swin-base-patch4-window7-224_rice-leaf-disease-augmented-v3_fft

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7864
• Accuracy: 0.9088

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.344	1.0	128	0.4355	0.8339
0.1697	2.0	256	0.3696	0.8893
0.0375	3.0	384	0.4804	0.9055
0.0104	4.0	512	0.5652	0.9055
0.0015	5.0	640	0.5460	0.9186
0.0009	6.0	768	0.5613	0.9283
0.0019	7.0	896	0.7415	0.8958
0.0147	8.0	1024	0.6602	0.8958
0.0032	9.0	1152	0.7006	0.9153
0.0005	10.0	1280	0.6299	0.9283
0.0002	11.0	1408	0.6441	0.9218
0.0027	12.0	1536	0.6161	0.9218
0.0113	13.0	1664	0.5309	0.9153
0.0011	14.0	1792	0.7052	0.9186
0.0004	15.0	1920	0.6677	0.9316
0.0001	16.0	2048	0.8119	0.9055
0.027	17.0	2176	0.7111	0.9121
0.0028	18.0	2304	0.7691	0.9088
0.0002	19.0	2432	0.7777	0.9153
0.0003	20.0	2560	0.7864	0.9088

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0