|
--- |
|
license: apache-2.0 |
|
tags: |
|
- image-classification |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: modeversion2_m7_e8 |
|
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. --> |
|
|
|
# modeversion2_m7_e8 |
|
|
|
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the sudo-s/herbier_mesuem7 dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1060 |
|
- Accuracy: 0.9761 |
|
|
|
## 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: 0.0002 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 8 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:| |
|
| 4.0231 | 0.06 | 100 | 3.8568 | 0.1883 | |
|
| 3.3863 | 0.12 | 200 | 3.2510 | 0.2596 | |
|
| 2.6187 | 0.18 | 300 | 2.6243 | 0.3882 | |
|
| 2.3097 | 0.23 | 400 | 2.2189 | 0.4527 | |
|
| 1.9016 | 0.29 | 500 | 1.9495 | 0.5244 | |
|
| 1.7478 | 0.35 | 600 | 1.6609 | 0.6091 | |
|
| 1.2345 | 0.41 | 700 | 1.4335 | 0.6426 | |
|
| 1.4129 | 0.47 | 800 | 1.3001 | 0.6752 | |
|
| 1.1722 | 0.53 | 900 | 1.2030 | 0.6785 | |
|
| 1.0808 | 0.59 | 1000 | 1.0051 | 0.7273 | |
|
| 0.8814 | 0.64 | 1100 | 1.0715 | 0.7063 | |
|
| 0.9831 | 0.7 | 1200 | 0.9283 | 0.7334 | |
|
| 0.8118 | 0.76 | 1300 | 0.8525 | 0.7631 | |
|
| 0.7203 | 0.82 | 1400 | 0.7849 | 0.7756 | |
|
| 0.8881 | 0.88 | 1500 | 0.8786 | 0.7487 | |
|
| 0.6407 | 0.94 | 1600 | 0.6896 | 0.8000 | |
|
| 0.7574 | 1.0 | 1700 | 0.7314 | 0.7754 | |
|
| 0.6063 | 1.06 | 1800 | 0.6312 | 0.8068 | |
|
| 0.4797 | 1.11 | 1900 | 0.5792 | 0.8296 | |
|
| 0.4973 | 1.17 | 2000 | 0.5846 | 0.8221 | |
|
| 0.4432 | 1.23 | 2100 | 0.7057 | 0.7905 | |
|
| 0.5518 | 1.29 | 2200 | 0.5621 | 0.8304 | |
|
| 0.3256 | 1.35 | 2300 | 0.5890 | 0.8143 | |
|
| 0.4284 | 1.41 | 2400 | 0.5204 | 0.8485 | |
|
| 0.3702 | 1.47 | 2500 | 0.5699 | 0.8256 | |
|
| 0.2858 | 1.52 | 2600 | 0.5815 | 0.8287 | |
|
| 0.3706 | 1.58 | 2700 | 0.4615 | 0.8571 | |
|
| 0.3484 | 1.64 | 2800 | 0.4812 | 0.8518 | |
|
| 0.2865 | 1.7 | 2900 | 0.4285 | 0.8638 | |
|
| 0.4474 | 1.76 | 3000 | 0.5217 | 0.8377 | |
|
| 0.2101 | 1.82 | 3100 | 0.4478 | 0.8589 | |
|
| 0.3545 | 1.88 | 3200 | 0.4444 | 0.8612 | |
|
| 0.2728 | 1.93 | 3300 | 0.4213 | 0.8645 | |
|
| 0.3525 | 1.99 | 3400 | 0.3551 | 0.8848 | |
|
| 0.0936 | 2.05 | 3500 | 0.4074 | 0.8748 | |
|
| 0.2118 | 2.11 | 3600 | 0.4089 | 0.8812 | |
|
| 0.2744 | 2.17 | 3700 | 0.3534 | 0.8894 | |
|
| 0.211 | 2.23 | 3800 | 0.4422 | 0.8599 | |
|
