segformer-b0-finetuned-segments-sidewalk-2

This model is a fine-tuned version of nvidia/mit-b0 on the segments/sidewalk-semantic dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.5277
• eval_mean_iou: 0.3177
• eval_mean_accuracy: 0.3837
• eval_overall_accuracy: 0.8556
• eval_accuracy_unlabeled: nan
• eval_accuracy_flat-road: 0.9247
• eval_accuracy_flat-sidewalk: 0.9046
• eval_accuracy_flat-crosswalk: 0.9004
• eval_accuracy_flat-cyclinglane: 0.9024
• eval_accuracy_flat-parkingdriveway: 0.5583
• eval_accuracy_flat-railtrack: nan
• eval_accuracy_flat-curb: 0.5870
• eval_accuracy_human-person: 0.7960
• eval_accuracy_human-rider: 0.0
• eval_accuracy_vehicle-car: 0.9470
• eval_accuracy_vehicle-truck: 0.0
• eval_accuracy_vehicle-bus: 0.0
• eval_accuracy_vehicle-tramtrain: 0.0
• eval_accuracy_vehicle-motorcycle: 0.0
• eval_accuracy_vehicle-bicycle: 0.6493
• eval_accuracy_vehicle-caravan: 0.0
• eval_accuracy_vehicle-cartrailer: 0.0
• eval_accuracy_construction-building: 0.8838
• eval_accuracy_construction-door: 0.0
• eval_accuracy_construction-wall: 0.4337
• eval_accuracy_construction-fenceguardrail: 0.3946
• eval_accuracy_construction-bridge: 0.0
• eval_accuracy_construction-tunnel: nan
• eval_accuracy_construction-stairs: 0.0
• eval_accuracy_object-pole: 0.2846
• eval_accuracy_object-trafficsign: 0.0
• eval_accuracy_object-trafficlight: 0.0
• eval_accuracy_nature-vegetation: 0.9430
• eval_accuracy_nature-terrain: 0.8979
• eval_accuracy_sky: 0.9559
• eval_accuracy_void-ground: 0.0
• eval_accuracy_void-dynamic: 0.0
• eval_accuracy_void-static: 0.3145
• eval_accuracy_void-unclear: 0.0
• eval_iou_unlabeled: nan
• eval_iou_flat-road: 0.7544
• eval_iou_flat-sidewalk: 0.8653
• eval_iou_flat-crosswalk: 0.6874
• eval_iou_flat-cyclinglane: 0.8249
• eval_iou_flat-parkingdriveway: 0.3996
• eval_iou_flat-railtrack: nan
• eval_iou_flat-curb: 0.4591
• eval_iou_human-person: 0.4778
• eval_iou_human-rider: 0.0
• eval_iou_vehicle-car: 0.8108
• eval_iou_vehicle-truck: 0.0
• eval_iou_vehicle-bus: 0.0
• eval_iou_vehicle-tramtrain: 0.0
• eval_iou_vehicle-motorcycle: 0.0
• eval_iou_vehicle-bicycle: 0.5337
• eval_iou_vehicle-caravan: 0.0
• eval_iou_vehicle-cartrailer: 0.0
• eval_iou_construction-building: 0.7103
• eval_iou_construction-door: 0.0
• eval_iou_construction-wall: 0.3464
• eval_iou_construction-fenceguardrail: 0.3532
• eval_iou_construction-bridge: 0.0
• eval_iou_construction-tunnel: nan
• eval_iou_construction-stairs: 0.0
• eval_iou_object-pole: 0.2448
• eval_iou_object-trafficsign: 0.0
• eval_iou_object-trafficlight: 0.0
• eval_iou_nature-vegetation: 0.8543
• eval_iou_nature-terrain: 0.7232
• eval_iou_sky: 0.9191
• eval_iou_void-ground: 0.0
• eval_iou_void-dynamic: 0.0
• eval_iou_void-static: 0.2024
• eval_iou_void-unclear: 0.0
• eval_runtime: 33.3831
• eval_samples_per_second: 5.991
• eval_steps_per_second: 1.498
• epoch: 11.3
• step: 2260

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2