learning_rate: 6e-05
train_batch_size: 16
eval_batch_size: 16
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: linear
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Per Category Iou	Per Category Accuracy
0.7163	3.3333	20	0.6298	0.4677	0.8962	0.8544	[0.8524830794058491, 0.08286296421009377]	[0.8532185363290071, 0.9392547548195109]
0.5862	6.6667	40	0.5343	0.4785	0.9238	0.8658	[0.8639249139547209, 0.09306240893340353]	[0.8641289068314079, 0.9833743045672144]
0.4895	10.0	60	0.3810	0.5262	0.9145	0.9184	[0.9173166354069936, 0.13510711817553558]	[0.9184813866782309, 0.9105964549100789]
0.4182	13.3333	80	0.3215	0.5458	0.9068	0.9345	[0.9336555702814376, 0.15803380202998418]	[0.9352691163472723, 0.8783154353732695]
0.3745	16.6667	100	0.2806	0.5498	0.9022	0.9376	[0.9368613530260278, 0.16274092539672888]	[0.9386474264388367, 0.8657652995212834]
0.3392	20.0	120	0.2478	0.5502	0.9051	0.9376	[0.9368321813064416, 0.16362488312508347]	[0.9385390118098444, 0.8717169103376893]
0.3166	23.3333	140	0.2353	0.5650	0.8927	0.9477	[0.9470789571619493, 0.1829323521086895]	[0.9492821665285452, 0.8362013197050071]
0.2868	26.6667	160	0.2257	0.5694	0.8980	0.9495	[0.9489114062839483, 0.18987047347686403]	[0.9510011815357025, 0.8449346616638633]
0.273	30.0	180	0.1981	0.5821	0.8754	0.9576	[0.9571070135309308, 0.20703144845806026]	[0.9599499822677768, 0.790852632940872]
0.2611	33.3333	200	0.2032	0.5776	0.8836	0.9549	[0.9543427027859125, 0.2008596356171414]	[0.9569152914240353, 0.8101953680941907]
0.258	36.6667	220	0.2011	0.5800	0.8854	0.9558	[0.9552498252089475, 0.20469017181011318]	[0.9577853647601005, 0.8131064820804761]
0.2635	40.0	240	0.1809	0.5845	0.8832	0.9579	[0.9574256373418557, 0.21150517563210589]	[0.9600593156648114, 0.8063138827791435]
0.2506	43.3333	260	0.1801	0.5927	0.8752	0.9619	[0.9614287546257282, 0.22395458985274874]	[0.9643490436543448, 0.7861301591408979]
0.2408	46.6667	280	0.1824	0.6003	0.8664	0.9653	[0.9648991073685136, 0.23568352209283777]	[0.968124261540186, 0.7646526070643033]
0.248	50.0	300	0.1839	0.5926	0.8811	0.9613	[0.9609061792276515, 0.22437109204595027]	[0.9636553737824026, 0.798550912149049]

Framework versions

Transformers 4.52.4
Pytorch 2.7.1+cu126
Datasets 3.6.0
Tokenizers 0.21.1

Jedrzej-Smok/2025-06-07_19-10-12

2025-06-07_19-10-12

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Jedrzej-Smok/
2025-06-07_19-10-12