README.md

metadata

license: other
tags:
  - generated_from_trainer
model-index:
  - name: segformer-b0-finetuned-segments-toolwear
    results: []

segformer-b0-finetuned-segments-toolwear

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

• Loss: 0.1501
• Mean Iou: 0.4560
• Mean Accuracy: 0.9040
• Overall Accuracy: 0.9643
• Accuracy Unlabeled: nan
• Accuracy Wear: 0.8404
• Accuracy Tool: 0.9675
• Iou Unlabeled: 0.0
• Iou Wear: 0.4034
• Iou Tool: 0.9646

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Wear	Accuracy Tool	Iou Unlabeled	Iou Wear	Iou Tool
0.4464	1.82	20	0.6527	0.3325	0.5116	0.9740	nan	0.0242	0.9990	0.0	0.0235	0.9740
0.3069	3.64	40	0.3300	0.4958	0.8505	0.9661	nan	0.7288	0.9723	0.0	0.5213	0.9662
0.276	5.45	60	0.2597	0.4089	0.9324	0.9368	nan	0.9278	0.9370	0.0	0.2909	0.9358
0.2648	7.27	80	0.2321	0.4338	0.8839	0.9567	nan	0.8071	0.9607	0.0	0.3441	0.9572
0.245	9.09	100	0.2298	0.4021	0.9265	0.9359	nan	0.9167	0.9364	0.0	0.2715	0.9348
0.2047	10.91	120	0.1897	0.4379	0.8814	0.9446	nan	0.8147	0.9480	0.0	0.3684	0.9455
0.1695	12.73	140	0.1681	0.4561	0.8444	0.9636	nan	0.7188	0.9701	0.0	0.4026	0.9657
0.1556	14.55	160	0.1741	0.4289	0.9060	0.9494	nan	0.8603	0.9517	0.0	0.3372	0.9497
0.1435	16.36	180	0.1528	0.4746	0.8851	0.9679	nan	0.7978	0.9723	0.0	0.4549	0.9689
0.1208	18.18	200	0.1648	0.4379	0.9126	0.9577	nan	0.8650	0.9601	0.0	0.3560	0.9577
0.1425	20.0	220	0.1587	0.4451	0.9116	0.9576	nan	0.8631	0.9601	0.0	0.3774	0.9578
0.1124	21.82	240	0.1515	0.4291	0.9044	0.9491	nan	0.8574	0.9515	0.0	0.3380	0.9493
0.1509	23.64	260	0.1501	0.4560	0.9040	0.9643	nan	0.8404	0.9675	0.0	0.4034	0.9646

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3