convnextv2-base-1k-224_rice-leaf-disease-augmented-v4_v5_fft

This model is a fine-tuned version of facebook/convnextv2-base-1k-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3987
• Accuracy: 0.9362

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
2.0296	0.5	64	1.8920	0.4799
1.5755	1.0	128	1.2129	0.7651
0.9059	1.5	192	0.6750	0.8289
0.4659	2.0	256	0.3395	0.9128
0.2039	2.5	320	0.2663	0.9362
0.1195	3.0	384	0.2652	0.9329
0.0556	3.5	448	0.2046	0.9396
0.0271	4.0	512	0.2504	0.9396
0.0096	4.5	576	0.2902	0.9329
0.0058	5.0	640	0.2492	0.9362
0.0032	5.5	704	0.2680	0.9329
0.0027	6.0	768	0.2695	0.9430
0.0026	6.5	832	0.2680	0.9396
0.0023	7.0	896	0.2648	0.9430
0.0014	7.5	960	0.2618	0.9463
0.001	8.0	1024	0.2782	0.9396
0.0008	8.5	1088	0.2870	0.9430
0.0007	9.0	1152	0.2964	0.9396
0.0006	9.5	1216	0.2964	0.9430
0.0006	10.0	1280	0.3007	0.9430
0.0006	10.5	1344	0.3016	0.9430
0.0005	11.0	1408	0.3014	0.9430
0.0005	11.5	1472	0.3089	0.9396
0.0005	12.0	1536	0.3132	0.9396
0.0004	12.5	1600	0.3153	0.9362
0.0003	13.0	1664	0.3209	0.9396
0.0003	13.5	1728	0.3287	0.9396
0.0003	14.0	1792	0.3284	0.9396
0.0003	14.5	1856	0.3342	0.9396
0.0003	15.0	1920	0.3333	0.9396
0.0003	15.5	1984	0.3340	0.9396
0.0003	16.0	2048	0.3342	0.9396
0.0003	16.5	2112	0.3338	0.9396
0.0002	17.0	2176	0.3515	0.9362
0.0002	17.5	2240	0.3447	0.9396
0.0002	18.0	2304	0.3551	0.9396
0.0002	18.5	2368	0.3551	0.9396
0.0002	19.0	2432	0.3570	0.9396
0.0002	19.5	2496	0.3606	0.9396
0.0002	20.0	2560	0.3593	0.9396
0.0002	20.5	2624	0.3592	0.9396
0.0002	21.0	2688	0.3612	0.9396
0.0001	21.5	2752	0.3726	0.9396
0.0001	22.0	2816	0.3690	0.9396
0.0001	22.5	2880	0.3777	0.9362
0.0001	23.0	2944	0.3725	0.9396
0.0001	23.5	3008	0.3751	0.9396
0.0001	24.0	3072	0.3782	0.9396
0.0001	24.5	3136	0.3805	0.9396
0.0001	25.0	3200	0.3804	0.9396
0.0001	25.5	3264	0.3824	0.9396
0.0001	26.0	3328	0.3891	0.9362
0.0001	26.5	3392	0.3921	0.9329
0.0001	27.0	3456	0.3896	0.9362
0.0001	27.5	3520	0.3978	0.9362
0.0001	28.0	3584	0.3952	0.9362
0.0001	28.5	3648	0.3994	0.9362
0.0001	29.0	3712	0.3989	0.9362
0.0001	29.5	3776	0.3985	0.9362
0.0001	30.0	3840	0.3987	0.9362

Framework versions

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