foot-finetune-28-jan

This model is a fine-tuned version of nvidia/mit-b0 on the suncy13/FootImg dataset. It achieves the following results on the evaluation set:

• Loss: 0.1107
• Mean Iou: 0.0
• Mean Accuracy: nan
• Overall Accuracy: nan
• Accuracy Foot: nan
• Iou Foot: 0.0

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: 6e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Foot	Iou Foot
0.356	2.0	20	0.5295	0.0	nan	nan	nan	0.0
0.2927	4.0	40	0.3244	0.0	nan	nan	nan	0.0
0.2511	6.0	60	0.2386	0.0	nan	nan	nan	0.0
0.2458	8.0	80	0.2305	0.0	nan	nan	nan	0.0
0.2152	10.0	100	0.2065	0.0	nan	nan	nan	0.0
0.1996	12.0	120	0.1905	0.0	nan	nan	nan	0.0
0.1878	14.0	140	0.1823	0.0	nan	nan	nan	0.0
0.1902	16.0	160	0.1743	0.0	nan	nan	nan	0.0
0.1646	18.0	180	0.1572	0.0	nan	nan	nan	0.0
0.1512	20.0	200	0.1552	0.0	nan	nan	nan	0.0
0.1438	22.0	220	0.1415	0.0	nan	nan	nan	0.0
0.1355	24.0	240	0.1424	0.0	nan	nan	nan	0.0
0.1342	26.0	260	0.1322	0.0	nan	nan	nan	0.0
0.1355	28.0	280	0.1307	0.0	nan	nan	nan	0.0
0.1198	30.0	300	0.1238	0.0	nan	nan	nan	0.0
0.1179	32.0	320	0.1229	0.0	nan	nan	nan	0.0
0.1108	34.0	340	0.1196	0.0	nan	nan	nan	0.0
0.1145	36.0	360	0.1182	0.0	nan	nan	nan	0.0
0.1097	38.0	380	0.1168	0.0	nan	nan	nan	0.0
0.1199	40.0	400	0.1164	0.0	nan	nan	nan	0.0
0.1185	42.0	420	0.1138	0.0	nan	nan	nan	0.0
0.1026	44.0	440	0.1115	0.0	nan	nan	nan	0.0
0.1039	46.0	460	0.1100	0.0	nan	nan	nan	0.0
0.1091	48.0	480	0.1107	0.0	nan	nan	nan	0.0
0.1074	50.0	500	0.1107	0.0	nan	nan	nan	0.0

Framework versions

• Transformers 4.37.1
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1