<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
segformer-b0-finetuned-metalography_D1
This model is a fine-tuned version of ironchanchellor/segformer-b0-finetuned-metalography_D1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0122
- Mean Iou: 0.7893
- Mean Accuracy: 0.9654
- Overall Accuracy: 0.9943
- Accuracy Background: nan
- Accuracy Haz: 0.9964
- Accuracy Matrix: 0.9824
- Accuracy Porosity: 0.8724
- Accuracy Carbides: 0.9779
- Accuracy Substrate: 0.9979
- Iou Background: 0.0
- Iou Haz: 0.9935
- Iou Matrix: 0.9707
- Iou Porosity: 0.8274
- Iou Carbides: 0.9486
- Iou Substrate: 0.9955
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Background | Accuracy Haz | Accuracy Matrix | Accuracy Porosity | Accuracy Carbides | Accuracy Substrate | Iou Background | Iou Haz | Iou Matrix | Iou Porosity | Iou Carbides | Iou Substrate |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0208 | 0.54 | 20 | 0.0141 | 0.7892 | 0.9684 | 0.9936 | nan | 0.9948 | 0.9836 | 0.8893 | 0.9768 | 0.9975 | 0.0 | 0.9914 | 0.9710 | 0.8299 | 0.9491 | 0.9939 |
| 0.016 | 1.08 | 40 | 0.0124 | 0.7896 | 0.9666 | 0.9941 | nan | 0.9947 | 0.9855 | 0.8819 | 0.9720 | 0.9987 | 0.0 | 0.9928 | 0.9709 | 0.8300 | 0.9489 | 0.9952 |
| 0.0144 | 1.62 | 60 | 0.0122 | 0.7893 | 0.9654 | 0.9943 | nan | 0.9964 | 0.9824 | 0.8724 | 0.9779 | 0.9979 | 0.0 | 0.9935 | 0.9707 | 0.8274 | 0.9486 | 0.9955 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.3
- Tokenizers 0.13.3
