<!-- 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_rust
This model is a fine-tuned version of nvidia/mit-b4 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1675
- Mean Iou: 0.7570
- Mean Accuracy: 0.8341
- Overall Accuracy: 0.9396
- Per Category Iou: [0.9341045714153958, 0.5798374949488075]
- Per Category Accuracy: [0.9728893841433082, 0.6952250066198535]
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: 8
- eval_batch_size: 4
- 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 | Per Category Iou | Per Category Accuracy |
|---|---|---|---|---|---|---|---|---|
| 0.1426 | 1.0 | 457 | 0.1936 | 0.6822 | 0.7438 | 0.9242 | [0.9193425271297406, 0.4449612711247542] | [0.9811465495367065, 0.5065393570873438] |
| 0.1066 | 2.0 | 914 | 0.1840 | 0.7247 | 0.8166 | 0.9281 | [0.921792369706966, 0.5275532773693746] | [0.9632516253169541, 0.6698574354887572] |
| 0.1902 | 3.0 | 1371 | 0.1825 | 0.7341 | 0.8195 | 0.9320 | [0.9260615811301273, 0.5422064806171729] | [0.9674973788201077, 0.6715653524785264] |
| 0.0337 | 4.0 | 1828 | 0.1754 | 0.7271 | 0.7949 | 0.9338 | [0.9285138951582051, 0.5256281586371581] | [0.9775754623882825, 0.6121479072473499] |
| 0.0282 | 5.0 | 2285 | 0.1959 | 0.7395 | 0.8653 | 0.9262 | [0.9184942871535651, 0.5605025991065423] | [0.9453582612880436, 0.7853124151415186] |
| 0.2816 | 6.0 | 2742 | 0.1763 | 0.7331 | 0.8035 | 0.9347 | [0.9293865372531065, 0.5367951843584413] | [0.9761229104877399, 0.6308764449372819] |
| 0.1378 | 7.0 | 3199 | 0.1707 | 0.7495 | 0.8309 | 0.9369 | [0.931185547791987, 0.5677946512496371] | [0.9703251796700836, 0.691472671396249] |
| 0.1596 | 8.0 | 3656 | 0.1654 | 0.7511 | 0.8228 | 0.9390 | [0.9337197231959533, 0.5685273561265333] | [0.9757118594786471, 0.6698855265722331] |
| 0.0751 | 9.0 | 4113 | 0.1658 | 0.7478 | 0.8151 | 0.9390 | [0.9338284931081033, 0.5617165036252598] | [0.9780982401336578, 0.6520209898525051] |
| 0.0682 | 10.0 | 4570 | 0.1675 | 0.7570 | 0.8341 | 0.9396 | [0.9341045714153958, 0.5798374949488075] | [0.9728893841433082, 0.6952250066198535] |
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
