swin-camdeboo-loc

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: 1.7931
• Accuracy: 0.6781

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 8
• 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
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.706	0.0633	100	1.7448	0.4844
1.3052	0.1267	200	1.9530	0.4844
1.1569	0.1900	300	1.6721	0.5363
1.4451	0.2533	400	1.6504	0.5812
0.9833	0.3167	500	2.2411	0.4998
1.1427	0.3800	600	1.5587	0.5720
1.0228	0.4433	700	1.9815	0.5817
0.8635	0.5066	800	1.6364	0.5953
0.8243	0.5700	900	1.6178	0.5834
0.6203	0.6333	1000	1.5926	0.5997
0.7647	0.6966	1100	1.9691	0.5491
0.6963	0.7600	1200	1.7098	0.6055
0.7896	0.8233	1300	1.6605	0.5962
0.8549	0.8866	1400	1.4914	0.6231
0.6353	0.9500	1500	2.0022	0.5495
0.5124	1.0133	1600	1.8427	0.5914
0.297	1.0766	1700	1.9349	0.6090
0.5861	1.1400	1800	1.7115	0.6667
0.7606	1.2033	1900	2.1216	0.6200
0.1236	1.2666	2000	2.1265	0.6196
0.3057	1.3300	2100	1.7538	0.6332
0.3497	1.3933	2200	2.1055	0.6072
0.5086	1.4566	2300	2.0850	0.6433
0.3447	1.5199	2400	1.8278	0.6583
0.6164	1.5833	2500	1.9431	0.6160
0.3085	1.6466	2600	1.5871	0.6874
0.5517	1.7099	2700	1.7441	0.6535
0.3404	1.7733	2800	1.6698	0.6742
0.1559	1.8366	2900	1.8379	0.6706
0.3857	1.8999	3000	1.8784	0.6583
0.2708	1.9633	3100	1.7931	0.6781

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1