File size: 20,659 Bytes
4da6238 e48b8ff b1902a4 e48b8ff 4da6238 e48b8ff b1902a4 e48b8ff f6d8282 e48b8ff f6d8282 b1902a4 e48b8ff |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 |
---
license: other
base_model: nvidia/segformer-b4-finetuned-cityscapes-1024-1024
tags:
- vision
- image-segmentation
- generated_from_trainer
model-index:
- name: segformer-b4-cityscapes-finetuned-coastTrain
results: []
---
<!-- 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-b4-cityscapes-finetuned-coastTrain
This model is a fine-tuned version of [nvidia/segformer-b4-finetuned-cityscapes-1024-1024](https://huggingface.co/nvidia/segformer-b4-finetuned-cityscapes-1024-1024) on the peldrak/coastTrain dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2345
- Mean Iou: 0.7920
- Mean Accuracy: 0.8609
- Overall Accuracy: 0.9360
- Accuracy Water: 0.9603
- Accuracy Whitewater: 0.6503
- Accuracy Sediment: 0.8872
- Accuracy Other Natural Terrain: 0.6902
- Accuracy Vegetation: 0.9383
- Accuracy Development: 0.9340
- Accuracy Unknown: 0.9659
- Iou Water: 0.9194
- Iou Whitewater: 0.5196
- Iou Sediment: 0.8231
- Iou Other Natural Terrain: 0.6344
- Iou Vegetation: 0.8728
- Iou Development: 0.8571
- Iou Unknown: 0.9179
- F1 Score: 0.9354
## 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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
### Training results
| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Water | Accuracy Whitewater | Accuracy Sediment | Accuracy Other Natural Terrain | Accuracy Vegetation | Accuracy Development | Accuracy Unknown | Iou Water | Iou Whitewater | Iou Sediment | Iou Other Natural Terrain | Iou Vegetation | Iou Development | Iou Unknown | F1 Score |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:--------------:|:-------------------:|:-----------------:|:------------------------------:|:-------------------:|:--------------------:|:----------------:|:---------:|:--------------:|:------------:|:-------------------------:|:--------------:|:---------------:|:-----------:|:--------:|
| 1.6229 | 0.16 | 20 | 1.5080 | 0.3461 | 0.4464 | 0.6629 | 0.7064 | 0.0117 | 0.3677 | 0.0094 | 0.9033 | 0.4311 | 0.6952 | 0.6022 | 0.0115 | 0.2505 | 0.0054 | 0.5058 | 0.3657 | 0.6818 | 0.6578 |
