swin-tiny-patch4-window7-224_rice-leaf-disease-augmented-v4_fft

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: 0.2585
• Accuracy: 0.9396

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_steps: 256
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.8997	0.5	64	1.5805	0.4966
1.0296	1.0	128	0.5906	0.8221
0.4294	1.5	192	0.3843	0.8557
0.2385	2.0	256	0.2437	0.9195
0.1146	2.5	320	0.2369	0.9094
0.0637	3.0	384	0.3286	0.9060
0.0363	3.5	448	0.3402	0.9195
0.0241	4.0	512	0.3084	0.8993
0.0099	4.5	576	0.2766	0.9161
0.007	5.0	640	0.2319	0.9396
0.0056	5.5	704	0.2576	0.9262
0.0035	6.0	768	0.2590	0.9228
0.0045	6.5	832	0.3203	0.9228
0.0103	7.0	896	0.3266	0.9228
0.0112	7.5	960	0.3506	0.9228
0.0077	8.0	1024	0.4189	0.9027
0.0034	8.5	1088	0.3010	0.9295
0.0032	9.0	1152	0.2840	0.9396
0.0027	9.5	1216	0.2497	0.9362
0.0013	10.0	1280	0.2358	0.9463
0.002	10.5	1344	0.2358	0.9497
0.0012	11.0	1408	0.3012	0.9262
0.0085	11.5	1472	0.4148	0.9262
0.0087	12.0	1536	0.3990	0.9228
0.0072	12.5	1600	0.2705	0.9497
0.0113	13.0	1664	0.3278	0.9295
0.0046	13.5	1728	0.2499	0.9463
0.0006	14.0	1792	0.3011	0.9463
0.0017	14.5	1856	0.2634	0.9362
0.001	15.0	1920	0.2585	0.9396

Framework versions

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