README.md

metadata

license: other
base_model: nvidia/mit-b0
tags:
  - vision
  - image-segmentation
  - generated_from_trainer
model-index:
  - name: segformer-b0-miic-tl
    results: []

segformer-b0-miic-tl

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

• Loss: 0.3532
• Mean Iou: 0.4001
• Mean Accuracy: 0.8002
• Overall Accuracy: 0.8002
• Accuracy Unlabeled: nan
• Accuracy Circuit: 0.8002
• Iou Unlabeled: 0.0
• Iou Circuit: 0.8002
• Dice Coefficient: 0.7456

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 Circuit	Iou Unlabeled	Iou Circuit	Dice Coefficient
0.5722	3.12	250	0.4296	0.3809	0.7618	0.7618	nan	0.7618	0.0	0.7618	0.6766
0.547	6.25	500	0.3983	0.3370	0.6739	0.6739	nan	0.6739	0.0	0.6739	0.6065
0.5147	9.38	750	0.3643	0.3487	0.6974	0.6974	nan	0.6974	0.0	0.6974	0.6477
0.5083	12.5	1000	0.3505	0.3006	0.6012	0.6012	nan	0.6012	0.0	0.6012	0.5586
0.4818	15.62	1250	0.3184	0.4400	0.8799	0.8799	nan	0.8799	0.0	0.8799	0.7758
0.4664	18.75	1500	0.3622	0.4347	0.8693	0.8693	nan	0.8693	0.0	0.8693	0.7755
0.4504	21.88	1750	0.3279	0.4327	0.8654	0.8654	nan	0.8654	0.0	0.8654	0.7792
0.4427	25.0	2000	0.3168	0.4386	0.8771	0.8771	nan	0.8771	0.0	0.8771	0.7840
0.4336	28.12	2250	0.2790	0.4100	0.8200	0.8200	nan	0.8200	0.0	0.8200	0.7636
0.4226	31.25	2500	0.3237	0.4148	0.8295	0.8295	nan	0.8295	0.0	0.8295	0.7641
0.4155	34.38	2750	0.3336	0.4169	0.8339	0.8339	nan	0.8339	0.0	0.8339	0.7664
0.4082	37.5	3000	0.3787	0.4267	0.8533	0.8533	nan	0.8533	0.0	0.8533	0.7760
0.403	40.62	3250	0.3541	0.3693	0.7387	0.7387	nan	0.7387	0.0	0.7387	0.6942
0.398	43.75	3500	0.3361	0.3864	0.7728	0.7728	nan	0.7728	0.0	0.7728	0.7244
0.3943	46.88	3750	0.3599	0.4053	0.8106	0.8106	nan	0.8106	0.0	0.8106	0.7519
0.3951	50.0	4000	0.3532	0.4001	0.8002	0.8002	nan	0.8002	0.0	0.8002	0.7456

Framework versions

• Transformers 4.36.2
• Pytorch 1.11.0+cu115
• Datasets 2.15.0
• Tokenizers 0.15.0