README.md

metadata

library_name: transformers
license: apache-2.0
base_model: microsoft/swin-small-patch4-window7-224
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - f1
model-index:
  - name: swin-small-patch4-window7-224
    results: []

swin-small-patch4-window7-224

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

• Loss: 0.1892
• Accuracy: 0.9778
• Precision: 0.9804
• Sensitivity: 1.0
• Specificity: 0.9733
• F1: 0.9783
• Auc: 0.9909
• Mcc: 0.9267
• J Stat: 0.9733
• Confusion Matrix: [[146, 4], [0, 30]]

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: 8.068420130106194e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.06206076180207542
• num_epochs: 10
• label_smoothing_factor: 0.06417838785936565

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Sensitivity	Specificity	F1	Auc	Mcc	J Stat	Confusion Matrix
No log	1.0	47	0.2904	0.9228	0.9256	0.7333	0.9892	0.9191	0.9754	0.7944	0.7225	[[1100, 12], [104, 286]]
0.4216	2.0	94	0.2247	0.9634	0.9640	0.8769	0.9937	0.9627	0.9866	0.9038	0.8706	[[1105, 7], [48, 342]]
0.2859	3.0	141	0.2064	0.9800	0.9800	0.9410	0.9937	0.9799	0.9914	0.9477	0.9347	[[1105, 7], [23, 367]]
0.2231	4.0	188	0.1778	0.9847	0.9847	0.9513	0.9964	0.9846	0.9956	0.9600	0.9477	[[1108, 4], [19, 371]]
0.1986	5.0	235	0.1590	0.9933	0.9933	0.9872	0.9955	0.9933	0.9982	0.9827	0.9827	[[1107, 5], [5, 385]]
0.174	6.0	282	0.1484	0.9973	0.9973	0.9923	0.9991	0.9973	0.9995	0.9931	0.9914	[[1111, 1], [3, 387]]
0.1526	7.0	329	0.1469	0.9980	0.9980	0.9949	0.9991	0.9980	0.9999	0.9948	0.9940	[[1111, 1], [2, 388]]
0.1518	8.0	376	0.1461	0.9987	0.9987	0.9949	1.0	0.9987	0.9997	0.9965	0.9949	[[1112, 0], [2, 388]]
0.148	9.0	423	0.1459	0.9987	0.9987	0.9949	1.0	0.9987	1.0000	0.9965	0.9949	[[1112, 0], [2, 388]]
0.1496	10.0	470	0.1458	0.9987	0.9987	0.9949	1.0	0.9987	1.0000	0.9965	0.9949	[[1112, 0], [2, 388]]

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.2