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
0.7217	3.3333	20	0.6479	0.4187	0.8710	0.7775	[0.7741554846363963, 0.06318359561278017]	[0.7745529798886959, 0.9674065420560748]
0.5822	6.6667	40	0.5667	0.4901	0.9146	0.8757	[0.8737915231451938, 0.10641857800404945]	[0.8744138497099614, 0.9547897196261682]
0.4715	10.0	60	0.4061	0.5395	0.9264	0.9228	[0.9216487567868655, 0.15741299040200857]	[0.92266220812108, 0.9301985981308412]
0.4017	13.3333	80	0.3289	0.5343	0.9306	0.9182	[0.9169649202323022, 0.15169531323377478]	[0.9177825299608374, 0.9433995327102803]
0.3538	16.6667	100	0.2727	0.5764	0.9257	0.9463	[0.9455599555306462, 0.20716139909545855]	[0.9469851447248314, 0.9043224299065421]
0.3136	20.0	120	0.2471	0.5845	0.9187	0.9510	[0.9502706863326824, 0.2187333150550513]	[0.9519862857395247, 0.8853971962616822]
0.2969	23.3333	140	0.2455	0.5766	0.9337	0.9457	[0.9448753190202712, 0.2082343312740905]	[0.9460447277759784, 0.9214369158878505]
0.274	26.6667	160	0.2239	0.5900	0.9242	0.9529	[0.9522827008599458, 0.2277052543935294]	[0.9538643591178116, 0.8945677570093458]
0.2669	30.0	180	0.2011	0.5985	0.9213	0.9568	[0.9561495842839874, 0.24086883238722193]	[0.9578845955655017, 0.8848130841121495]
0.2754	33.3333	200	0.2037	0.5839	0.9281	0.9498	[0.9491200570396355, 0.21875793572416127]	[0.9505296516592562, 0.9057242990654205]
0.257	36.6667	220	0.1999	0.6014	0.9239	0.9577	[0.9570845806698737, 0.24570948836758746]	[0.9587587600482906, 0.8889602803738318]
0.2448	40.0	240	0.1826	0.6102	0.9159	0.9615	[0.9609953369978641, 0.2593548387096774]	[0.9629814345867318, 0.8688084112149532]
0.2398	43.3333	260	0.1703	0.6178	0.9080	0.9646	[0.9641522441796739, 0.2713600656802194]	[0.9664385250728778, 0.8494742990654206]
0.2313	46.6667	280	0.1885	0.6157	0.9158	0.9634	[0.9628763079239877, 0.2684652408352332]	[0.9648990754100292, 0.8666471962616823]
0.2485	50.0	300	0.1832	0.6175	0.9145	0.9640	[0.9635604705571247, 0.27136195057467044]	[0.9656352139218516, 0.8633177570093458]

Framework versions

Transformers 4.52.4
Pytorch 2.7.1+cu126
Datasets 3.6.0
Tokenizers 0.21.1

Jedrzej-Smok/2025-06-07_17-30-30

2025-06-07_17-30-30

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Jedrzej-Smok/
2025-06-07_17-30-30