| 0.1684 | 2.29 | 3900 | 0.3705 | 0.8858 | |
|
| 0.1885 | 2.34 | 4000 | 0.3651 | 0.8862 | |
|
| 0.249 | 2.4 | 4100 | 0.4234 | 0.8687 | |
|
| 0.1485 | 2.46 | 4200 | 0.3784 | 0.8798 | |
|
| 0.1188 | 2.52 | 4300 | 0.3589 | 0.8873 | |
|
| 0.1274 | 2.58 | 4400 | 0.3570 | 0.8917 | |
|
| 0.2206 | 2.64 | 4500 | 0.3377 | 0.8920 | |
|
| 0.1287 | 2.7 | 4600 | 0.3170 | 0.9023 | |
|
| 0.1805 | 2.75 | 4700 | 0.3469 | 0.8934 | |
|
| 0.1505 | 2.81 | 4800 | 0.4258 | 0.8757 | |
|
| 0.1592 | 2.87 | 4900 | 0.3415 | 0.8948 | |
|
| 0.1297 | 2.93 | 5000 | 0.3168 | 0.9028 | |
|
| 0.1284 | 2.99 | 5100 | 0.3060 | 0.9089 | |
|
| 0.0833 | 3.05 | 5200 | 0.2610 | 0.9207 | |
|
| 0.0334 | 3.11 | 5300 | 0.2766 | 0.9197 | |
|
| 0.0847 | 3.17 | 5400 | 0.3366 | 0.9016 | |
|
| 0.1112 | 3.22 | 5500 | 0.3098 | 0.9079 | |
|
| 0.0477 | 3.28 | 5600 | 0.3385 | 0.9041 | |
|
| 0.0419 | 3.34 | 5700 | 0.2944 | 0.9139 | |
|
| 0.0827 | 3.4 | 5800 | 0.2715 | 0.9239 | |
|
| 0.0659 | 3.46 | 5900 | 0.2695 | 0.9230 | |
|
| 0.0244 | 3.52 | 6000 | 0.3050 | 0.9147 | |
|
| 0.0883 | 3.58 | 6100 | 0.2862 | 0.9203 | |
|
| 0.0527 | 3.63 | 6200 | 0.2383 | 0.9319 | |
|
| 0.0828 | 3.69 | 6300 | 0.2984 | 0.9182 | |
|
| 0.0678 | 3.75 | 6400 | 0.2135 | 0.9436 | |
|
| 0.0492 | 3.81 | 6500 | 0.2605 | 0.9296 | |
|
| 0.0374 | 3.87 | 6600 | 0.2192 | 0.9380 | |
|
| 0.1846 | 3.93 | 6700 | 0.2804 | 0.9187 | |
|
| 0.0557 | 3.99 | 6800 | 0.2599 | 0.9253 | |
|
| 0.0127 | 4.04 | 6900 | 0.2412 | 0.9336 | |
|
| 0.0203 | 4.1 | 7000 | 0.2214 | 0.9415 | |
|
| 0.0272 | 4.16 | 7100 | 0.2322 | 0.9356 | |
|
| 0.066 | 4.22 | 7200 | 0.2643 | 0.9325 | |
|
| 0.0628 | 4.28 | 7300 | 0.2170 | 0.9406 | |
|
| 0.0108 | 4.34 | 7400 | 0.2388 | 0.9405 | |
|
| 0.026 | 4.4 | 7500 | 0.2533 | 0.9372 | |
|
| 0.0401 | 4.45 | 7600 | 0.2407 | 0.9358 | |
|
| 0.0493 | 4.51 | 7700 | 0.2213 | 0.9415 | |
|
| 0.0951 | 4.57 | 7800 | 0.3016 | 0.9237 | |
|
| 0.0017 | 4.63 | 7900 | 0.2183 | 0.9448 | |
|
| 0.0561 | 4.69 | 8000 | 0.1962 | 0.9492 | |
|
| 0.0063 | 4.75 | 8100 | 0.1868 | 0.9522 | |
|
| 0.0054 | 4.81 | 8200 | 0.2068 | 0.9459 | |
|
| 0.0519 | 4.87 | 8300 | 0.2141 | 0.9429 | |
|
| 0.027 | 4.92 | 8400 | 0.2138 | 0.9438 | |
|
| 0.0034 | 4.98 | 8500 | 0.1774 | 0.9529 | |
|
| 0.0096 | 5.04 | 8600 | 0.1778 | 0.9512 | |
|
| 0.0011 | 5.1 | 8700 | 0.1854 | 0.9512 | |
|
| 0.0195 | 5.16 | 8800 | 0.1914 | 0.9483 | |
|
| 0.0245 | 5.22 | 8900 | 0.2156 | 0.9471 | |
