Boya1_SGD_1-e3_20Epoch_Deit-tiny-patch16_fold5

This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.7210
• Accuracy: 0.4071

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: 0.001
• train_batch_size: 16
• eval_batch_size: 16
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.4832	1.0	924	2.4588	0.1991
2.3933	2.0	1848	2.2987	0.2438
2.1601	3.0	2772	2.1633	0.2951
2.0165	4.0	3696	2.0604	0.3192
2.1447	5.0	4620	1.9841	0.3358
1.9413	6.0	5544	1.9278	0.3558
1.8772	7.0	6468	1.8829	0.3599
1.9245	8.0	7392	1.8510	0.3710
1.8852	9.0	8316	1.8234	0.3721
1.6188	10.0	9240	1.7999	0.3854
1.8268	11.0	10164	1.7826	0.3841
1.7105	12.0	11088	1.7673	0.3944
1.5952	13.0	12012	1.7552	0.3979
1.7598	14.0	12936	1.7451	0.3982
1.6066	15.0	13860	1.7368	0.4036
1.7336	16.0	14784	1.7320	0.4025
1.7215	17.0	15708	1.7264	0.4082
1.5878	18.0	16632	1.7236	0.4074
1.7164	19.0	17556	1.7216	0.4066
1.581	20.0	18480	1.7210	0.4071

Framework versions

• Transformers 4.40.1
• Pytorch 2.1.0
• Datasets 2.19.0
• Tokenizers 0.19.1