File size: 5,281 Bytes
85fda79 256b697 85fda79 |
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 |
---
license: other
tags:
- generated_from_trainer
base_model: nvidia/mit-b0
model-index:
- name: Aerial-Drone-Image-Segmentation
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. -->
# Aerial-Drone-Image-Segmentation
This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8852
- Mean Iou: 0.2994
- Mean Accuracy: 0.3923
- Overall Accuracy: 0.7774
## 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: 5e-05
- train_batch_size: 24
- eval_batch_size: 8
- 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 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|
| 2.7923 | 1.25 | 20 | 2.8338 | 0.0954 | 0.1626 | 0.5529 |
| 2.219 | 2.5 | 40 | 2.1391 | 0.1036 | 0.1666 | 0.5929 |
| 1.9451 | 3.75 | 60 | 1.7919 | 0.1154 | 0.1782 | 0.6129 |
| 1.7558 | 5.0 | 80 | 1.6767 | 0.1300 | 0.1961 | 0.6396 |
| 1.6381 | 6.25 | 100 | 1.5817 | 0.1383 | 0.2055 | 0.6550 |
| 1.5338 | 7.5 | 120 | 1.4816 | 0.1464 | 0.2140 | 0.6729 |
| 1.4478 | 8.75 | 140 | 1.4231 | 0.1529 | 0.2219 | 0.6823 |
| 1.361 | 10.0 | 160 | 1.3300 | 0.1637 | 0.2315 | 0.6975 |
| 1.306 | 11.25 | 180 | 1.3034 | 0.1737 | 0.2419 | 0.7060 |
| 1.2611 | 12.5 | 200 | 1.2692 | 0.1755 | 0.2450 | 0.7093 |
| 1.2317 | 13.75 | 220 | 1.2190 | 0.1821 | 0.2501 | 0.7145 |
| 1.1868 | 15.0 | 240 | 1.2063 | 0.1862 | 0.2539 | 0.7188 |
| 1.1628 | 16.25 | 260 | 1.1832 | 0.1909 | 0.2612 | 0.7234 |
| 1.1149 | 17.5 | 280 | 1.1368 | 0.2048 | 0.2739 | 0.7317 |
| 1.1009 | 18.75 | 300 | 1.1117 | 0.2232 | 0.2938 | 0.7387 |
| 1.0532 | 20.0 | 320 | 1.0923 | 0.2315 | 0.2997 | 0.7414 |
| 1.0464 | 21.25 | 340 | 1.0821 | 0.2408 | 0.3147 | 0.7480 |
| 1.0278 | 22.5 | 360 | 1.0541 | 0.2517 | 0.3277 | 0.7530 |
| 0.9945 | 23.75 | 380 | 1.0352 | 0.2612 | 0.3398 | 0.7573 |
| 0.9729 | 25.0 | 400 | 1.0207 | 0.2671 | 0.3511 | 0.7609 |
| 0.9527 | 26.25 | 420 | 1.0067 | 0.2684 | 0.3547 | 0.7609 |
| 0.9494 | 27.5 | 440 | 0.9870 | 0.2713 | 0.3548 | 0.7627 |
| 0.9287 | 28.75 | 460 | 0.9729 | 0.2745 | 0.3619 | 0.7640 |
| 0.9089 | 30.0 | 480 | 0.9561 | 0.2791 | 0.3640 | 0.7680 |
| 0.9064 | 31.25 | 500 | 0.9500 | 0.2799 | 0.3712 | 0.7672 |
| 0.8681 | 32.5 | 520 | 0.9397 | 0.2845 | 0.3749 | 0.7696 |
| 0.8677 | 33.75 | 540 | 0.9340 | 0.2835 | 0.3737 | 0.7692 |
| 0.8663 | 35.0 | 560 | 0.9243 | 0.2862 | 0.3755 | 0.7716 |
| 0.8629 | 36.25 | 580 | 0.9173 | 0.2869 | 0.3766 | 0.7719 |
| 0.8542 | 37.5 | 600 | 0.9112 | 0.2908 | 0.3810 | 0.7740 |
| 0.8391 | 38.75 | 620 | 0.9050 | 0.2904 | 0.3812 | 0.7734 |
| 0.8392 | 40.0 | 640 | 0.9027 | 0.2917 | 0.3818 | 0.7734 |
| 0.8306 | 41.25 | 660 | 0.8949 | 0.2941 | 0.3841 | 0.7755 |
| 0.8213 | 42.5 | 680 | 0.8936 | 0.2958 | 0.3875 | 0.7760 |
| 0.8406 | 43.75 | 700 | 0.8910 | 0.2964 | 0.3879 | 0.7763 |
| 0.8254 | 45.0 | 720 | 0.8889 | 0.2981 | 0.3897 | 0.7764 |
| 0.8202 | 46.25 | 740 | 0.8880 | 0.2985 | 0.3917 | 0.7767 |
| 0.8013 | 47.5 | 760 | 0.8891 | 0.2989 | 0.3923 | 0.7767 |
| 0.8188 | 48.75 | 780 | 0.8861 | 0.2994 | 0.3926 | 0.7772 |
| 0.8089 | 50.0 | 800 | 0.8852 | 0.2994 | 0.3923 | 0.7774 |
### Framework versions
- Transformers 4.38.1
- Pytorch 2.1.2
- Datasets 2.1.0
- Tokenizers 0.15.2
|