swinv2-tiny-patch4-window8-256-DMAE-U

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: 1.0462
  • Accuracy: 0.6522

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 40

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 0.86 3 1.3654 0.4783
No log 2.0 7 1.3027 0.4565
1.3356 2.86 10 1.2580 0.4565
1.3356 4.0 14 1.2157 0.4565
1.3356 4.86 17 1.2121 0.4565
1.202 6.0 21 1.2014 0.4565
1.202 6.86 24 1.2013 0.4565
1.202 8.0 28 1.1949 0.4565
1.1884 8.86 31 1.1934 0.4565
1.1884 10.0 35 1.1916 0.4565
1.1884 10.86 38 1.1829 0.4565
1.1351 12.0 42 1.1568 0.4565
1.1351 12.86 45 1.1371 0.4565
1.1351 14.0 49 1.1238 0.4783
1.132 14.86 52 1.1183 0.5217
1.132 16.0 56 1.0962 0.6087
1.132 16.86 59 1.0737 0.6087
1.0659 18.0 63 1.0462 0.6522
1.0659 18.86 66 1.0217 0.6304
1.0299 20.0 70 0.9955 0.6522
1.0299 20.86 73 0.9767 0.6304
1.0299 22.0 77 0.9495 0.6304
0.9684 22.86 80 0.9328 0.6304
0.9684 24.0 84 0.9176 0.6304
0.9684 24.86 87 0.9078 0.6304
0.9301 26.0 91 0.8966 0.6304
0.9301 26.86 94 0.8951 0.6304
0.9301 28.0 98 0.8894 0.6522
0.9258 28.86 101 0.8820 0.6304
0.9258 30.0 105 0.8771 0.6304
0.9258 30.86 108 0.8776 0.6522
0.8877 32.0 112 0.8754 0.6522
0.8877 32.86 115 0.8732 0.6522
0.8877 34.0 119 0.8721 0.6522
0.8953 34.29 120 0.8719 0.6522

Framework versions

  • Transformers 4.36.2
  • Pytorch 2.1.2+cu118
  • Datasets 2.16.1
  • Tokenizers 0.15.0
Downloads last month
6
Safetensors
Model size
27.6M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.

Model tree for Augusto777/swinv2-tiny-patch4-window8-256-DMAE-U

Finetuned
(121)
this model

Evaluation results