segformer-b0-finetuned-agriculture

This model is a fine-tuned version of nvidia/mit-b0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3305
• Mean Iou: 0.4242
• Mean Accuracy: 0.5107
• Overall Accuracy: 0.6733
• Accuracy Unlabeled: nan
• Accuracy Nutrient Deficiency: 0.6872
• Accuracy Planter Skip: 0.1915
• Accuracy Water: 0.8549
• Accuracy Waterway: 0.1797
• Accuracy Weed Cluster: 0.6404
• Iou Unlabeled: 0.0
• Iou Nutrient Deficiency: 0.6865
• Iou Planter Skip: 0.1914
• Iou Water: 0.8475
• Iou Waterway: 0.1795
• Iou Weed Cluster: 0.6401

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: 1e-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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Nutrient Deficiency	Accuracy Planter Skip	Accuracy Water	Accuracy Waterway	Accuracy Weed Cluster	Iou Unlabeled	Iou Nutrient Deficiency	Iou Planter Skip	Iou Water	Iou Waterway	Iou Weed Cluster
0.2889	1.0	8145	0.4127	0.2578	0.3110	0.4484	nan	0.3062	0.0	0.7988	0.0007	0.4496	0.0	0.3062	0.0	0.7913	0.0007	0.4485
0.3157	2.0	16290	0.3877	0.3374	0.4070	0.5970	nan	0.5241	0.0023	0.8816	0.0303	0.5969	0.0	0.5237	0.0023	0.8715	0.0301	0.5968
0.2637	3.0	24435	0.3531	0.3717	0.4480	0.6171	nan	0.5638	0.0409	0.8804	0.1563	0.5984	0.0	0.5633	0.0409	0.8723	0.1554	0.5982
0.4715	4.0	32580	0.3337	0.3653	0.4398	0.6073	nan	0.6172	0.1068	0.8077	0.0976	0.5698	0.0	0.6164	0.1068	0.8015	0.0976	0.5696
0.0668	5.0	40725	0.3305	0.4242	0.5107	0.6733	nan	0.6872	0.1915	0.8549	0.1797	0.6404	0.0	0.6865	0.1914	0.8475	0.1795	0.6401

Framework versions

• Transformers 4.39.1
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2