segformer-b0-finetuned-segments-sidewalk-2

This model is a fine-tuned version of segformer on the condensed_dataset_4_2 dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.0307
• eval_overall_accuracy: 0.9450
• eval_mean_per_class_accuracy: 0.8402
• eval_accuracy_Panel: 0.9704
• eval_accuracy_Asphalt: 0.9612
• eval_accuracy_Brick: 0.9642
• eval_accuracy_Cracked_Deteriorated: 0.7501
• eval_accuracy_Obstruction: 0.6817
• eval_accuracy_Gutter: 0.7340
• eval_accuracy_DW: 0.9319
• eval_accuracy_Gravel: 0.7284
• eval_runtime: 128.3265
• eval_samples_per_second: 38.854
• eval_steps_per_second: 1.216
• epoch: 3.5728
• step: 9000

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: 64
• eval_batch_size: 32
• 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: 20

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.5.0
• Tokenizers 0.21.0