vit-msn-small-lateral_flow_ivalidation_train_test_1

This model is a fine-tuned version of Melo1512/vit-msn-small-lateral_flow_ivalidation_train_test_1 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.3874
• Accuracy: 0.9084

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-06
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 10
• total_train_batch_size: 640
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: reduce_lr_on_plateau
• lr_scheduler_warmup_ratio: 0.5
• num_epochs: 100
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3744	0.7692	1	0.3874	0.9084
0.3574	1.5385	2	0.3874	0.9084
0.3627	2.3077	3	0.3875	0.9084
0.3367	3.8462	5	0.3875	0.9084
0.3752	4.6154	6	0.3875	0.9084
0.3696	5.3846	7	0.3876	0.9084
0.3563	6.9231	9	0.3877	0.9084
0.3244	7.6923	10	0.3877	0.9084
0.3727	8.4615	11	0.3878	0.9084
0.3395	10.0	13	0.3879	0.9084
0.35	10.7692	14	0.3880	0.9084
0.3527	11.5385	15	0.3881	0.9084
0.3539	12.3077	16	0.3881	0.9084
0.348	13.8462	18	0.3883	0.9084
0.3584	14.6154	19	0.3884	0.9084
0.3491	15.3846	20	0.3884	0.9084
0.3603	16.9231	22	0.3886	0.9084
0.382	17.6923	23	0.3887	0.9084
0.3567	18.4615	24	0.3888	0.9084
0.3397	20.0	26	0.3889	0.9084
0.3682	20.7692	27	0.3890	0.9084
0.3336	21.5385	28	0.3891	0.9084
0.3399	22.3077	29	0.3892	0.9084
0.3656	23.8462	31	0.3893	0.9084
0.348	24.6154	32	0.3894	0.9084
0.3466	25.3846	33	0.3895	0.9084
0.3614	26.9231	35	0.3897	0.9084
0.3522	27.6923	36	0.3898	0.9084
0.3506	28.4615	37	0.3898	0.9084
0.3623	30.0	39	0.3900	0.9084
0.3562	30.7692	40	0.3901	0.9084
0.3515	31.5385	41	0.3902	0.9084
0.3744	32.3077	42	0.3903	0.9084
0.3382	33.8462	44	0.3904	0.9084
0.3467	34.6154	45	0.3905	0.9084
0.3713	35.3846	46	0.3906	0.9084
0.3653	36.9231	48	0.3908	0.9084
0.3359	37.6923	49	0.3909	0.9084
0.3745	38.4615	50	0.3910	0.9084
0.3594	40.0	52	0.3911	0.9084
0.3567	40.7692	53	0.3912	0.9084
0.3332	41.5385	54	0.3913	0.9084
0.3424	42.3077	55	0.3914	0.9084
0.3485	43.8462	57	0.3915	0.9084
0.3795	44.6154	58	0.3916	0.9084
0.321	45.3846	59	0.3917	0.9084
0.3314	46.9231	61	0.3918	0.9084
0.3582	47.6923	62	0.3919	0.9084
0.3355	48.4615	63	0.3920	0.9084
0.3688	50.0	65	0.3922	0.9084
0.3414	50.7692	66	0.3923	0.9084
0.3292	51.5385	67	0.3923	0.9084
0.3622	52.3077	68	0.3924	0.9048
0.3446	53.8462	70	0.3926	0.9048
0.3678	54.6154	71	0.3927	0.9011
0.3632	55.3846	72	0.3927	0.9011
0.3459	56.9231	74	0.3929	0.9011
0.3435	57.6923	75	0.3929	0.9011
0.3461	58.4615	76	0.3930	0.9011
0.3423	60.0	78	0.3931	0.9011
0.3632	60.7692	79	0.3932	0.8974
0.3536	61.5385	80	0.3933	0.8974
0.3556	62.3077	81	0.3934	0.8974
0.3391	63.8462	83	0.3935	0.8974
0.3482	64.6154	84	0.3936	0.8974
0.3339	65.3846	85	0.3937	0.8974
0.3438	66.9231	87	0.3938	0.8974
0.3198	67.6923	88	0.3939	0.8974
0.3399	68.4615	89	0.3939	0.8974
0.3388	70.0	91	0.3941	0.8974
0.3519	70.7692	92	0.3942	0.8974
0.3445	71.5385	93	0.3942	0.8974
0.3423	72.3077	94	0.3943	0.8974
0.3392	73.8462	96	0.3945	0.8974
0.3576	74.6154	97	0.3945	0.8974
0.3526	75.3846	98	0.3946	0.8974
0.3714	76.9231	100	0.3947	0.8974

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1