|
--- |
|
license: apache-2.0 |
|
base_model: microsoft/swinv2-tiny-patch4-window8-256 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: swinv2-tiny-patch4-window8-256-dmae-va-U5-42B |
|
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. --> |
|
|
|
# swinv2-tiny-patch4-window8-256-dmae-va-U5-42B |
|
|
|
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0633 |
|
- Accuracy: 0.6333 |
|
|
|
## 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: 4e-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: 42 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| No log | 0.9 | 7 | 7.8663 | 0.1167 | |
|
| 6.936 | 1.94 | 15 | 7.7572 | 0.1167 | |
|
| 6.936 | 2.97 | 23 | 7.1789 | 0.1167 | |
|
| 6.7016 | 4.0 | 31 | 5.9142 | 0.1167 | |
|
| 5.5418 | 4.9 | 38 | 4.6065 | 0.1167 | |
|
| 5.5418 | 5.94 | 46 | 3.2783 | 0.1167 | |
|
| 3.6439 | 6.97 | 54 | 2.1984 | 0.1167 | |
|
| 2.2477 | 8.0 | 62 | 1.5631 | 0.45 | |
|
| 2.2477 | 8.9 | 69 | 1.3158 | 0.45 | |
|
| 1.5076 | 9.94 | 77 | 1.3665 | 0.3333 | |
|
| 1.3865 | 10.97 | 85 | 1.3561 | 0.35 | |
|
| 1.3865 | 12.0 | 93 | 1.2673 | 0.4667 | |
|
| 1.3436 | 12.9 | 100 | 1.5020 | 0.2833 | |
|
| 1.3187 | 13.94 | 108 | 1.3018 | 0.4667 | |
|
| 1.3187 | 14.97 | 116 | 1.2582 | 0.4833 | |
|
| 1.2132 | 16.0 | 124 | 1.2250 | 0.5 | |
|
| 1.2132 | 16.9 | 131 | 1.2071 | 0.5167 | |
|
| 1.2041 | 17.94 | 139 | 1.1806 | 0.5167 | |
|
| 1.1756 | 18.97 | 147 | 1.1720 | 0.55 | |
|
| 1.1756 | 20.0 | 155 | 1.1319 | 0.5333 | |
|
| 1.1107 | 20.9 | 162 | 1.0851 | 0.5833 | |
|
| 1.0651 | 21.94 | 170 | 1.0554 | 0.5833 | |
|
| 1.0651 | 22.97 | 178 | 1.0580 | 0.5 | |
|
| 1.0419 | 24.0 | 186 | 1.1471 | 0.5833 | |
|
| 0.9804 | 24.9 | 193 | 1.0633 | 0.6333 | |
|
| 0.9804 | 25.94 | 201 | 1.0597 | 0.5833 | |
|
| 0.9195 | 26.97 | 209 | 0.9563 | 0.6333 | |
|
| 0.9053 | 28.0 | 217 | 0.9839 | 0.5167 | |
|
| 0.9053 | 28.9 | 224 | 0.9914 | 0.5167 | |
|
| 0.8645 | 29.94 | 232 | 0.9520 | 0.5833 | |
|
| 0.8139 | 30.97 | 240 | 0.9573 | 0.5333 | |
|
| 0.8139 | 32.0 | 248 | 0.9510 | 0.5167 | |
|
| 0.8151 | 32.9 | 255 | 0.9469 | 0.5167 | |
|
| 0.8151 | 33.94 | 263 | 0.9872 | 0.5333 | |
|
| 0.7837 | 34.97 | 271 | 0.9279 | 0.5833 | |
|
| 0.7659 | 36.0 | 279 | 0.9174 | 0.5667 | |
|
| 0.7659 | 36.9 | 286 | 0.9347 | 0.5667 | |
|
| 0.7835 | 37.94 | 294 | 0.9372 | 0.5667 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.2 |
|
- Pytorch 2.1.2+cu118 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
|
|
