efficientvit_m4.r224_in1k_rice-leaf-disease-augmented-v4_v5_fft

This model is a fine-tuned version of timm/efficientvit_m4.r224_in1k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4767
• Accuracy: 0.8658

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
2.0834	0.5	64	2.0788	0.1577
2.0669	1.0	128	2.0488	0.1711
2.0293	1.5	192	2.0006	0.2785
1.9774	2.0	256	1.9270	0.3792
1.8929	2.5	320	1.8456	0.4597
1.8113	3.0	384	1.7539	0.5235
1.7181	3.5	448	1.6894	0.5570
1.6738	4.0	512	1.6412	0.5839
1.6197	4.5	576	1.6008	0.6141
1.5808	5.0	640	1.5555	0.6208
1.5563	5.5	704	1.5397	0.6174
1.5374	6.0	768	1.5281	0.6409
1.5426	6.5	832	1.5175	0.6309
1.5093	7.0	896	1.4774	0.6376
1.4656	7.5	960	1.4045	0.6443
1.3943	8.0	1024	1.3379	0.6577
1.3244	8.5	1088	1.2769	0.6879
1.2782	9.0	1152	1.2230	0.6946
1.2293	9.5	1216	1.2051	0.6980
1.1952	10.0	1280	1.1664	0.7114
1.1759	10.5	1344	1.1598	0.7215
1.1638	11.0	1408	1.1507	0.7248
1.1612	11.5	1472	1.1345	0.7282
1.1221	12.0	1536	1.0794	0.7383
1.0554	12.5	1600	1.0158	0.7584
0.9903	13.0	1664	0.9986	0.7651
0.9281	13.5	1728	0.9145	0.7718
0.9074	14.0	1792	0.8825	0.7919
0.86	14.5	1856	0.8671	0.7919
0.8338	15.0	1920	0.8936	0.7785
0.8242	15.5	1984	0.8743	0.7886
0.8269	16.0	2048	0.8563	0.7886
0.8116	16.5	2112	0.8288	0.7987
0.7591	17.0	2176	0.7901	0.7987
0.7088	17.5	2240	0.7543	0.8087
0.6646	18.0	2304	0.7242	0.8221
0.6291	18.5	2368	0.7118	0.8188
0.6018	19.0	2432	0.6792	0.8255
0.5824	19.5	2496	0.6707	0.8289
0.5794	20.0	2560	0.6707	0.8322
0.5722	20.5	2624	0.6688	0.8356
0.5643	21.0	2688	0.6503	0.8255
0.5286	21.5	2752	0.6360	0.8221
0.5141	22.0	2816	0.6289	0.8289
0.4557	22.5	2880	0.5956	0.8255
0.4438	23.0	2944	0.5746	0.8389
0.4084	23.5	3008	0.5673	0.8490
0.4007	24.0	3072	0.5566	0.8456
0.3777	24.5	3136	0.5547	0.8456
0.3824	25.0	3200	0.5598	0.8523
0.3807	25.5	3264	0.5528	0.8523
0.3549	26.0	3328	0.5500	0.8490
0.3357	26.5	3392	0.5255	0.8523
0.3206	27.0	3456	0.5039	0.8523
0.2941	27.5	3520	0.4959	0.8725
0.2754	28.0	3584	0.4910	0.8523
0.2536	28.5	3648	0.4837	0.8658
0.2583	29.0	3712	0.4753	0.8658
0.2519	29.5	3776	0.4844	0.8792
0.2458	30.0	3840	0.4767	0.8658

Framework versions

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