vit-base-butterflies

sasha/vit-base-butterflies

vit-base-butterflies is a image classification model from sasha. Use it when you need a label for an image. It is set up for transformers. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

2% of all-time downloads

All-time downloads

910

Public

Parameters

86M

29.9 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors344 MB · 100%

At a glance

Task: Image Classification
Library: transformers
License: apache-2.0
Model type: vit
Access: Public
Created: Oct 20, 2025
Updated: Oct 21, 2025
SHA: 80632ed9

Base models

google/vit-base-patch16-224-in21k

Datasets

imagefolder

Spaces

Model card

From the Hugging Face model README

vit-base-butterflies

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 11.2307
Accuracy: 0.0034

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: 0.0002
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
4.036	0.0722	100	6.5320	0.0
3.203	0.1443	200	6.9035	0.0
2.5923	0.2165	300	7.2400	0.0018
2.2606	0.2886	400	7.7074	0.0023
1.9191	0.3608	500	7.7788	0.0029
1.7853	0.4329	600	7.9713	0.0036
1.7658	0.5051	700	8.1260	0.0038
1.4418	0.5772	800	8.3715	0.0034
1.4053	0.6494	900	8.3429	0.0032
1.3018	0.7215	1000	8.3995	0.0036
1.177	0.7937	1100	8.7005	0.0036
1.2162	0.8658	1200	9.0641	0.0022
1.2031	0.9380	1300	9.0564	0.0034
0.8526	1.0101	1400	9.1382	0.0034
0.7236	1.0823	1500	9.2894	0.0032
0.705	1.1544	1600	9.6825	0.0018
1.0321	1.2266	1700	9.0059	0.0040
0.6998	1.2987	1800	9.7627	0.0036
0.6525	1.3709	1900	9.6668	0.0032
0.7351	1.4430	2000	9.5337	0.0032
0.6919	1.5152	2100	10.3186	0.0018
0.6289	1.5873	2200	9.9239	0.0029
0.7725	1.6595	2300	10.0128	0.0034
0.7261	1.7316	2400	10.3410	0.0031
0.5443	1.8038	2500	9.7265	0.0032
0.5457	1.8759	2600	10.0234	0.0032
0.6076	1.9481	2700	9.8331	0.0038
0.4203	2.0202	2800	10.2837	0.0034
0.3361	2.0924	2900	9.7588	0.0027
0.3477	2.1645	3000	9.7134	0.0036
0.2561	2.2367	3100	10.5175	0.0005
0.2844	2.3088	3200	9.9866	0.0022
0.3417	2.3810	3300	10.2690	0.0027
0.321	2.4531	3400	10.6317	0.0036
0.2614	2.5253	3500	10.8217	0.0032
0.3315	2.5974	3600	10.6792	0.0034
0.3768	2.6696	3700	10.2700	0.0031
0.337	2.7417	3800	10.4120	0.0027
0.4078	2.8139	3900	10.5717	0.0018
0.3645	2.8860	4000	10.4502	0.0034
0.3641	2.9582	4100	10.4831	0.0036
0.2849	3.0303	4200	10.4292	0.0032
0.1721	3.1025	4300	10.5754	0.0032
0.1726	3.1746	4400	10.8464	0.0032
0.2392	3.2468	4500	10.5688	0.0034
0.1985	3.3189	4600	11.1911	0.0034
0.1706	3.3911	4700	10.6939	0.0034
0.1761	3.4632	4800	10.7841	0.0034
0.1707	3.5354	4900	10.8447	0.0034
0.0998	3.6075	5000	10.9400	0.0036
0.1283	3.6797	5100	11.0827	0.0032
0.2074	3.7518	5200	11.2594	0.0032
0.2001	3.8240	5300	10.8110	0.0032
0.1822	3.8961	5400	11.2189	0.0036
0.2206	3.9683	5500	11.2027	0.0034
0.1137	4.0404	5600	11.2689	0.0031
0.1276	4.1126	5700	11.2888	0.0032
0.1044	4.1847	5800	11.0306	0.0032
0.1623	4.2569	5900	10.8323	0.0038
0.1596	4.3290	6000	10.9011	0.0032
0.1345	4.4012	6100	11.0262	0.0034
0.1414	4.4733	6200	11.1875	0.0034
0.1513	4.5455	6300	11.2167	0.0034
0.1444	4.6176	6400	11.2639	0.0034
0.1772	4.6898	6500	11.2617	0.0034
0.1297	4.7619	6600	11.2941	0.0034
0.1204	4.8341	6700	11.1882	0.0034
0.1548	4.9062	6800	11.2022	0.0034
0.1051	4.9784	6900	11.2307	0.0034

Framework versions

Transformers 4.57.1
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.1

Task: Image Classification
Library: transformers
Type: vit
License: apache-2.0
Created: Oct 20, 2025
Updated: Oct 21, 2025