|
--- |
|
license: apache-2.0 |
|
base_model: microsoft/swinv2-base-patch4-window8-256 |
|
tags: |
|
- image-classification |
|
- vision |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
model-index: |
|
- name: swinv2-base-panorama-IQA |
|
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-base-panorama-IQA |
|
|
|
This model is a fine-tuned version of [microsoft/swinv2-base-patch4-window8-256](https://huggingface.co/microsoft/swinv2-base-patch4-window8-256) on the isiqa-2019-hf dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0254 |
|
- Srocc: 0.1552 |
|
- Lcc: 0.2308 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 32 |
|
- seed: 10 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 50.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Srocc | Lcc | |
|
|:-------------:|:-------:|:----:|:---------------:|:-------:|:-------:| |
|
| No log | 0.8571 | 3 | 0.2656 | -0.1585 | -0.1388 | |
|
| No log | 2.0 | 7 | 0.0658 | -0.1810 | -0.1277 | |
|
| 0.2144 | 2.8571 | 10 | 0.1341 | -0.1684 | -0.1185 | |
|
| 0.2144 | 4.0 | 14 | 0.0562 | -0.2382 | -0.1379 | |
|
| 0.2144 | 4.8571 | 17 | 0.0616 | -0.1657 | -0.1428 | |
|
| 0.0575 | 6.0 | 21 | 0.0594 | -0.1501 | -0.1185 | |
|
| 0.0575 | 6.8571 | 24 | 0.0513 | -0.1204 | -0.1122 | |
|
| 0.0575 | 8.0 | 28 | 0.0417 | -0.0693 | -0.0987 | |
|
| 0.0201 | 8.8571 | 31 | 0.0406 | -0.0714 | -0.0813 | |
|
| 0.0201 | 10.0 | 35 | 0.0343 | -0.0236 | -0.0385 | |
|
| 0.0201 | 10.8571 | 38 | 0.0315 | 0.0074 | 0.0011 | |
|
| 0.0142 | 12.0 | 42 | 0.0309 | 0.0218 | 0.0326 | |
|
| 0.0142 | 12.8571 | 45 | 0.0310 | 0.0381 | 0.0466 | |
|
| 0.0142 | 14.0 | 49 | 0.0299 | 0.0535 | 0.0681 | |
|
| 0.0097 | 14.8571 | 52 | 0.0314 | 0.0604 | 0.0749 | |
|
| 0.0097 | 16.0 | 56 | 0.0288 | 0.0785 | 0.1049 | |
|
| 0.0097 | 16.8571 | 59 | 0.0283 | 0.0944 | 0.1269 | |
|
| 0.0083 | 18.0 | 63 | 0.0298 | 0.1022 | 0.1448 | |
|
| 0.0083 | 18.8571 | 66 | 0.0274 | 0.1118 | 0.1651 | |
|
| 0.0063 | 20.0 | 70 | 0.0286 | 0.1224 | 0.1703 | |
|
| 0.0063 | 20.8571 | 73 | 0.0283 | 0.1371 | 0.1833 | |
|
| 0.0063 | 22.0 | 77 | 0.0282 | 0.1317 | 0.1944 | |
|
| 0.0059 | 22.8571 | 80 | 0.0277 | 0.1382 | 0.2035 | |
|
| 0.0059 | 24.0 | 84 | 0.0270 | 0.1479 | 0.2146 | |
|
| 0.0059 | 24.8571 | 87 | 0.0263 | 0.1500 | 0.2197 | |
|
| 0.0046 | 26.0 | 91 | 0.0269 | 0.1364 | 0.2199 | |
|
| 0.0046 | 26.8571 | 94 | 0.0259 | 0.1406 | 0.2251 | |
|
| 0.0046 | 28.0 | 98 | 0.0254 | 0.1552 | 0.2308 | |
|
| 0.0039 | 28.8571 | 101 | 0.0267 | 0.1480 | 0.2261 | |
|
| 0.0039 | 30.0 | 105 | 0.0270 | 0.1489 | 0.2247 | |
|
| 0.0039 | 30.8571 | 108 | 0.0261 | 0.1576 | 0.2319 | |
|
| 0.0041 | 32.0 | 112 | 0.0268 | 0.1630 | 0.2311 | |
|
| 0.0041 | 32.8571 | 115 | 0.0282 | 0.1570 | 0.2257 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.42.3 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
|
|
