segformer-vineyard-rows

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

• Loss: 0.0012
• Mean Iou: 0.5000
• Mean Accuracy: 1.0000
• Overall Accuracy: 1.0000
• Accuracy Unlabeled: 1.0000
• Accuracy Vineyard-row: nan
• Iou Unlabeled: 1.0000
• Iou Vineyard-row: 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: 8
• eval_batch_size: 8
• 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Vineyard-row	Iou Unlabeled	Iou Vineyard-row
0.0333	0.3509	20	0.0552	0.4994	0.9987	0.9987	0.9987	nan	0.9987	0.0
0.0104	0.7018	40	0.0056	0.4999	0.9998	0.9998	0.9998	nan	0.9998	0.0
0.0116	1.0526	60	0.0053	0.4999	0.9998	0.9998	0.9998	nan	0.9998	0.0
0.008	1.4035	80	0.0054	0.4999	0.9998	0.9998	0.9998	nan	0.9998	0.0
0.0043	1.7544	100	0.0053	0.4999	0.9998	0.9998	0.9998	nan	0.9998	0.0
0.004	2.1053	120	0.0040	0.4999	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0035	2.4561	140	0.0030	0.4999	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0036	2.8070	160	0.0039	0.4999	0.9998	0.9998	0.9998	nan	0.9998	0.0
0.0036	3.1579	180	0.0029	0.4999	0.9998	0.9998	0.9998	nan	0.9998	0.0
0.002	3.5088	200	0.0030	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0026	3.8596	220	0.0026	0.4999	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0025	4.2105	240	0.0020	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0026	4.5614	260	0.0022	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0019	4.9123	280	0.0018	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0016	5.2632	300	0.0018	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0025	5.6140	320	0.0016	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0009	5.9649	340	0.0017	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0021	6.3158	360	0.0017	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0014	6.6667	380	0.0017	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0021	7.0175	400	0.0015	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.003	7.3684	420	0.0017	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0013	7.7193	440	0.0014	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0009	8.0702	460	0.0014	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0024	8.4211	480	0.0015	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0014	8.7719	500	0.0014	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0021	9.1228	520	0.0014	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0016	9.4737	540	0.0013	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0011	9.8246	560	0.0012	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0a0+df5bbc09d1.nv24.12
• Datasets 3.3.2
• Tokenizers 0.21.0