|
--- |
|
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: 0.0246 |
|
- Srocc: 0.0896 |
|
- Lcc: 0.1773 |
|
|
|
## 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 | 0.2685 | -0.1661 | -0.1400 | |
|
| No log | 2.0 | 7 | 0.0675 | -0.2071 | -0.1319 | |
|
| 0.223 | 2.8571 | 10 | 0.1380 | -0.1972 | -0.1144 | |
|
| 0.223 | 4.0 | 14 | 0.0639 | -0.2362 | -0.1162 | |
|
| 0.223 | 4.8571 | 17 | 0.0601 | -0.1760 | -0.1097 | |
|
| 0.0607 | 6.0 | 21 | 0.0627 | -0.1290 | -0.0852 | |
|
| 0.0607 | 6.8571 | 24 | 0.0543 | -0.1050 | -0.0791 | |
|
| 0.0607 | 8.0 | 28 | 0.0408 | -0.0683 | -0.0702 | |
|
| 0.0212 | 8.8571 | 31 | 0.0419 | -0.0692 | -0.0567 | |
|
| 0.0212 | 10.0 | 35 | 0.0343 | -0.0370 | -0.0274 | |
|
| 0.0212 | 10.8571 | 38 | 0.0307 | -0.0339 | -0.0013 | |
|
| 0.0168 | 12.0 | 42 | 0.0299 | -0.0281 | 0.0233 | |
|
| 0.0168 | 12.8571 | 45 | 0.0300 | -0.0428 | 0.0326 | |
|
| 0.0168 | 14.0 | 49 | 0.0286 | -0.0238 | 0.0517 | |
|
| 0.0143 | 14.8571 | 52 | 0.0283 | -0.0186 | 0.0601 | |
|
| 0.0143 | 16.0 | 56 | 0.0273 | -0.0024 | 0.0868 | |
|
| 0.0143 | 16.8571 | 59 | 0.0257 | 0.0283 | 0.1119 | |
|
| 0.013 | 18.0 | 63 | 0.0247 | 0.0542 | 0.1404 | |
|
| 0.013 | 18.8571 | 66 | 0.0247 | 0.0703 | 0.1533 | |
|
| 0.0111 | 20.0 | 70 | 0.0246 | 0.0800 | 0.1670 | |
|
| 0.0111 | 20.8571 | 73 | 0.0246 | 0.0896 | 0.1773 | |
|
| 0.0111 | 22.0 | 77 | 0.0257 | 0.0998 | 0.1835 | |
|
| 0.0104 | 22.8571 | 80 | 0.0255 | 0.1017 | 0.1943 | |
|
| 0.0104 | 24.0 | 84 | 0.0255 | 0.1149 | 0.2085 | |
|
| 0.0104 | 24.8571 | 87 | 0.0255 | 0.1245 | 0.2155 | |
|
| 0.0088 | 26.0 | 91 | 0.0262 | 0.1319 | 0.2258 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.42.3 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
|
|
