segformer-b0-finetuned-segments-sidewalk-oct-22

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: 0.5925
• eval_mean_iou: 0.2753
• eval_mean_accuracy: 0.3327
• eval_overall_accuracy: 0.8401
• eval_accuracy_unlabeled: nan
• eval_accuracy_flat-road: 0.8405
• eval_accuracy_flat-sidewalk: 0.9533
• eval_accuracy_flat-crosswalk: 0.6601
• eval_accuracy_flat-cyclinglane: 0.7992
• eval_accuracy_flat-parkingdriveway: 0.5578
• eval_accuracy_flat-railtrack: nan
• eval_accuracy_flat-curb: 0.4836
• eval_accuracy_human-person: 0.6161
• eval_accuracy_human-rider: 0.0
• eval_accuracy_vehicle-car: 0.9299
• eval_accuracy_vehicle-truck: 0.0
• eval_accuracy_vehicle-bus: 0.0
• eval_accuracy_vehicle-tramtrain: nan
• eval_accuracy_vehicle-motorcycle: 0.0
• eval_accuracy_vehicle-bicycle: 0.0003
• eval_accuracy_vehicle-caravan: 0.0
• eval_accuracy_vehicle-cartrailer: 0.0
• eval_accuracy_construction-building: 0.8840
• eval_accuracy_construction-door: 0.0
• eval_accuracy_construction-wall: 0.3660
• eval_accuracy_construction-fenceguardrail: 0.3076
• eval_accuracy_construction-bridge: 0.0
• eval_accuracy_construction-tunnel: 0.0
• eval_accuracy_construction-stairs: 0.0
• eval_accuracy_object-pole: 0.2707
• eval_accuracy_object-trafficsign: 0.0
• eval_accuracy_object-trafficlight: 0.0
• eval_accuracy_nature-vegetation: 0.9456
• eval_accuracy_nature-terrain: 0.8426
• eval_accuracy_sky: 0.9610
• eval_accuracy_void-ground: 0.0
• eval_accuracy_void-dynamic: 0.0
• eval_accuracy_void-static: 0.2296
• eval_accuracy_void-unclear: 0.0
• eval_iou_unlabeled: nan
• eval_iou_flat-road: 0.7077
• eval_iou_flat-sidewalk: 0.8656
• eval_iou_flat-crosswalk: 0.5379
• eval_iou_flat-cyclinglane: 0.7062
• eval_iou_flat-parkingdriveway: 0.4285
• eval_iou_flat-railtrack: nan
• eval_iou_flat-curb: 0.3675
• eval_iou_human-person: 0.3194
• eval_iou_human-rider: 0.0
• eval_iou_vehicle-car: 0.7878
• eval_iou_vehicle-truck: 0.0
• eval_iou_vehicle-bus: 0.0
• eval_iou_vehicle-tramtrain: nan
• eval_iou_vehicle-motorcycle: 0.0
• eval_iou_vehicle-bicycle: 0.0003
• eval_iou_vehicle-caravan: 0.0
• eval_iou_vehicle-cartrailer: 0.0
• eval_iou_construction-building: 0.6784
• eval_iou_construction-door: 0.0
• eval_iou_construction-wall: 0.2711
• eval_iou_construction-fenceguardrail: 0.2716
• eval_iou_construction-bridge: 0.0
• eval_iou_construction-tunnel: 0.0
• eval_iou_construction-stairs: 0.0
• eval_iou_object-pole: 0.2059
• eval_iou_object-trafficsign: 0.0
• eval_iou_object-trafficlight: 0.0
• eval_iou_nature-vegetation: 0.8358
• eval_iou_nature-terrain: 0.7375
• eval_iou_sky: 0.9064
• eval_iou_void-ground: 0.0
• eval_iou_void-dynamic: 0.0
• eval_iou_void-static: 0.1826
• eval_iou_void-unclear: 0.0
• eval_runtime: 11.1228
• eval_samples_per_second: 17.981
• eval_steps_per_second: 2.248
• epoch: 17.4
• step: 1740

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0