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: 0.2911
• Accuracy: 0.9228

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
2.0561	0.5	64	2.0213	0.2886
1.9819	1.0	128	1.8788	0.5503
1.771	1.5	192	1.5291	0.6107
1.3911	2.0	256	1.0706	0.7349
1.0026	2.5	320	0.7560	0.8054
0.7657	3.0	384	0.6180	0.8356
0.6082	3.5	448	0.5422	0.8389
0.5313	4.0	512	0.4946	0.8523
0.4623	4.5	576	0.4512	0.8758
0.4212	5.0	640	0.4322	0.8792
0.4025	5.5	704	0.4259	0.8893
0.3892	6.0	768	0.4238	0.8859
0.3959	6.5	832	0.4083	0.8859
0.3279	7.0	896	0.3750	0.8826
0.2793	7.5	960	0.3350	0.8993
0.222	8.0	1024	0.3208	0.8960
0.1862	8.5	1088	0.3128	0.8993
0.1717	9.0	1152	0.3049	0.9027
0.1408	9.5	1216	0.3010	0.9027
0.1507	10.0	1280	0.3240	0.9161
0.1369	10.5	1344	0.3063	0.9060
0.1389	11.0	1408	0.3045	0.9060
0.1199	11.5	1472	0.3062	0.9094
0.1003	12.0	1536	0.3131	0.9128
0.0756	12.5	1600	0.3002	0.9228
0.0636	13.0	1664	0.3177	0.9128
0.058	13.5	1728	0.3143	0.9228
0.0566	14.0	1792	0.3136	0.9195
0.0516	14.5	1856	0.3447	0.9161
0.0426	15.0	1920	0.2911	0.9228

Framework versions

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