924	65.0410
7.974	250.0	1500	0.4900	66.5468
7.974	266.6667	1600	0.4902	67.8736
8.0177	283.3333	1700	0.4851	68.4019
7.9356	300.0	1800	0.4894	74.3255
7.9356	316.6667	1900	0.4900	73.2780
7.8784	333.3333	2000	0.4899	75.2934
7.8445	350.0	2100	0.4878	77.9780
7.8445	366.6667	2200	0.4868	76.0033
7.7889	383.3333	2300	0.4879	78.4704
7.7828	400.0	2400	0.4860	77.1425
7.7828	416.6667	2500	0.4915	78.8530
7.7601	433.3333	2600	0.4875	81.9598
7.738	450.0	2700	0.4865	79.7205
7.738	466.6667	2800	0.4866	79.9854
7.7139	483.3333	2900	0.4869	82.6984
7.7043	500.0	3000	0.4847	81.8537
7.7043	516.6667	3100	0.4855	81.6806
7.6982	533.3333	3200	0.4850	85.2813