mask2former-finetuned-ER-Mito-LD4

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:

• Loss: 33.9148
• Dummy: 1.0

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: 0.0004
• 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 results

Training Loss	Epoch	Step	Validation Loss	Dummy
53.0042	1.0	129	41.6702	1.0
41.5032	2.0	258	35.3152	1.0
37.8318	3.0	387	33.2929	1.0
33.1734	4.0	516	31.6052	1.0
31.0889	5.0	645	32.0792	1.0
30.5091	6.0	774	29.4252	1.0
27.7742	7.0	903	29.3660	1.0
27.1136	8.0	1032	28.6043	1.0
25.1614	9.0	1161	28.0848	1.0
24.7794	10.0	1290	28.1507	1.0
23.636	11.0	1419	28.3853	1.0
22.7494	12.0	1548	27.2592	1.0
22.7129	13.0	1677	29.8838	1.0
21.1747	14.0	1806	28.1624	1.0
20.9589	15.0	1935	27.9121	1.0
20.2591	16.0	2064	26.6467	1.0
20.1436	17.0	2193	26.9901	1.0
19.5047	18.0	2322	29.2895	1.0
18.4257	19.0	2451	27.0489	1.0
18.6316	20.0	2580	27.3730	1.0
18.037	21.0	2709	28.0853	1.0
17.6324	22.0	2838	26.6344	1.0
17.19	23.0	2967	28.1709	1.0
17.5784	24.0	3096	26.3646	1.0
16.3714	25.0	3225	28.6477	1.0
16.2177	26.0	3354	29.9328	1.0
15.8326	27.0	3483	27.1418	1.0
15.7345	28.0	3612	28.5265	1.0
14.918	29.0	3741	30.8378	1.0
15.2316	30.0	3870	28.5173	1.0
14.6576	31.0	3999	29.0688	1.0
14.5837	32.0	4128	29.7354	1.0
13.7819	33.0	4257	28.6140	1.0
14.851	34.0	4386	30.7131	1.0
14.1454	35.0	4515	29.3673	1.0
13.5445	36.0	4644	30.1412	1.0
13.3725	37.0	4773	29.7489	1.0
13.8976	38.0	4902	32.2482	1.0
13.2317	39.0	5031	33.3837	1.0
12.8382	40.0	5160	31.9261	1.0
12.8798	41.0	5289	31.0644	1.0
12.5615	42.0	5418	32.6052	1.0
12.4595	43.0	5547	32.6710	1.0
12.9861	44.0	5676	32.3271	1.0
12.3429	45.0	5805	33.1802	1.0
11.6031	46.0	5934	33.3981	1.0
12.7182	47.0	6063	33.2806	1.0
11.8251	48.0	6192	33.9491	1.0
12.4439	49.0	6321	33.4338	1.0
11.9834	50.0	6450	33.8444	1.0

Framework versions

• Transformers 4.50.0.dev0
• Pytorch 2.4.1
• Datasets 3.3.2
• Tokenizers 0.21.0