matthieulel/swin-tiny-patch4-window7-224-finetuned-galaxy10-decals

swin-tiny-patch4-window7-224-finetuned-galaxy

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

• Loss: 0.5336
• Accuracy: 0.8516

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3447	0.9978	112	0.5080	0.8384
0.3511	1.9955	224	0.5638	0.8372
0.2986	2.9933	336	0.5371	0.8341
0.3484	4.0	449	0.6118	0.8303
0.2946	4.9978	561	0.5799	0.8334
0.3378	5.9955	673	0.6073	0.8215
0.3418	6.9933	785	0.5790	0.8272
0.3625	8.0	898	0.5552	0.8378
0.3532	8.9978	1010	0.5870	0.8297
0.3402	9.9955	1122	0.5448	0.8372
0.2695	10.9933	1234	0.5966	0.8347
0.2787	12.0	1347	0.5727	0.8322
0.3075	12.9978	1459	0.5451	0.8403
0.3008	13.9955	1571	0.5888	0.8328
0.2825	14.9933	1683	0.5411	0.8466
0.3083	16.0	1796	0.5473	0.8460
0.3319	16.9978	1908	0.5336	0.8516
0.2984	17.9955	2020	0.5299	0.8460
0.2926	18.9933	2132	0.5276	0.8503
0.2845	19.9555	2240	0.5281	0.8516

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1