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: 0.8536
Accuracy: 0.6667

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	1	1.9028	0.3333
No log	2.0	2	1.8803	0.3333
No log	3.0	3	1.8603	0.3333
No log	4.0	4	1.8293	0.3333
No log	5.0	5	1.8093	0.3333
No log	6.0	6	1.7682	0.3333
No log	7.0	7	1.7140	0.3333
No log	8.0	8	1.6566	0.3333
No log	9.0	9	1.6020	0.3333
0.8416	10.0	10	1.5466	0.3333
0.8416	11.0	11	1.4709	0.3333
0.8416	12.0	12	1.3894	0.3333
0.8416	13.0	13	1.3049	0.5
0.8416	14.0	14	1.2186	0.6667
0.8416	15.0	15	1.1344	0.6667
0.8416	16.0	16	1.0706	0.6667
0.8416	17.0	17	1.0184	0.6667
0.8416	18.0	18	0.9807	0.6667
0.8416	19.0	19	0.9455	0.6667
0.433	20.0	20	0.9182	0.6667
0.433	21.0	21	0.8942	0.6667
0.433	22.0	22	0.8766	0.6667
0.433	23.0	23	0.8654	0.6667
0.433	24.0	24	0.8578	0.6667
0.433	25.0	25	0.8536	0.6667

Framework versions

Transformers 4.40.1
Pytorch 2.2.1+cu121
Datasets 2.19.0
Tokenizers 0.19.1

Veda0718/swin-tiny-patch4-window7-224-finetuned-fish

swin-tiny-patch4-window7-224-finetuned-fish

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Veda0718/
swin-tiny-patch4-window7-224-finetuned-fish