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: 0.4461
• Accuracy: 0.9295

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: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
1.8997	0.5	64	0.4966	1.5805
1.0296	1.0	128	0.8221	0.5906
0.4294	1.5	192	0.8557	0.3843
0.2385	2.0	256	0.9195	0.2437
0.1146	2.5	320	0.9094	0.2369
0.0637	3.0	384	0.9060	0.3286
0.0363	3.5	448	0.9195	0.3402
0.0241	4.0	512	0.8993	0.3084
0.0099	4.5	576	0.9161	0.2766
0.007	5.0	640	0.9396	0.2319
0.0056	5.5	704	0.9262	0.2576
0.0035	6.0	768	0.9228	0.2590
0.0045	6.5	832	0.9228	0.3203
0.0103	7.0	896	0.9228	0.3266
0.0112	7.5	960	0.9228	0.3506
0.0077	8.0	1024	0.9027	0.4189
0.0034	8.5	1088	0.9295	0.3010
0.0032	9.0	1152	0.9396	0.2840
0.0027	9.5	1216	0.9362	0.2497
0.0013	10.0	1280	0.9463	0.2358
0.002	10.5	1344	0.9497	0.2358
0.0012	11.0	1408	0.9262	0.3012
0.0085	11.5	1472	0.9262	0.4148
0.0087	12.0	1536	0.9228	0.3990
0.0072	12.5	1600	0.9497	0.2705
0.0113	13.0	1664	0.9295	0.3278
0.0046	13.5	1728	0.9463	0.2499
0.0006	14.0	1792	0.9463	0.3011
0.0017	14.5	1856	0.9362	0.2634
0.001	15.0	1920	0.9396	0.2585
0.0024	15.5	1984	0.4167	0.9262
0.0038	16.0	2048	0.3021	0.9497
0.0015	16.5	2112	0.4016	0.9329
0.0007	17.0	2176	0.2542	0.9430
0.0008	17.5	2240	0.3506	0.9362
0.0004	18.0	2304	0.3287	0.9362
0.0004	18.5	2368	0.2918	0.9497
0.0006	19.0	2432	0.2974	0.9463
0.0005	19.5	2496	0.2861	0.9530
0.0003	20.0	2560	0.2850	0.9530
0.0002	20.5	2624	0.2851	0.9530
0.0088	21.0	2688	0.4212	0.9161
0.0232	21.5	2752	0.5788	0.9094
0.0134	22.0	2816	0.3361	0.9295
0.0087	22.5	2880	0.4262	0.9195
0.012	23.0	2944	0.6413	0.9094
0.0075	23.5	3008	0.4673	0.9128
0.0045	24.0	3072	0.4002	0.9295
0.0019	24.5	3136	0.4725	0.9295
0.0023	25.0	3200	0.4702	0.9262
0.0002	25.5	3264	0.4981	0.9128
0.0002	26.0	3328	0.4631	0.9228
0.0005	26.5	3392	0.4899	0.9094
0.0004	27.0	3456	0.4424	0.9295
0.0014	27.5	3520	0.4532	0.9329
0.0001	28.0	3584	0.4459	0.9329
0.0001	28.5	3648	0.4449	0.9295
0.0001	29.0	3712	0.4455	0.9262
0.0001	29.5	3776	0.4463	0.9262
0.0005	30.0	3840	0.4461	0.9295

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.1