This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.5357
Accuracy: 0.8696

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: 5.5e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 80

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.97	7	1.3872	0.1304
1.3844	1.93	14	1.3826	0.1739
1.3552	2.9	21	1.3500	0.2391
1.3552	4.0	29	1.2528	0.2174
1.2458	4.97	36	1.1474	0.2391
1.0668	5.93	43	1.1376	0.3913
0.9335	6.9	50	1.0063	0.4348
0.9335	8.0	58	0.9238	0.5870
0.8059	8.97	65	0.8241	0.8043
0.6774	9.93	72	0.7625	0.7826
0.6774	10.9	79	0.7096	0.8043
0.5346	12.0	87	0.6368	0.8261
0.4427	12.97	94	0.5741	0.8261
0.3557	13.93	101	0.5441	0.8261
0.3557	14.9	108	0.5258	0.8478
0.2637	16.0	116	0.5430	0.8261
0.2356	16.97	123	0.5773	0.8261
0.1844	17.93	130	0.7222	0.7391
0.1844	18.9	137	0.6537	0.7826
0.1765	20.0	145	0.5458	0.8043
0.1362	20.97	152	0.5777	0.8478
0.1362	21.93	159	0.6256	0.7826
0.1467	22.9	166	0.7330	0.7826
0.1614	24.0	174	0.7743	0.7609
0.1246	24.97	181	0.5763	0.8261
0.1246	25.93	188	0.5994	0.8261
0.1058	26.9	195	0.6926	0.8043
0.0943	28.0	203	0.6406	0.8478
0.1	28.97	210	0.6940	0.7609
0.1	29.93	217	0.6193	0.8261
0.0865	30.9	224	0.5357	0.8696
0.0852	32.0	232	0.8015	0.7826
0.0852	32.97	239	0.6680	0.8478
0.0721	33.93	246	0.8469	0.7826
0.0749	34.9	253	0.6682	0.8261
0.0876	36.0	261	0.7474	0.8261
0.0876	36.97	268	0.6501	0.8696
0.0677	37.93	275	0.6918	0.8043
0.0574	38.9	282	0.7001	0.8478
0.0573	40.0	290	0.7119	0.8261
0.0573	40.97	297	0.8317	0.8043
0.0663	41.93	304	0.7456	0.8043
0.0685	42.9	311	0.7242	0.8261
0.0685	44.0	319	0.6971	0.8043
0.0431	44.97	326	0.7439	0.8261
0.0529	45.93	333	0.8210	0.8043
0.0698	46.9	340	0.7114	0.8043
0.0698	48.0	348	0.6985	0.8478
0.054	48.97	355	0.8860	0.8261
0.0528	49.93	362	0.8942	0.8043
0.0528	50.9	369	0.9411	0.8043
0.0478	52.0	377	0.8705	0.7826
0.041	52.97	384	0.8130	0.8261
0.0321	53.93	391	0.7682	0.8043
0.0321	54.9	398	0.8696	0.7826
0.0318	56.0	406	0.9598	0.8043
0.0416	56.97	413	0.7291	0.8261
0.0477	57.93	420	0.6869	0.8478
0.0477	58.9	427	0.7055	0.8478
0.0307	60.0	435	0.7415	0.8478
0.032	60.97	442	0.8024	0.8261
0.032	61.93	449	0.7856	0.8478
0.0232	62.9	456	0.7251	0.8043
0.0267	64.0	464	0.7231	0.8478
0.0456	64.97	471	0.7326	0.8696
0.0456	65.93	478	0.7300	0.8696
0.0359	66.9	485	0.7293	0.8696
0.0199	68.0	493	0.7361	0.8696
0.0235	68.97	500	0.7362	0.8696
0.0235	69.93	507	0.7513	0.8696
0.0368	70.9	514	0.7513	0.8696
0.0254	72.0	522	0.7586	0.8696
0.0254	72.97	529	0.7574	0.8696
0.029	73.93	536	0.7685	0.8478
0.0302	74.9	543	0.7653	0.8478
0.0305	76.0	551	0.7637	0.8261
0.0305	76.97	558	0.7645	0.8478
0.0301	77.24	560	0.7645	0.8478

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0

Augusto777/vit-base-patch16-224-ve-U15-b-80

vit-base-patch16-224-ve-U15-b-80

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Augusto777/
vit-base-patch16-224-ve-U15-b-80