This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.5958
Accuracy: 0.8782

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
5.175	0.96	16	4.7967	0.1345
4.1158	1.97	33	2.9977	0.3824
2.0676	2.99	50	1.5415	0.6807
1.4395	4.0	67	0.9951	0.8151
0.9396	4.96	83	0.8235	0.8277
0.7456	5.97	100	0.7195	0.8361
0.666	6.99	117	0.6406	0.8613
0.5893	8.0	134	0.6045	0.8739
0.4704	8.96	150	0.6016	0.8655
0.4475	9.97	167	0.5958	0.8782
0.3937	10.99	184	0.5856	0.8782
0.3327	12.0	201	0.5761	0.8782
0.3277	12.96	217	0.5758	0.8782
0.2928	13.97	234	0.5754	0.8739
0.2545	14.99	251	0.5711	0.8739
0.2657	16.0	268	0.5851	0.8739
0.2457	16.96	284	0.5805	0.8655
0.2359	17.97	301	0.5762	0.8697
0.2849	18.99	318	0.5792	0.8739
0.223	19.1	320	0.5792	0.8739

Framework versions

Transformers 4.37.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.1

tectonatech/swin-tiny-patch4-window7-224-fine_tune

swin-tiny-patch4-window7-224-fine_tune

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

tectonatech/
swin-tiny-patch4-window7-224-fine_tune