vit-msn-small-lateral_flow_ivalidation_train_test_4

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

• Loss: 0.3980
• Accuracy: 0.8938

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-07
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.3
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.8038	1.0	13	0.8368	0.4029
0.6874	2.0	26	0.8356	0.4029
0.6487	3.0	39	0.8336	0.3810
0.773	4.0	52	0.8307	0.3700
0.7002	5.0	65	0.8270	0.3480
0.6991	6.0	78	0.8223	0.3407
0.6809	7.0	91	0.8164	0.3480
0.7359	8.0	104	0.8093	0.3516
0.771	9.0	117	0.8017	0.3443
0.6855	10.0	130	0.7934	0.3443
0.6674	11.0	143	0.7851	0.3480
0.6296	12.0	156	0.7746	0.3810
0.5597	13.0	169	0.7643	0.3956
0.5636	14.0	182	0.7519	0.4066
0.5718	15.0	195	0.7382	0.4432
0.5527	16.0	208	0.7256	0.4579
0.5646	17.0	221	0.7115	0.5055
0.4843	18.0	234	0.6966	0.5275
0.492	19.0	247	0.6805	0.5788
0.4865	20.0	260	0.6630	0.6117
0.4198	21.0	273	0.6448	0.6410
0.4203	22.0	286	0.6280	0.6740
0.4547	23.0	299	0.6083	0.6923
0.3916	24.0	312	0.5909	0.7143
0.4329	25.0	325	0.5768	0.7289
0.4645	26.0	338	0.5629	0.7399
0.3376	27.0	351	0.5536	0.7436
0.4417	28.0	364	0.5417	0.7729
0.3908	29.0	377	0.5262	0.7619
0.3715	30.0	390	0.5130	0.7729
0.438	31.0	403	0.5059	0.7912
0.2937	32.0	416	0.4937	0.8022
0.2944	33.0	429	0.4871	0.8022
0.3474	34.0	442	0.4820	0.8059
0.2302	35.0	455	0.4776	0.7949
0.3543	36.0	468	0.4690	0.8022
0.3325	37.0	481	0.4640	0.8059
0.4004	38.0	494	0.4584	0.8095
0.3031	39.0	507	0.4548	0.8132
0.4862	40.0	520	0.4520	0.8095
0.2609	41.0	533	0.4498	0.8278
0.1859	42.0	546	0.4450	0.8462
0.2712	43.0	559	0.4408	0.8462
0.221	44.0	572	0.4387	0.8425
0.2328	45.0	585	0.4371	0.8498
0.3004	46.0	598	0.4339	0.8425
0.2036	47.0	611	0.4318	0.8462
0.1925	48.0	624	0.4299	0.8498
0.4543	49.0	637	0.4266	0.8498
0.4056	50.0	650	0.4251	0.8462
0.2326	51.0	663	0.4247	0.8498
0.327	52.0	676	0.4224	0.8571
0.2385	53.0	689	0.4193	0.8571
0.2876	54.0	702	0.4183	0.8571
0.2257	55.0	715	0.4162	0.8718
0.252	56.0	728	0.4150	0.8755
0.4299	57.0	741	0.4129	0.8645
0.3146	58.0	754	0.4124	0.8755
0.1993	59.0	767	0.4124	0.8755
0.2507	60.0	780	0.4118	0.8791
0.324	61.0	793	0.4101	0.8535
0.2303	62.0	806	0.4090	0.8718
0.2767	63.0	819	0.4072	0.8608
0.3318	64.0	832	0.4071	0.8681
0.1946	65.0	845	0.4064	0.8681
0.4204	66.0	858	0.4055	0.8608
0.3351	67.0	871	0.4031	0.8608
0.2772	68.0	884	0.4013	0.8645
0.2969	69.0	897	0.4000	0.8681
0.2755	70.0	910	0.4021	0.8901
0.2835	71.0	923	0.4005	0.8608
0.2487	72.0	936	0.3998	0.8608
0.2447	73.0	949	0.3987	0.8571
0.3512	74.0	962	0.3970	0.8718
0.2303	75.0	975	0.3975	0.8681
0.2271	76.0	988	0.3976	0.8791
0.2325	77.0	1001	0.3980	0.8938
0.2517	78.0	1014	0.3965	0.8901
0.2839	79.0	1027	0.3956	0.8938
0.1994	80.0	1040	0.3940	0.8828
0.4525	81.0	1053	0.3934	0.8864
0.2178	82.0	1066	0.3930	0.8828
0.2784	83.0	1079	0.3929	0.8901
0.1956	84.0	1092	0.3930	0.8901
0.2713	85.0	1105	0.3922	0.8828
0.2331	86.0	1118	0.3920	0.8828
0.3294	87.0	1131	0.3917	0.8864
0.2998	88.0	1144	0.3911	0.8864
0.3767	89.0	1157	0.3909	0.8864
0.3126	90.0	1170	0.3908	0.8828
0.2427	91.0	1183	0.3903	0.8791
0.2696	92.0	1196	0.3898	0.8828
0.2664	93.0	1209	0.3897	0.8828
0.3718	94.0	1222	0.3898	0.8828
0.2813	95.0	1235	0.3899	0.8828
0.3105	96.0	1248	0.3898	0.8828
0.2452	97.0	1261	0.3901	0.8828
0.2775	98.0	1274	0.3900	0.8828
0.3814	99.0	1287	0.3901	0.8828
0.2861	100.0	1300	0.3901	0.8828

Framework versions

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