swin-tiny-patch4-window7-224-finetuned-leukemia.v2.2

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

• Loss: 0.5507
• Accuracy: 0.7638

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: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2349	0.9984	312	0.5575	0.7698
0.2191	2.0	625	0.5572	0.7618
0.2124	2.9984	937	0.5580	0.769
0.2207	4.0	1250	0.5500	0.763
0.2143	4.9984	1562	0.5575	0.7652
0.2191	6.0	1875	0.5486	0.7728
0.2063	6.9984	2187	0.5594	0.7615
0.207	8.0	2500	0.5405	0.7695
0.2273	8.9984	2812	0.5568	0.7672
0.2136	10.0	3125	0.5483	0.7728
0.2184	10.9984	3437	0.5606	0.7665
0.212	12.0	3750	0.5578	0.761
0.1903	12.9984	4062	0.5371	0.769
0.2487	14.0	4375	0.5582	0.7645
0.2025	14.9984	4687	0.5414	0.7778
0.2207	16.0	5000	0.5376	0.7685
0.2012	16.9984	5312	0.5489	0.7702
0.2198	18.0	5625	0.5560	0.7752
0.2171	18.9984	5937	0.5570	0.7725
0.2116	20.0	6250	0.5622	0.7625
0.2162	20.9984	6562	0.5587	0.7668
0.224	22.0	6875	0.5456	0.7712
0.212	22.9984	7187	0.5647	0.7652
0.2084	24.0	7500	0.5533	0.7672
0.2226	24.9984	7812	0.5434	0.7705
0.2173	26.0	8125	0.5738	0.7675
0.2216	26.9984	8437	0.5557	0.7672
0.1918	28.0	8750	0.5502	0.7705
0.199	28.9984	9062	0.5456	0.7675
0.21	29.9520	9360	0.5483	0.7715

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1