efficientformer_l1.snap_dist_in1k_rice-leaf-disease-augmented-v4_v5_fft

This model is a fine-tuned version of timm/efficientformer_l1.snap_dist_in1k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4269
• Accuracy: 0.9060

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	Validation Loss	Accuracy
2.0721	0.5	64	2.0503	0.1812
1.9705	1.0	128	1.8812	0.4128
1.7172	1.5	192	1.5160	0.5570
1.3261	2.0	256	1.0740	0.6779
0.8641	2.5	320	0.7390	0.7718
0.5772	3.0	384	0.5308	0.8289
0.3856	3.5	448	0.4667	0.8389
0.2981	4.0	512	0.4052	0.8523
0.213	4.5	576	0.3778	0.8591
0.1795	5.0	640	0.3505	0.8792
0.1435	5.5	704	0.3455	0.8859
0.1309	6.0	768	0.3440	0.8826
0.1243	6.5	832	0.3309	0.8893
0.1142	7.0	896	0.3252	0.8758
0.0837	7.5	960	0.3259	0.8893
0.059	8.0	1024	0.3085	0.9060
0.0296	8.5	1088	0.2963	0.8960
0.0208	9.0	1152	0.3109	0.8993
0.0092	9.5	1216	0.3261	0.9027
0.0098	10.0	1280	0.3265	0.8960
0.0056	10.5	1344	0.3280	0.9027
0.0068	11.0	1408	0.3289	0.9060
0.005	11.5	1472	0.3590	0.8893
0.0058	12.0	1536	0.3379	0.9060
0.0025	12.5	1600	0.3744	0.9094
0.0026	13.0	1664	0.3851	0.9060
0.0016	13.5	1728	0.3950	0.9027
0.0011	14.0	1792	0.3766	0.9128
0.0007	14.5	1856	0.3729	0.9161
0.0011	15.0	1920	0.3591	0.9027
0.0006	15.5	1984	0.3769	0.8993
0.0006	16.0	2048	0.3660	0.9094
0.0005	16.5	2112	0.3687	0.9195
0.0006	17.0	2176	0.3933	0.9060
0.0006	17.5	2240	0.3849	0.9128
0.0006	18.0	2304	0.4178	0.9027
0.0009	18.5	2368	0.4092	0.9027
0.0002	19.0	2432	0.4117	0.9094
0.0003	19.5	2496	0.4075	0.9060
0.0003	20.0	2560	0.4116	0.9094
0.0002	20.5	2624	0.3974	0.9094
0.0004	21.0	2688	0.4266	0.8993
0.0004	21.5	2752	0.4172	0.9128
0.0004	22.0	2816	0.4450	0.9027
0.0003	22.5	2880	0.4505	0.9060
0.0002	23.0	2944	0.4213	0.9027
0.0001	23.5	3008	0.4285	0.9027
0.0001	24.0	3072	0.4368	0.9027
0.0002	24.5	3136	0.4330	0.9060
0.0002	25.0	3200	0.4294	0.9060
0.0001	25.5	3264	0.4395	0.9027
0.0006	26.0	3328	0.4304	0.9060
0.0001	26.5	3392	0.4203	0.9161
0.0001	27.0	3456	0.4403	0.9094
0.0002	27.5	3520	0.4447	0.9027
0.0001	28.0	3584	0.4348	0.9094
0.0001	28.5	3648	0.4200	0.9094
0.0001	29.0	3712	0.4340	0.9094
0.0001	29.5	3776	0.4402	0.9094
0.0001	30.0	3840	0.4269	0.9060

Framework versions

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