|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: google/vit-base-patch16-224-in21k |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: vit-base-patch16-224-in21k-finetuned-biopsy |
|
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. --> |
|
|
|
# vit-base-patch16-224-in21k-finetuned-biopsy |
|
|
|
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1249 |
|
- Accuracy: 0.9682 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 128 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 50 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 1.1332 | 1.0 | 42 | 1.0712 | 0.5494 | |
|
| 0.6556 | 2.0 | 84 | 0.5742 | 0.8492 | |
|
| 0.3987 | 3.0 | 126 | 0.3950 | 0.8894 | |
|
| 0.2825 | 4.0 | 168 | 0.3924 | 0.8777 | |
|
| 0.3662 | 5.0 | 210 | 0.3622 | 0.8861 | |
|
| 0.2218 | 6.0 | 252 | 0.2706 | 0.9246 | |
|
| 0.2236 | 7.0 | 294 | 0.2283 | 0.9347 | |
|
| 0.2224 | 8.0 | 336 | 0.2367 | 0.9313 | |
|
| 0.1754 | 9.0 | 378 | 0.2139 | 0.9296 | |
|
| 0.1707 | 10.0 | 420 | 0.1829 | 0.9497 | |
|
| 0.1619 | 11.0 | 462 | 0.2172 | 0.9464 | |
|
| 0.1547 | 12.0 | 504 | 0.1960 | 0.9380 | |
|
| 0.1213 | 13.0 | 546 | 0.1484 | 0.9581 | |
|
| 0.1388 | 14.0 | 588 | 0.1689 | 0.9581 | |
|
| 0.1009 | 15.0 | 630 | 0.1494 | 0.9581 | |
|
| 0.124 | 16.0 | 672 | 0.1564 | 0.9581 | |
|
| 0.1078 | 17.0 | 714 | 0.1728 | 0.9514 | |
|
| 0.102 | 18.0 | 756 | 0.1669 | 0.9447 | |
|
| 0.1006 | 19.0 | 798 | 0.1405 | 0.9581 | |
|
| 0.0791 | 20.0 | 840 | 0.1179 | 0.9665 | |
|
| 0.0694 | 21.0 | 882 | 0.1424 | 0.9631 | |
|
| 0.0627 | 22.0 | 924 | 0.1224 | 0.9665 | |
|
| 0.0883 | 23.0 | 966 | 0.1602 | 0.9447 | |
|
| 0.074 | 24.0 | 1008 | 0.1315 | 0.9615 | |
|
| 0.0708 | 25.0 | 1050 | 0.1331 | 0.9631 | |
|
| 0.06 | 26.0 | 1092 | 0.1191 | 0.9665 | |
|
| 0.083 | 27.0 | 1134 | 0.1583 | 0.9531 | |
|
| 0.0584 | 28.0 | 1176 | 0.1348 | 0.9564 | |
|
| 0.0627 | 29.0 | 1218 | 0.1270 | 0.9564 | |
|
| 0.0627 | 30.0 | 1260 | 0.1411 | 0.9564 | |
|
| 0.038 | 31.0 | 1302 | 0.1208 | 0.9665 | |
|
| 0.0569 | 32.0 | 1344 | 0.1587 | 0.9514 | |
|
| 0.0502 | 33.0 | 1386 | 0.1501 | 0.9497 | |
|
| 0.0464 | 34.0 | 1428 | 0.1508 | 0.9615 | |
|
| 0.0317 | 35.0 | 1470 | 0.1309 | 0.9631 | |
|
| 0.0552 | 36.0 | 1512 | 0.1372 | 0.9598 | |
|
| 0.031 | 37.0 | 1554 | 0.1258 | 0.9598 | |
|
| 0.0383 | 38.0 | 1596 | 0.1249 | 0.9682 | |
|
| 0.036 | 39.0 | 1638 | 0.1312 | 0.9665 | |
|
| 0.0405 | 40.0 | 1680 | 0.1207 | 0.9665 | |
|
| 0.0343 | 41.0 | 1722 | 0.1233 | 0.9648 | |
|
| 0.0325 | 42.0 | 1764 | 0.1286 | 0.9631 | |
|
| 0.0293 | 43.0 | 1806 | 0.1135 | 0.9682 | |
|
| 0.0306 | 44.0 | 1848 | 0.1258 | 0.9615 | |
|
| 0.0267 | 45.0 | 1890 | 0.1261 | 0.9648 | |
|
| 0.0338 | 46.0 | 1932 | 0.1209 | 0.9665 | |
|
| 0.0213 | 47.0 | 1974 | 0.1157 | 0.9665 | |
|
| 0.0285 | 48.0 | 2016 | 0.1203 | 0.9631 | |
|
| 0.0287 | 49.0 | 2058 | 0.1240 | 0.9648 | |
|
| 0.0183 | 50.0 | 2100 | 0.1224 | 0.9665 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.1+cu121 |
|
- Datasets 3.0.1 |
|
- Tokenizers 0.19.1 |
|
|