| 1.6972 | 0.31 | 40 | 1.1157 | 0.4168 | 0.5175 | 0.7412 | 0.7367 | 0.0 | 0.5983 | 0.0000 | 0.9738 | 0.5606 | 0.7528 | 0.6767 | 0.0 | 0.4343 | 0.0000 | 0.5731 | 0.4817 | 0.7519 | 0.7355 |
| 1.1969 | 0.47 | 60 | 0.8521 | 0.4777 | 0.5606 | 0.8157 | 0.8778 | 0.0001 | 0.5850 | 0.0000 | 0.9674 | 0.5962 | 0.8978 | 0.7865 | 0.0001 | 0.4970 | 0.0000 | 0.6699 | 0.4942 | 0.8960 | 0.8025 |
| 1.2209 | 0.62 | 80 | 0.6990 | 0.5024 | 0.6009 | 0.8277 | 0.8306 | 0.0 | 0.7406 | 0.0 | 0.9174 | 0.7970 | 0.9208 | 0.7942 | 0.0 | 0.5641 | 0.0 | 0.6800 | 0.5612 | 0.9175 | 0.8201 |
| 0.793 | 0.78 | 100 | 0.5543 | 0.5256 | 0.6016 | 0.8590 | 0.9409 | 0.0 | 0.6744 | 0.0 | 0.9365 | 0.7280 | 0.9312 | 0.8480 | 0.0 | 0.5903 | 0.0 | 0.7405 | 0.5737 | 0.9264 | 0.8458 |
| 1.1597 | 0.93 | 120 | 0.4957 | 0.5524 | 0.6231 | 0.8720 | 0.9201 | 0.0 | 0.7323 | 0.0 | 0.9560 | 0.8134 | 0.9401 | 0.8588 | 0.0 | 0.6428 | 0.0 | 0.7344 | 0.6976 | 0.9335 | 0.8602 |
| 0.689 | 1.09 | 140 | 0.4376 | 0.5692 | 0.6374 | 0.8844 | 0.9221 | 0.0 | 0.7930 | 0.0000 | 0.9586 | 0.8469 | 0.9411 | 0.8748 | 0.0 | 0.6899 | 0.0000 | 0.7542 | 0.7310 | 0.9348 | 0.8730 |
| 0.9423 | 1.24 | 160 | 0.3968 | 0.5715 | 0.6433 | 0.8882 | 0.9328 | 0.0000 | 0.8273 | 0.0 | 0.9201 | 0.8664 | 0.9564 | 0.8794 | 0.0000 | 0.6963 | 0.0 | 0.7687 | 0.7135 | 0.9424 | 0.8766 |
| 1.2176 | 1.4 | 180 | 0.3838 | 0.5673 | 0.6397 | 0.8811 | 0.9107 | 0.0000 | 0.7816 | 0.0 | 0.9461 | 0.8832 | 0.9561 | 0.8578 | 0.0000 | 0.6427 | 0.0 | 0.7728 | 0.7550 | 0.9429 | 0.8695 |
| 0.4714 | 1.55 | 200 | 0.4459 | 0.5380 | 0.6092 | 0.8570 | 0.9730 | 0.0 | 0.5500 | 0.0 | 0.8806 | 0.9121 | 0.9490 | 0.7998 | 0.0 | 0.4883 | 0.0 | 0.7722 | 0.7618 | 0.9442 | 0.8398 |
| 0.5087 | 1.71 | 220 | 0.4062 | 0.5677 | 0.6359 | 0.8827 | 0.9365 | 0.0008 | 0.7216 | 0.0 | 0.9511 | 0.8918 | 0.9499 | 0.8844 | 0.0008 | 0.6722 | 0.0 | 0.7302 | 0.7463 | 0.9402 | 0.8709 |
| 0.484 | 1.86 | 240 | 0.3121 | 0.5926 | 0.6518 | 0.9017 | 0.9688 | 0.0001 | 0.7858 | 0.0 | 0.9341 | 0.9243 | 0.9498 | 0.8788 | 0.0001 | 0.7236 | 0.0 | 0.8069 | 0.7972 | 0.9420 | 0.8886 |
| 0.443 | 2.02 | 260 | 0.3554 | 0.5811 | 0.6575 | 0.8904 | 0.8939 | 0.0001 | 0.8842 | 0.0 | 0.9316 | 0.9332 | 0.9599 | 0.8541 | 0.0001 | 0.6697 | 0.0 | 0.8205 | 0.7724 | 0.9512 | 0.8798 |
