vit-msn-small-lateral_flow_ivalidation_green

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: 0.2917
Accuracy: 0.9602

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
gradient_accumulation_steps: 4
total_train_batch_size: 256
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.9231	6	0.5034	0.9239
0.704	2.0	13	0.4524	0.9410
0.704	2.9231	19	0.2633	0.9594
0.505	4.0	26	0.4748	0.8092
0.4456	4.9231	32	0.2917	0.9602
0.4456	6.0	39	0.2621	0.9222
0.3908	6.9231	45	0.4519	0.8191
0.3628	8.0	52	0.4093	0.8623
0.3628	8.9231	58	0.2705	0.9354
0.372	10.0	65	0.4137	0.8546
0.36	10.9231	71	0.3493	0.8815
0.36	12.0	78	0.2190	0.9457
0.36	12.9231	84	0.3190	0.9033
0.3363	14.0	91	0.3380	0.8948
0.3363	14.9231	97	0.3342	0.8982
0.327	16.0	104	0.4212	0.8328
0.3257	16.9231	110	0.5167	0.7844
0.3257	18.0	117	0.5848	0.7275
0.3175	18.9231	123	0.4091	0.8336
0.3377	20.0	130	0.2838	0.9162
0.3377	20.9231	136	0.6106	0.7263
0.3129	22.0	143	0.6295	0.7164
0.3129	22.9231	149	0.7898	0.5932
0.3138	24.0	156	0.9408	0.4846
0.3106	24.9231	162	0.3485	0.8832
0.3106	26.0	169	0.5201	0.7866
0.3157	26.9231	175	0.7210	0.6672
0.2896	28.0	182	0.7981	0.6330
0.2896	28.9231	188	0.7667	0.6429
0.2867	30.0	195	0.7687	0.6544
0.2786	30.9231	201	1.1714	0.5210
0.2786	32.0	208	1.1744	0.4273
0.2823	32.9231	214	0.9260	0.5445
0.2864	34.0	221	0.7140	0.6920
0.2864	34.9231	227	0.6098	0.7331
0.2707	36.0	234	0.6993	0.6784
0.2921	36.9231	240	0.8719	0.6176
0.2921	38.0	247	0.8337	0.6061
0.2849	38.9231	253	0.4396	0.8255
0.2657	40.0	260	1.0982	0.5017
0.2657	40.9231	266	1.0934	0.5175
0.2659	42.0	273	0.8629	0.6369
0.2659	42.9231	279	1.4602	0.4140
0.2645	44.0	286	1.9095	0.3422
0.2424	44.9231	292	1.2180	0.4397
0.2424	46.0	299	0.7686	0.6424
0.2495	46.9231	305	0.9899	0.5796
0.2454	48.0	312	1.0291	0.5535
0.2454	48.9231	318	0.7534	0.6822
0.2473	50.0	325	0.6591	0.7092
0.2716	50.9231	331	0.5840	0.7455
0.2716	52.0	338	1.2430	0.4765
0.234	52.9231	344	1.2993	0.5145
0.2482	54.0	351	0.6042	0.7173
0.2482	54.9231	357	0.8892	0.6027
0.2339	56.0	364	1.8546	0.3161
0.2461	56.9231	370	1.0859	0.5359
0.2461	58.0	377	0.8690	0.6176
0.2395	58.9231	383	0.7557	0.6694
0.2159	60.0	390	1.0534	0.5701
0.2159	60.9231	396	0.9856	0.5813
0.2309	62.0	403	1.0000	0.5500
0.2309	62.9231	409	1.1940	0.5180
0.2117	64.0	416	1.1581	0.5154
0.2307	64.9231	422	0.9987	0.5338
0.2307	66.0	429	1.0850	0.5415
0.2068	66.9231	435	0.9428	0.6014
0.2126	68.0	442	1.2380	0.5115
0.2126	68.9231	448	0.9993	0.5860
0.2176	70.0	455	1.1910	0.5021
0.2096	70.9231	461	1.2468	0.5120
0.2096	72.0	468	0.7588	0.6920
0.2092	72.9231	474	0.9003	0.6309
0.1968	74.0	481	1.1697	0.5646
0.1968	74.9231	487	0.8789	0.6446
0.2027	76.0	494	1.1352	0.5599
0.1965	76.9231	500	1.0836	0.5599
0.1965	78.0	507	1.0188	0.5902
0.2267	78.9231	513	1.0287	0.5975
0.1967	80.0	520	0.8465	0.6544
0.1967	80.9231	526	1.1881	0.5470
0.1842	82.0	533	1.2352	0.5368
0.1842	82.9231	539	1.1064	0.5701
0.1952	84.0	546	0.8088	0.6608
0.1873	84.9231	552	0.9342	0.6086
0.1873	86.0	559	0.9807	0.6056
0.185	86.9231	565	1.0165	0.5898
0.1993	88.0	572	1.1511	0.5475
0.1993	88.9231	578	1.1766	0.5406
0.1707	90.0	585	1.1201	0.5663
0.1852	90.9231	591	1.1162	0.5701
0.1852	92.0	598	1.1273	0.5680
0.1904	92.3077	600	1.1280	0.5680

Framework versions

Transformers 4.44.2
Pytorch 2.4.1+cu121
Datasets 3.2.0
Tokenizers 0.19.1

Melo1512/vit-msn-small-lateral_flow_ivalidation_green

vit-msn-small-lateral_flow_ivalidation_green

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions