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

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: 0.9332
• Accuracy: 0.6923

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
No log	0.8421	4	0.9536	0.6538
No log	1.8947	9	0.9803	0.6154
0.7659	2.9474	14	0.9831	0.6346
0.7659	4.0	19	1.0565	0.6346
0.7069	4.8421	23	0.9332	0.6923
0.7069	5.8947	28	0.9039	0.6731
0.5937	6.9474	33	0.8719	0.6538
0.5937	8.0	38	0.9287	0.6923
0.5467	8.8421	42	0.9408	0.6538
0.5467	9.8947	47	0.9465	0.6346
0.5226	10.9474	52	1.0464	0.6346
0.5226	12.0	57	1.0826	0.6538
0.4715	12.8421	61	0.9619	0.6154
0.4715	13.8947	66	0.9822	0.6154
0.4752	14.9474	71	0.9639	0.6538
0.4752	16.0	76	0.9449	0.6731
0.4688	16.8421	80	0.9450	0.6346

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1