jpodivin/upernet-convnext-small-finetuned

来源：🤗有风险🏴‍☠️ 救援者：isiliconimage-segmentationmit82M 参数327 MBsafetensors✓ 2 个校验和980 天前更新

upernet-convnext-small-finetuned

This model is a fine-tuned version of openmmlab/upernet-convnext-small on the jpodivin/plantorgans dataset. It achieves the following results on the evaluation set:

Loss: 0.2874
Mean Iou: 0.4231
Mean Accuracy: 0.5343
Overall Accuracy: 0.7437
Accuracy Void: nan
Accuracy Fruit: 0.8642
Accuracy Leaf: 0.7167
Accuracy Flower: 0.0
Accuracy Stem: 0.5563
Iou Void: 0.0
Iou Fruit: 0.8605
Iou Leaf: 0.7108
Iou Flower: 0.0
Iou Stem: 0.5440
Median Iou: 0.5440

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: 0.0006
train_batch_size: 10
eval_batch_size: 10
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Void	Accuracy Fruit	Accuracy Leaf	Accuracy Stem	Iou Fruit	Iou Leaf	Iou Stem	Median Iou
0.8456	1.0	575	0.3074	0.3946	0.4987	0.7054	nan	0.8110	0.6951	0.4888	0.8088	0.6852	0.4791	0.4791
0.3006	2.0	1150	0.2868	0.3945	0.4965	0.7227	nan	0.8533	0.7186	0.4139	0.8494	0.7139	0.4092	0.4092
0.3315	3.0	1725	0.2874	0.4231	0.5343	0.7437	nan	0.8642	0.7167	0.5563	0.8605	0.7108	0.5440	0.5440

Framework versions

Transformers 4.38.0.dev0
Pytorch 2.1.2+cu121
Datasets 2.16.1
Tokenizers 0.15.0