| 0.466 | 2.17 | 280 | 0.3265 | 0.5830 | 0.6553 | 0.8954 | 0.9347 | 0.0021 | 0.8131 | 0.0 | 0.9256 | 0.9487 | 0.9627 | 0.8786 | 0.0021 | 0.6993 | 0.0 | 0.7950 | 0.7566 | 0.9494 | 0.8833 |
| 0.7117 | 2.33 | 300 | 0.4096 | 0.5634 | 0.6494 | 0.8672 | 0.8367 | 0.0076 | 0.9133 | 0.0 | 0.9048 | 0.9233 | 0.9600 | 0.8000 | 0.0076 | 0.5985 | 0.0 | 0.8080 | 0.7798 | 0.9503 | 0.8595 |
| 0.3095 | 2.48 | 320 | 0.3111 | 0.5858 | 0.6494 | 0.8986 | 0.9655 | 0.0070 | 0.7517 | 0.0 | 0.9363 | 0.9210 | 0.9644 | 0.8850 | 0.0070 | 0.6808 | 0.0 | 0.8089 | 0.7790 | 0.9399 | 0.8851 |
| 0.3843 | 2.64 | 340 | 0.3076 | 0.6033 | 0.6589 | 0.9050 | 0.9648 | 0.0448 | 0.7872 | 0.0 | 0.9478 | 0.9052 | 0.9628 | 0.8786 | 0.0443 | 0.6973 | 0.0 | 0.8322 | 0.8187 | 0.9516 | 0.8924 |
| 0.4158 | 2.79 | 360 | 0.2985 | 0.5965 | 0.6553 | 0.9026 | 0.9557 | 0.0266 | 0.8130 | 0.0 | 0.9561 | 0.8939 | 0.9415 | 0.8820 | 0.0265 | 0.7245 | 0.0 | 0.8147 | 0.7892 | 0.9386 | 0.8903 |
| 0.3492 | 2.95 | 380 | 0.2709 | 0.6251 | 0.6863 | 0.9126 | 0.9524 | 0.1545 | 0.8818 | 0.0 | 0.9347 | 0.9218 | 0.9587 | 0.9003 | 0.1514 | 0.7515 | 0.0 | 0.8313 | 0.7895 | 0.9520 | 0.9026 |
| 0.2384 | 3.1 | 400 | 0.2531 | 0.6348 | 0.6950 | 0.9169 | 0.9541 | 0.1876 | 0.8731 | 0.0015 | 0.9446 | 0.9407 | 0.9632 | 0.9061 | 0.1830 | 0.7663 | 0.0015 | 0.8395 | 0.7944 | 0.9524 | 0.9071 |
| 0.2227 | 3.26 | 420 | 0.2772 | 0.6388 | 0.6939 | 0.9186 | 0.9648 | 0.1767 | 0.9028 | 0.0013 | 0.9230 | 0.9247 | 0.9637 | 0.9015 | 0.1735 | 0.7597 | 0.0013 | 0.8503 | 0.8312 | 0.9544 | 0.9087 |
| 0.6677 | 3.41 | 440 | 0.2861 | 0.6306 | 0.6874 | 0.9091 | 0.9640 | 0.1876 | 0.7780 | 0.0559 | 0.9635 | 0.9022 | 0.9610 | 0.8965 | 0.1838 | 0.7228 | 0.0558 | 0.8215 | 0.7825 | 0.9509 | 0.8996 |
| 0.3552 | 3.57 | 460 | 0.2795 | 0.6408 | 0.6945 | 0.9130 | 0.9641 | 0.2606 | 0.8234 | 0.0001 | 0.9487 | 0.8958 | 0.9690 | 0.8932 | 0.2536 | 0.7201 | 0.0001 | 0.8427 | 0.8204 | 0.9553 | 0.9032 |
| 0.3258 | 3.72 | 480 | 0.3075 | 0.6306 | 0.6891 | 0.9042 | 0.9730 | 0.2962 | 0.7091 | 0.0068 | 0.9628 | 0.9131 | 0.9625 | 0.8945 | 0.2617 | 0.6816 | 0.0068 | 0.8003 | 0.8150 | 0.9546 | 0.8939 |
| 0.4778 | 3.88 | 500 | 0.2449 | 0.6570 | 0.7137 | 0.9188 | 0.9583 | 0.2928 | 0.8807 | 0.0347 | 0.9383 | 0.9280 | 0.9633 | 0.9028 | 0.2768 | 0.7677 | 0.0347 | 0.8407 | 0.8223 | 0.9541 | 0.9106 |
