mask2former-finetuned-ER-Mito-LD5

This model is a fine-tuned version of facebook/mask2former-swin-base-IN21k-ade-semantic on the Dnq2025/Mask2former_Pretrain 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: 6450

Training Loss	Epoch	Step	Validation Loss
53.498	1.0	129	39.6802
39.3157	2.0	258	35.9394
37.2096	3.0	387	32.0225
31.9877	4.0	516	33.2635
30.1511	5.0	645	29.8756
28.3667	6.0	774	30.3257
26.7492	7.0	903	27.9416
25.6035	8.0	1032	27.4391
24.5091	9.0	1161	28.4225
23.8578	10.0	1290	26.4271
22.6785	11.0	1419	26.4148
22.0847	12.0	1548	26.6679
22.0106	13.0	1677	26.7030
20.45	14.0	1806	26.1600
20.1949	15.0	1935	26.2444
19.1922	16.0	2064	27.0105
18.9458	17.0	2193	24.9449
18.46	18.0	2322	27.8372
17.3966	19.0	2451	27.0517
17.5908	20.0	2580	28.5696
16.9413	21.0	2709	27.3707
16.3963	22.0	2838	26.4041
16.2948	23.0	2967	25.3316
16.2511	24.0	3096	27.9766
15.4496	25.0	3225	27.6993
15.1992	26.0	3354	27.9919
14.9445	27.0	3483	25.4937
14.8226	28.0	3612	28.7659
14.264	29.0	3741	26.7018
14.348	30.0	3870	28.9018
13.936	31.0	3999	28.2813
13.7577	32.0	4128	30.0501
13.1629	33.0	4257	28.0087
14.1035	34.0	4386	28.3435
13.4379	35.0	4515	28.9629
12.9478	36.0	4644	29.8509
12.8114	37.0	4773	28.9036
13.2322	38.0	4902	29.9045
12.7433	39.0	5031	31.3430
12.3428	40.0	5160	31.3746
12.3295	41.0	5289	31.6009
12.1459	42.0	5418	31.6387
11.8999	43.0	5547	32.2195
12.4076	44.0	5676	32.5034
11.7797	45.0	5805	32.9062
11.1345	46.0	5934	32.4447
12.3552	47.0	6063	32.7274
11.3111	48.0	6192	33.0397
11.8742	49.0	6321	33.3195
11.5268	50.0	6450	33.3223