metadata
license: apache-2.0
base_model: microsoft/swin-tiny-patch4-window7-224
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: Human-action-swin
results: []
Human-action-swin
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5231
- Accuracy: 0.8377
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: 12
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.8248 | 0.99 | 78 | 0.7299 | 0.7667 |
| 0.806 | 1.99 | 157 | 0.6541 | 0.7944 |
| 0.8544 | 3.0 | 236 | 0.6443 | 0.7996 |
| 0.8503 | 4.0 | 315 | 0.5965 | 0.8179 |
| 0.7538 | 4.99 | 393 | 0.5674 | 0.8282 |
| 0.6921 | 5.99 | 472 | 0.5941 | 0.8175 |
| 0.7214 | 7.0 | 551 | 0.5721 | 0.8246 |
| 0.6402 | 8.0 | 630 | 0.5433 | 0.8361 |
| 0.5899 | 8.99 | 708 | 0.5323 | 0.8425 |
| 0.6688 | 9.99 | 787 | 0.5213 | 0.8409 |
| 0.5948 | 11.0 | 866 | 0.5244 | 0.8385 |
| 0.5818 | 11.89 | 936 | 0.5231 | 0.8377 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2