| 0.4817 | 4.03 | 520 | 0.2365 | 0.6790 | 0.7381 | 0.9235 | 0.9611 | 0.4327 | 0.8871 | 0.0484 | 0.9393 | 0.9328 | 0.9651 | 0.9121 | 0.3866 | 0.7842 | 0.0483 | 0.8459 | 0.8230 | 0.9529 | 0.9165 |
| 0.3363 | 4.19 | 540 | 0.2273 | 0.6783 | 0.7315 | 0.9243 | 0.9635 | 0.4529 | 0.8805 | 0.0058 | 0.9546 | 0.8945 | 0.9685 | 0.9131 | 0.4101 | 0.7847 | 0.0058 | 0.8451 | 0.8339 | 0.9554 | 0.9163 |
| 0.4825 | 4.34 | 560 | 0.2615 | 0.6791 | 0.7482 | 0.9180 | 0.9406 | 0.5008 | 0.9124 | 0.0441 | 0.9206 | 0.9512 | 0.9675 | 0.8996 | 0.4385 | 0.7400 | 0.0441 | 0.8552 | 0.8198 | 0.9562 | 0.9119 |
| 0.3482 | 4.5 | 580 | 0.2336 | 0.6965 | 0.7537 | 0.9276 | 0.9695 | 0.4845 | 0.8611 | 0.1015 | 0.9492 | 0.9455 | 0.9648 | 0.9193 | 0.4258 | 0.7901 | 0.1006 | 0.8498 | 0.8349 | 0.9550 | 0.9215 |
| 0.5311 | 4.65 | 600 | 0.2592 | 0.6858 | 0.7484 | 0.9200 | 0.9477 | 0.4867 | 0.8797 | 0.0974 | 0.9463 | 0.9109 | 0.9703 | 0.9136 | 0.4329 | 0.7687 | 0.0970 | 0.8438 | 0.8232 | 0.9215 | 0.9142 |
| 0.3754 | 4.81 | 620 | 0.2345 | 0.7039 | 0.7629 | 0.9265 | 0.9641 | 0.5201 | 0.8692 | 0.1376 | 0.9387 | 0.9338 | 0.9769 | 0.9233 | 0.4557 | 0.7998 | 0.1369 | 0.8395 | 0.8404 | 0.9315 | 0.9211 |
| 0.236 | 4.96 | 640 | 0.2342 | 0.7061 | 0.7604 | 0.9268 | 0.9669 | 0.4754 | 0.8856 | 0.1790 | 0.9329 | 0.9039 | 0.9790 | 0.9218 | 0.4330 | 0.8088 | 0.1763 | 0.8374 | 0.8329 | 0.9327 | 0.9218 |
| 0.3496 | 5.12 | 660 | 0.2061 | 0.7264 | 0.7870 | 0.9313 | 0.9622 | 0.5779 | 0.9263 | 0.2075 | 0.9189 | 0.9404 | 0.9757 | 0.9243 | 0.5092 | 0.8247 | 0.2026 | 0.8525 | 0.8402 | 0.9315 | 0.9273 |
| 0.1729 | 5.27 | 680 | 0.2289 | 0.7086 | 0.7793 | 0.9245 | 0.9541 | 0.6032 | 0.8708 | 0.1744 | 0.9351 | 0.9392 | 0.9781 | 0.9202 | 0.4789 | 0.7981 | 0.1706 | 0.8377 | 0.8265 | 0.9282 | 0.9203 |
| 0.2636 | 5.43 | 700 | 0.2071 | 0.7739 | 0.8389 | 0.9335 | 0.9623 | 0.6448 | 0.8970 | 0.5348 | 0.9262 | 0.9396 | 0.9676 | 0.9245 | 0.5537 | 0.8125 | 0.4941 | 0.8551 | 0.8473 | 0.9304 | 0.9324 |
| 0.1594 | 5.58 | 720 | 0.2175 | 0.7447 | 0.8114 | 0.9284 | 0.9632 | 0.6275 | 0.8570 | 0.3850 | 0.9306 | 0.9383 | 0.9783 | 0.9172 | 0.5130 | 0.7999 | 0.3765 | 0.8467 | 0.8207 | 0.9389 | 0.9262 |
