wangzfsh/upernet-convnext-base-AIData-0.8533

upernet-convnext-base-AIData

This model is a fine-tuned version of openmmlab/upernet-convnext-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0005
• Mean Iou: 0.8533
• Mean Accuracy: 0.9119
• Overall Accuracy: 0.9999
• Per Category Iou: [0.9998919682572691, 0.7067961165048544]
• Per Category Accuracy: [0.9999476505782227, 0.8238400603545831]

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use 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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Per Category Iou	Per Category Accuracy
0.0005	11.7647	200	0.0005	0.8681	0.9390	0.9999	[0.999900551853742, 0.7362428842504743]	[0.9999390647960632, 0.8781591852131271]
0.0004	23.5294	400	0.0005	0.8520	0.9004	0.9999	[0.9998936383955853, 0.7041459369817579]	[0.9999565941013053, 0.8008298755186722]
0.0004	35.2941	600	0.0005	0.8515	0.9047	0.9999	[0.9998919687467731, 0.7031454783748362]	[0.9999521819632512, 0.8095058468502452]
0.0004	47.0588	800	0.0005	0.8533	0.9119	0.9999	[0.9998919682572691, 0.7067961165048544]	[0.9999476505782227, 0.8238400603545831]

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.0+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3