This model is a fine-tuned version of nvidia/mit-b0 on the bfwggggg/image-with-puzzle dataset. It achieves the following results on the evaluation set:

Loss: 0.2327
Mean Iou: 0.4736
Mean Accuracy: 0.9472
Overall Accuracy: 0.9472
Accuracy Unlabeled: nan
Accuracy Missing-puzzle: 0.9472
Iou Unlabeled: 0.0
Iou Missing-puzzle: 0.9472

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: 2
eval_batch_size: 2
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Missing-puzzle	Iou Missing-puzzle
0.5575	5.0	20	0.6601	0.4998	0.9996	0.9996	nan	0.9996	0.9996
0.365	10.0	40	0.5628	0.4980	0.9960	0.9960	nan	0.9960	0.9960
0.2788	15.0	60	0.3816	0.4812	0.9624	0.9624	nan	0.9624	0.9624
0.2527	20.0	80	0.3869	0.4806	0.9611	0.9611	nan	0.9611	0.9611
0.2145	25.0	100	0.2733	0.4663	0.9326	0.9326	nan	0.9326	0.9326
0.206	30.0	120	0.2672	0.4739	0.9478	0.9478	nan	0.9478	0.9478
0.1866	35.0	140	0.2351	0.4667	0.9334	0.9334	nan	0.9334	0.9334
0.1696	40.0	160	0.2099	0.4749	0.9497	0.9497	nan	0.9497	0.9497
0.1639	45.0	180	0.2058	0.4723	0.9445	0.9445	nan	0.9445	0.9445
0.1719	50.0	200	0.2327	0.4736	0.9472	0.9472	nan	0.9472	0.9472

Framework versions

Transformers 4.38.2
Pytorch 2.1.0+cu121
Datasets 2.18.0
Tokenizers 0.15.2

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

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

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

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