trashbot

This model is a fine-tuned version of nvidia/mit-b5 on the mraottth/all_locations_pooled dataset. It achieves the following results on the evaluation set:

• Loss: 0.0189
• Mean Iou: 0.4050
• Mean Accuracy: 0.8101
• Overall Accuracy: 0.8101
• Accuracy Unlabeled: nan
• Accuracy Trash: 0.8101
• Iou Unlabeled: 0.0
• Iou Trash: 0.8101

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: 3
• eval_batch_size: 3
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Trash	Iou Unlabeled	Iou Trash
0.0592	1.0	90	0.0387	0.3723	0.7446	0.7446	nan	0.7446	0.0	0.7446
0.0402	2.0	180	0.0281	0.4123	0.8247	0.8247	nan	0.8247	0.0	0.8247
0.0209	3.0	270	0.0246	0.3691	0.7382	0.7382	nan	0.7382	0.0	0.7382
0.0117	4.0	360	0.0210	0.3882	0.7763	0.7763	nan	0.7763	0.0	0.7763
0.019	5.0	450	0.0198	0.3822	0.7644	0.7644	nan	0.7644	0.0	0.7644
0.0445	6.0	540	0.0199	0.3771	0.7542	0.7542	nan	0.7542	0.0	0.7542
0.0195	7.0	630	0.0191	0.4177	0.8354	0.8354	nan	0.8354	0.0	0.8354
0.008	8.0	720	0.0191	0.4060	0.8119	0.8119	nan	0.8119	0.0	0.8119
0.0268	9.0	810	0.0188	0.4083	0.8166	0.8166	nan	0.8166	0.0	0.8166
0.0061	10.0	900	0.0189	0.4050	0.8101	0.8101	nan	0.8101	0.0	0.8101

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.13.2