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	1.1364	100	0.3146	37.8845
46.6024	2.2727	200	0.3627	36.0673
33.1722	3.4091	300	0.3700	34.6181
33.1722	4.5455	400	0.3653	34.6646
28.6661	5.6818	500	0.3817	33.8163
25.6461	6.8182	600	0.3686	34.5651
25.6461	7.9545	700	0.3872	34.6853
22.9594	9.0909	800	0.4054	34.6685
20.9253	10.2273	900	0.4176	33.7321
20.9253	11.3636	1000	0.4234	33.3689
19.4883	12.5	1100	0.4244	34.2274
18.0467	13.6364	1200	0.3938	36.7450
18.0467	14.7727	1300	0.4114	35.1488
16.8268	15.9091	1400	0.3904	36.7316
16.15	17.0455	1500	0.3792	36.5393
16.15	18.1818	1600	0.3835	36.0319
15.2578	19.3182	1700	0.4292	36.3998
14.6085	20.4545	1800	0.3808	36.4097
14.6085	21.5909	1900	0.4252	37.5680
13.9805	22.7273	2000	0.3874	36.9057
13.4133	23.8636	2100	0.3949	37.5872
13.4133	25.0	2200	0.4442	36.8461
12.9534	26.1364	2300	0.3961	37.5527
12.4781	27.2727	2400	0.4437	36.9928
12.4781	28.4091	2500	0.3937	38.2435
12.0214	29.5455	2600	0.3901	39.0895
11.791	30.6818	2700	0.3845	40.6167
11.791	31.8182	2800	0.4469	37.9631
11.5261	32.9545	2900	0.3883	40.8935
11.3609	34.0909	3000	0.3869	41.0478
11.3609	35.2273	3100	0.3972	39.0594
11.1008	36.3636	3200	0.3880	40.7524