swinv2-tiny-patch4-window8-256-dmae-humeda-DAV60

RobertoSonic/swinv2-tiny-patch4-window8-256-dmae-humeda-DAV60

swinv2-tiny-patch4-window8-256-dmae-humeda-DAV60 is a image classification model from RobertoSonic. 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

19% of all-time downloads

All-time downloads

Public

Parameters

27.6M

4.4 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors110 MB · 100%

At a glance

Task: Image Classification
Library: transformers
License: apache-2.0
Model type: swinv2
Access: Public
Created: Mar 10, 2025
Updated: Mar 10, 2025
SHA: 1cf008d9

Base models

microsoft/swinv2-tiny-patch4-window8-256

Model card

From the Hugging Face model README

swinv2-tiny-patch4-window8-256-dmae-humeda-DAV60

This model is a fine-tuned version of microsoft/swinv2-tiny-patch4-window8-256 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.5934
Accuracy: 0.8333

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: 4e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 40
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5779	0.9630	13	1.5147	0.3509
1.532	1.9630	26	1.3726	0.5395
1.459	2.9630	39	0.9852	0.6184
1.0357	3.9630	52	0.8558	0.6140
0.9678	4.9630	65	0.7997	0.6623
0.9602	5.9630	78	0.7789	0.6711
0.7664	6.9630	91	0.8890	0.6184
0.8537	7.9630	104	0.6860	0.7105
0.7566	8.9630	117	0.6138	0.7588
0.7025	9.9630	130	0.5207	0.7939
0.6081	10.9630	143	0.5644	0.7763
0.631	11.9630	156	0.5859	0.7544
0.6163	12.9630	169	0.7182	0.7105
0.5743	13.9630	182	0.5643	0.7763
0.5752	14.9630	195	0.5028	0.7939
0.461	15.9630	208	0.5465	0.7807
0.4145	16.9630	221	0.5868	0.7719
0.4065	17.9630	234	0.5470	0.7807
0.501	18.9630	247	0.5406	0.7939
0.4374	19.9630	260	0.5534	0.7939
0.4614	20.9630	273	0.5485	0.8158
0.3836	21.9630	286	0.6217	0.7851
0.4474	22.9630	299	0.6069	0.7763
0.3893	23.9630	312	0.5981	0.7939
0.3548	24.9630	325	0.6003	0.7895
0.3454	25.9630	338	0.5897	0.8114
0.2857	26.9630	351	0.6031	0.8158
0.3282	27.9630	364	0.6140	0.7763
0.3088	28.9630	377	0.5934	0.8333
0.2943	29.9630	390	0.6545	0.7895
0.2857	30.9630	403	0.6423	0.7851
0.2882	31.9630	416	0.6610	0.7939
0.3342	32.9630	429	0.6815	0.7807
0.2679	33.9630	442	0.6481	0.8114
0.2756	34.9630	455	0.6518	0.8158
0.2733	35.9630	468	0.6514	0.8026
0.2749	36.9630	481	0.6525	0.8070
0.2416	37.9630	494	0.6447	0.8202
0.2766	38.9630	507	0.6359	0.8202
0.2528	39.9630	520	0.6481	0.8158

Framework versions

Transformers 4.48.3
Pytorch 2.5.1+cu124
Datasets 3.3.2
Tokenizers 0.21.0

Task: Image Classification
Library: transformers
Type: swinv2
License: apache-2.0
Created: Mar 10, 2025
Updated: Mar 10, 2025