vit-ytr-demo-v3

sharren/vit-ytr-demo-v3

vit-ytr-demo-v3 is a image classification model from sharren. 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

286

Public

Parameters

85.8M

43.3 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors343 MB · 100%

At a glance

Task: Image Classification
Library: transformers
License: apache-2.0
Model type: vit
Access: Public
Created: Feb 23, 2024
Updated: Feb 24, 2024
SHA: acc1bff1

Base models

google/vit-base-patch16-224-in21k

Model card

From the Hugging Face model README

vit-ytr-demo-v3

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

Loss: 0.5923
Accuracy: 0.8712

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 15
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7103	0.31	100	0.7668	0.7464
0.8059	0.62	200	0.7692	0.7374
0.8991	0.93	300	0.6572	0.7639
0.7064	1.25	400	0.6511	0.7718
0.6537	1.56	500	0.7332	0.7609
0.5387	1.87	600	0.6083	0.7743
0.5508	2.18	700	0.5943	0.7938
0.5509	2.49	800	0.5373	0.8118
0.5104	2.8	900	0.5612	0.7973
0.3535	3.12	1000	0.5531	0.8113
0.2151	3.43	1100	0.6024	0.7958
0.3758	3.74	1200	0.5272	0.8198
0.4745	4.05	1300	0.4738	0.8318
0.4279	4.36	1400	0.4848	0.8248
0.4029	4.67	1500	0.4958	0.8243
0.3744	4.98	1600	0.5503	0.8078
0.4498	5.3	1700	0.5024	0.8382
0.3793	5.61	1800	0.4926	0.8283
0.3461	5.92	1900	0.4819	0.8367
0.3781	6.23	2000	0.5399	0.8183
0.3122	6.54	2100	0.5061	0.8397
0.2431	6.85	2200	0.4794	0.8372
0.2484	7.17	2300	0.5713	0.8342
0.284	7.48	2400	0.5165	0.8308
0.1815	7.79	2500	0.4973	0.8407
0.0996	8.1	2600	0.5248	0.8467
0.2379	8.41	2700	0.5535	0.8472
0.2408	8.72	2800	0.5308	0.8362
0.2151	9.03	2900	0.4568	0.8607
0.1258	9.35	3000	0.5181	0.8507
0.1132	9.66	3100	0.5151	0.8477
0.09	9.97	3200	0.5395	0.8532
0.1324	10.28	3300	0.5641	0.8452
0.0447	10.59	3400	0.5345	0.8537
0.0795	10.9	3500	0.5789	0.8577
0.1806	11.21	3600	0.6749	0.8442
0.0822	11.53	3700	0.5518	0.8622
0.0915	11.84	3800	0.5960	0.8607
0.1171	12.15	3900	0.5547	0.8682
0.1169	12.46	4000	0.5873	0.8647
0.1077	12.77	4100	0.5914	0.8652
0.035	13.08	4200	0.5513	0.8642
0.088	13.4	4300	0.5972	0.8627
0.0419	13.71	4400	0.5934	0.8632
0.0762	14.02	4500	0.5979	0.8692
0.0773	14.33	4600	0.6091	0.8652
0.0611	14.64	4700	0.5911	0.8647
0.0623	14.95	4800	0.5923	0.8712

Framework versions

Transformers 4.38.1
Pytorch 2.1.0+cu121
Datasets 2.17.1
Tokenizers 0.15.2

Task: Image Classification
Library: transformers
Type: vit
License: apache-2.0
Created: Feb 23, 2024
Updated: Feb 24, 2024