|
| 0.0055 | 5.28 | 9000 | 0.1640 | 0.9574 | |
|
| 0.0166 | 5.33 | 9100 | 0.1770 | 0.9568 | |
|
| 0.0217 | 5.39 | 9200 | 0.2011 | 0.9479 | |
|
| 0.0017 | 5.45 | 9300 | 0.2210 | 0.9462 | |
|
| 0.0161 | 5.51 | 9400 | 0.1510 | 0.9621 | |
|
| 0.0193 | 5.57 | 9500 | 0.1643 | 0.9586 | |
|
| 0.0121 | 5.63 | 9600 | 0.1716 | 0.9535 | |
|
| 0.0146 | 5.69 | 9700 | 0.1720 | 0.9554 | |
|
| 0.0071 | 5.74 | 9800 | 0.1831 | 0.9541 | |
|
| 0.0018 | 5.8 | 9900 | 0.2076 | 0.9485 | |
|
| 0.0007 | 5.86 | 10000 | 0.1636 | 0.9599 | |
|
| 0.0005 | 5.92 | 10100 | 0.1625 | 0.9602 | |
|
| 0.0277 | 5.98 | 10200 | 0.1874 | 0.9546 | |
|
| 0.0005 | 6.04 | 10300 | 0.1790 | 0.9579 | |
|
| 0.0012 | 6.1 | 10400 | 0.1840 | 0.9544 | |
|
| 0.0431 | 6.15 | 10500 | 0.1571 | 0.9628 | |
|
| 0.0332 | 6.21 | 10600 | 0.1599 | 0.9591 | |
|
| 0.0014 | 6.27 | 10700 | 0.1493 | 0.9632 | |
|
| 0.0014 | 6.33 | 10800 | 0.1366 | 0.9661 | |
|
| 0.0006 | 6.39 | 10900 | 0.1582 | 0.9609 | |
|
| 0.0005 | 6.45 | 11000 | 0.1704 | 0.9589 | |
|
| 0.0004 | 6.51 | 11100 | 0.1376 | 0.9671 | |
|
| 0.0755 | 6.57 | 11200 | 0.1375 | 0.9654 | |
|
| 0.0002 | 6.62 | 11300 | 0.1361 | 0.9661 | |
|
| 0.0006 | 6.68 | 11400 | 0.1323 | 0.9675 | |
|
| 0.0009 | 6.74 | 11500 | 0.1239 | 0.9692 | |
|
| 0.0004 | 6.8 | 11600 | 0.1514 | 0.9631 | |
|
| 0.0002 | 6.86 | 11700 | 0.1386 | 0.9664 | |
|
| 0.0004 | 6.92 | 11800 | 0.1368 | 0.9659 | |
|
| 0.0004 | 6.98 | 11900 | 0.1276 | 0.9684 | |
|
| 0.0002 | 7.03 | 12000 | 0.1171 | 0.9712 | |
|
| 0.0002 | 7.09 | 12100 | 0.1142 | 0.9711 | |
|
| 0.0001 | 7.15 | 12200 | 0.1183 | 0.9727 | |
|
| 0.0002 | 7.21 | 12300 | 0.1167 | 0.9732 | |
|
| 0.0002 | 7.27 | 12400 | 0.1143 | 0.9737 | |
|
| 0.0001 | 7.33 | 12500 | 0.1129 | 0.9737 | |
|
| 0.0002 | 7.39 | 12600 | 0.1116 | 0.9742 | |
|
| 0.0002 | 7.44 | 12700 | 0.1126 | 0.9745 | |
|
| 0.0002 | 7.5 | 12800 | 0.1111 | 0.9748 | |
|
| 0.0002 | 7.56 | 12900 | 0.1102 | 0.9747 | |
|
| 0.0001 | 7.62 | 13000 | 0.1094 | 0.9747 | |
|
| 0.0001 | 7.68 | 13100 | 0.1086 | 0.9742 | |
|
| 0.0001 | 7.74 | 13200 | 0.1079 | 0.9748 | |
|
| 0.0002 | 7.8 | 13300 | 0.1062 | 0.9754 | |
|
| 0.0002 | 7.85 | 13400 | 0.1068 | 0.9757 | |
|
| 0.0001 | 7.91 | 13500 | 0.1061 | 0.9762 | |
|
| 0.0001 | 7.97 | 13600 | 0.1060 | 0.9761 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.20.1 |
|
- Pytorch 1.12.0 |
|
- Datasets 2.3.2 |
|
- Tokenizers 0.12.1 |
|
|
