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

Loss: 8.0201
Accuracy: 0.0980

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: 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: 0.1
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.57	1	8.0201	0.0980
No log	1.71	3	8.0044	0.0980
No log	2.86	5	7.9306	0.0980
No log	4.0	7	7.7713	0.0980
No log	4.57	8	7.6511	0.0980
7.7785	5.71	10	7.3653	0.0980
7.7785	6.86	12	7.0246	0.0980
7.7785	8.0	14	6.6413	0.0980
7.7785	8.57	15	6.4670	0.0980
7.7785	9.71	17	6.1321	0.0980
7.7785	10.86	19	5.8360	0.0980
6.5357	12.0	21	5.5743	0.0980
6.5357	12.57	22	5.4552	0.0980
6.5357	13.71	24	5.2367	0.0980
6.5357	14.86	26	5.0418	0.0980
6.5357	16.0	28	4.8706	0.0980
6.5357	16.57	29	4.7939	0.0980
5.2494	17.71	31	4.6596	0.0980
5.2494	18.86	33	4.5508	0.0980
5.2494	20.0	35	4.4676	0.0980
5.2494	20.57	36	4.4356	0.0980
5.2494	21.71	38	4.3906	0.0980
4.5614	22.86	40	4.3714	0.0980

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0

Augusto777/swinv2-tiny-patch4-window8-256-ve-Ub

swinv2-tiny-patch4-window8-256-ve-Ub

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Augusto777/
swinv2-tiny-patch4-window8-256-ve-Ub