swin-tiny-patch4-window7-224-finetuned-leukemia.v2.0

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: 0.5694
• Accuracy: 0.8855

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
0.4734	0.9984	312	0.7528	0.5968
0.3596	2.0	625	0.8091	0.688
0.2991	2.9984	937	0.9220	0.6335
0.2658	4.0	1250	0.7774	0.7137
0.2511	4.9984	1562	0.4364	0.8267
0.2218	6.0	1875	0.6225	0.7837
0.1691	6.9984	2187	0.3587	0.8718
0.1721	8.0	2500	0.6494	0.7987
0.1393	8.9984	2812	0.6802	0.818
0.1109	10.0	3125	0.5511	0.834
0.1213	10.9984	3437	0.5982	0.8417
0.0971	12.0	3750	0.8005	0.814
0.1121	12.9984	4062	0.6397	0.8407
0.0947	14.0	4375	1.0869	0.768
0.1022	14.9984	4687	0.5969	0.8515
0.0801	16.0	5000	0.5839	0.8732
0.0951	16.9984	5312	0.8599	0.827
0.0716	18.0	5625	0.8355	0.822
0.0859	18.9984	5937	0.7547	0.8427
0.0661	20.0	6250	0.7206	0.851
0.0543	20.9984	6562	0.8396	0.8363
0.0646	22.0	6875	0.5467	0.881
0.0563	22.9984	7187	0.5694	0.8855
0.042	24.0	7500	0.8404	0.8492
0.0638	24.9984	7812	0.9300	0.84
0.0455	26.0	8125	0.9865	0.8393
0.037	26.9984	8437	0.8503	0.8525
0.0469	28.0	8750	0.8272	0.8602
0.0409	28.9984	9062	0.8988	0.8502
0.0438	29.9520	9360	0.8338	0.858

Framework versions

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