efficientvit_b1.r224_in1k_rice-leaf-disease-augmented-v4_v5_fft

This model is a fine-tuned version of timm/efficientvit_b1.r224_in1k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2422
• Accuracy: 0.9262

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: 40
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1301	0.5	64	2.0282	0.1711
2.0177	1.0	128	1.8643	0.3121
1.7993	1.5	192	1.6252	0.4765
1.5315	2.0	256	1.3194	0.5940
1.1986	2.5	320	1.0574	0.6611
0.9772	3.0	384	0.8973	0.7349
0.7861	3.5	448	0.7841	0.7685
0.6808	4.0	512	0.6916	0.7886
0.5661	4.5	576	0.6288	0.8289
0.5042	5.0	640	0.5646	0.8456
0.4319	5.5	704	0.5420	0.8423
0.4057	6.0	768	0.5023	0.8591
0.3552	6.5	832	0.4788	0.8490
0.3201	7.0	896	0.4474	0.8591
0.3034	7.5	960	0.4353	0.8591
0.2745	8.0	1024	0.4315	0.8658
0.2588	8.5	1088	0.4181	0.8624
0.2601	9.0	1152	0.4154	0.8591
0.2374	9.5	1216	0.4123	0.8591
0.2464	10.0	1280	0.4054	0.8725
0.237	10.5	1344	0.4071	0.8591
0.2334	11.0	1408	0.3991	0.8658
0.236	11.5	1472	0.3978	0.8792
0.2267	12.0	1536	0.3735	0.8826
0.1949	12.5	1600	0.3543	0.8758
0.1715	13.0	1664	0.3416	0.8859
0.1381	13.5	1728	0.3150	0.8926
0.1274	14.0	1792	0.2979	0.8926
0.1026	14.5	1856	0.3080	0.8960
0.0888	15.0	1920	0.2842	0.8993
0.0722	15.5	1984	0.2812	0.9027
0.0723	16.0	2048	0.2804	0.8960
0.0615	16.5	2112	0.2734	0.9094
0.053	17.0	2176	0.2662	0.8993
0.0481	17.5	2240	0.2615	0.9060
0.0472	18.0	2304	0.2646	0.9027
0.0425	18.5	2368	0.2603	0.9060
0.041	19.0	2432	0.2545	0.9060
0.0379	19.5	2496	0.2536	0.9027
0.0427	20.0	2560	0.2567	0.9060
0.0403	20.5	2624	0.2510	0.9094
0.0374	21.0	2688	0.2571	0.9027
0.039	21.5	2752	0.2548	0.9027
0.0332	22.0	2816	0.2584	0.9027
0.0236	22.5	2880	0.2474	0.9161
0.0225	23.0	2944	0.2417	0.9161
0.016	23.5	3008	0.2442	0.9128
0.0135	24.0	3072	0.2411	0.9128
0.0112	24.5	3136	0.2378	0.9060
0.01	25.0	3200	0.2369	0.9161
0.008	25.5	3264	0.2391	0.9161
0.0069	26.0	3328	0.2333	0.9228
0.0062	26.5	3392	0.2339	0.9195
0.0065	27.0	3456	0.2366	0.9094
0.0056	27.5	3520	0.2390	0.9195
0.0057	28.0	3584	0.2387	0.9161
0.005	28.5	3648	0.2417	0.9128
0.005	29.0	3712	0.2420	0.9128
0.0056	29.5	3776	0.2457	0.9161
0.005	30.0	3840	0.2392	0.9195
0.0052	30.5	3904	0.2323	0.9161
0.0048	31.0	3968	0.2432	0.9195
0.0042	31.5	4032	0.2426	0.9161
0.0037	32.0	4096	0.2395	0.9195
0.0028	32.5	4160	0.2380	0.9228
0.0028	33.0	4224	0.2485	0.9195
0.0025	33.5	4288	0.2410	0.9262
0.0021	34.0	4352	0.2392	0.9228
0.0017	34.5	4416	0.2463	0.9228
0.0017	35.0	4480	0.2443	0.9161
0.0016	35.5	4544	0.2463	0.9262
0.0016	36.0	4608	0.2478	0.9161
0.0015	36.5	4672	0.2484	0.9195
0.0013	37.0	4736	0.2486	0.9228
0.0013	37.5	4800	0.2449	0.9161
0.0014	38.0	4864	0.2479	0.9228
0.0013	38.5	4928	0.2553	0.9161
0.0012	39.0	4992	0.2456	0.9228
0.0013	39.5	5056	0.2507	0.9161
0.0013	40.0	5120	0.2422	0.9262

Framework versions

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