mobilevit-small_rice-leaf-disease-augmented-v4_fft

This model is a fine-tuned version of apple/mobilevit-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4089
• 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
2.0561	0.5	64	0.2886	2.0213
1.9819	1.0	128	0.5503	1.8788
1.771	1.5	192	0.6107	1.5291
1.3911	2.0	256	0.7349	1.0706
1.0026	2.5	320	0.8054	0.7560
0.7657	3.0	384	0.8356	0.6180
0.6082	3.5	448	0.8389	0.5422
0.5313	4.0	512	0.8523	0.4946
0.4623	4.5	576	0.8758	0.4512
0.4212	5.0	640	0.8792	0.4322
0.4025	5.5	704	0.8893	0.4259
0.3892	6.0	768	0.8859	0.4238
0.3959	6.5	832	0.8859	0.4083
0.3279	7.0	896	0.8826	0.3750
0.2793	7.5	960	0.8993	0.3350
0.222	8.0	1024	0.8960	0.3208
0.1862	8.5	1088	0.8993	0.3128
0.1717	9.0	1152	0.9027	0.3049
0.1408	9.5	1216	0.9027	0.3010
0.1507	10.0	1280	0.9161	0.3240
0.1369	10.5	1344	0.9060	0.3063
0.1389	11.0	1408	0.9060	0.3045
0.1199	11.5	1472	0.9094	0.3062
0.1003	12.0	1536	0.9128	0.3131
0.0756	12.5	1600	0.9228	0.3002
0.0636	13.0	1664	0.9128	0.3177
0.058	13.5	1728	0.9228	0.3143
0.0566	14.0	1792	0.9195	0.3136
0.0516	14.5	1856	0.9161	0.3447
0.0426	15.0	1920	0.9228	0.2911
0.0513	15.5	1984	0.3028	0.9228
0.0447	16.0	2048	0.3328	0.9195
0.0332	16.5	2112	0.3193	0.9262
0.0358	17.0	2176	0.3385	0.9161
0.0343	17.5	2240	0.3297	0.9295
0.0291	18.0	2304	0.3518	0.9161
0.0287	18.5	2368	0.3224	0.9195
0.0197	19.0	2432	0.3099	0.9228
0.0223	19.5	2496	0.3305	0.9295
0.0282	20.0	2560	0.3378	0.9161
0.0231	20.5	2624	0.3077	0.9228
0.0251	21.0	2688	0.3520	0.9161
0.021	21.5	2752	0.3506	0.9228
0.0222	22.0	2816	0.3561	0.9128
0.016	22.5	2880	0.3482	0.9195
0.0163	23.0	2944	0.3429	0.9228
0.0114	23.5	3008	0.3839	0.9329
0.0106	24.0	3072	0.4066	0.9262
0.0111	24.5	3136	0.4003	0.9329
0.009	25.0	3200	0.4000	0.9262
0.0088	25.5	3264	0.3667	0.9228
0.0057	26.0	3328	0.3587	0.9195
0.0073	26.5	3392	0.3686	0.9329
0.0085	27.0	3456	0.3676	0.9195
0.0087	27.5	3520	0.4251	0.9262
0.0061	28.0	3584	0.3879	0.9195
0.0062	28.5	3648	0.3865	0.9195
0.0068	29.0	3712	0.3943	0.9262
0.0092	29.5	3776	0.4064	0.9228
0.0078	30.0	3840	0.4089	0.9295

Framework versions

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