Boya1_RMSProp_1-e5_10Epoch_swinv2-large-patch4_fold3
This model is a fine-tuned version of microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.8858
- Accuracy: 0.6835
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: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.8473 | 1.0 | 1846 | 1.1167 | 0.6219 |
0.8985 | 2.0 | 3692 | 0.9760 | 0.6565 |
0.687 | 3.0 | 5538 | 0.9578 | 0.6795 |
0.5802 | 4.0 | 7384 | 0.9833 | 0.6995 |
0.359 | 5.0 | 9230 | 1.1314 | 0.6857 |
0.4094 | 6.0 | 11076 | 1.2902 | 0.6914 |
0.397 | 7.0 | 12922 | 1.5573 | 0.6833 |
0.1913 | 8.0 | 14768 | 1.7605 | 0.6811 |
0.0583 | 9.0 | 16614 | 1.8168 | 0.6900 |
0.1804 | 10.0 | 18460 | 1.8858 | 0.6835 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 10