| 0.6799 | 5.74 | 740 | 0.1965 | 0.7704 | 0.8330 | 0.9379 | 0.9650 | 0.6576 | 0.9113 | 0.4469 | 0.9289 | 0.9430 | 0.9783 | 0.9303 | 0.5574 | 0.8353 | 0.4280 | 0.8695 | 0.8384 | 0.9338 | 0.9364 |
| 0.4955 | 5.89 | 760 | 0.2184 | 0.7480 | 0.8065 | 0.9318 | 0.9569 | 0.5099 | 0.9085 | 0.4344 | 0.9266 | 0.9255 | 0.9841 | 0.9146 | 0.4472 | 0.8109 | 0.4160 | 0.8654 | 0.8514 | 0.9308 | 0.9297 |
| 0.218 | 6.05 | 780 | 0.2025 | 0.7870 | 0.8508 | 0.9364 | 0.9678 | 0.5859 | 0.8690 | 0.6870 | 0.9368 | 0.9360 | 0.9732 | 0.9282 | 0.4918 | 0.8198 | 0.6437 | 0.8651 | 0.8324 | 0.9282 | 0.9355 |
| 0.4344 | 6.2 | 800 | 0.2128 | 0.7816 | 0.8399 | 0.9361 | 0.9620 | 0.6012 | 0.8908 | 0.5972 | 0.9446 | 0.9092 | 0.9743 | 0.9243 | 0.5193 | 0.8213 | 0.5626 | 0.8684 | 0.8520 | 0.9233 | 0.9350 |
| 0.5841 | 6.36 | 820 | 0.2412 | 0.7965 | 0.8614 | 0.9378 | 0.9587 | 0.6847 | 0.8888 | 0.6398 | 0.9360 | 0.9408 | 0.9808 | 0.9254 | 0.5838 | 0.8259 | 0.6002 | 0.8758 | 0.8440 | 0.9204 | 0.9371 |
| 0.3048 | 6.51 | 840 | 0.2336 | 0.7869 | 0.8580 | 0.9344 | 0.9576 | 0.6990 | 0.8928 | 0.6175 | 0.9281 | 0.9381 | 0.9728 | 0.9160 | 0.5612 | 0.8190 | 0.5724 | 0.8720 | 0.8484 | 0.9193 | 0.9337 |
| 0.2002 | 6.67 | 860 | 0.2373 | 0.7929 | 0.8605 | 0.9343 | 0.9512 | 0.6492 | 0.8691 | 0.6968 | 0.9424 | 0.9290 | 0.9861 | 0.9216 | 0.5520 | 0.8087 | 0.6401 | 0.8685 | 0.8437 | 0.9155 | 0.9337 |
| 0.2093 | 6.82 | 880 | 0.2335 | 0.7918 | 0.8528 | 0.9364 | 0.9649 | 0.6226 | 0.8682 | 0.6653 | 0.9414 | 0.9311 | 0.9758 | 0.9210 | 0.5272 | 0.8166 | 0.6269 | 0.8720 | 0.8550 | 0.9235 | 0.9355 |
| 0.1581 | 6.98 | 900 | 0.2279 | 0.7995 | 0.8786 | 0.9368 | 0.9509 | 0.7124 | 0.8959 | 0.7476 | 0.9347 | 0.9273 | 0.9812 | 0.9159 | 0.5266 | 0.8244 | 0.6672 | 0.8800 | 0.8627 | 0.9198 | 0.9367 |
| 0.1209 | 7.13 | 920 | 0.2432 | 0.7832 | 0.8475 | 0.9327 | 0.9548 | 0.5902 | 0.8897 | 0.6726 | 0.9357 | 0.9173 | 0.9725 | 0.9104 | 0.5047 | 0.8096 | 0.6187 | 0.8725 | 0.8463 | 0.9201 | 0.9319 |
| 0.1492 | 7.29 | 940 | 0.2345 | 0.7920 | 0.8609 | 0.9360 | 0.9603 | 0.6503 | 0.8872 | 0.6902 | 0.9383 | 0.9340 | 0.9659 | 0.9194 | 0.5196 | 0.8231 | 0.6344 | 0.8728 | 0.8571 | 0.9179 | 0.9354 |
### Framework versions
- Transformers 4.37.0
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.1
|