Tianmu28/segformer-b0-segments-floods

segformer-b0-segments-sidewalk-finetuned

This model is a fine-tuned version of nvidia/segformer-b0-finetuned-ade-512-512 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2387
• Mean Iou: 0.8160
• Mean Accuracy: 0.8955
• Overall Accuracy: 0.9070
• Accuracy Background: 0.9351
• Accuracy Target: 0.8559
• Iou Background: 0.8665
• Iou Target: 0.7655

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: 4
• eval_batch_size: 4
• 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: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Background	Accuracy Target	Iou Background	Iou Target
0.2223	1.0	51	0.2299	0.8346	0.9082	0.9123	0.9335	0.8830	0.8608	0.8085
0.1991	2.0	102	0.2313	0.8371	0.9103	0.9136	0.9307	0.8900	0.8622	0.8120
0.1905	3.0	153	0.2269	0.8398	0.9112	0.9153	0.9368	0.8856	0.8653	0.8143
0.2218	4.0	204	0.2287	0.8407	0.9119	0.9158	0.9361	0.8877	0.8659	0.8155
0.2145	5.0	255	0.2275	0.8397	0.9125	0.9150	0.9279	0.8971	0.8637	0.8156
0.1905	6.0	306	0.2301	0.8395	0.9108	0.9152	0.9383	0.8832	0.8654	0.8137
0.2056	7.0	357	0.2278	0.8413	0.9116	0.9163	0.9410	0.8821	0.8672	0.8155

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.20.3