vit-finetuned2

This model is a fine-tuned version of microsoft/resnet-18 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8828
• Accuracy: 0.746

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: 64
• eval_batch_size: 128
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	211	3.2058	0.2
No log	2.0	422	2.7863	0.27
3.5109	3.0	633	2.6225	0.306
3.5109	4.0	844	2.3383	0.392
2.6956	5.0	1055	2.1045	0.456
2.6956	6.0	1266	1.8551	0.504
2.6956	7.0	1477	1.6949	0.54
2.213	8.0	1688	1.5866	0.576
2.213	9.0	1899	1.3373	0.646
1.8406	10.0	2110	1.2958	0.64
1.8406	11.0	2321	1.3066	0.652
1.5618	12.0	2532	1.1972	0.664
1.5618	13.0	2743	1.1654	0.67
1.5618	14.0	2954	1.0900	0.7
1.3308	15.0	3165	1.0244	0.704
1.3308	16.0	3376	1.0534	0.706
1.1426	17.0	3587	0.9758	0.732
1.1426	18.0	3798	0.9583	0.716
1.0085	19.0	4009	0.9191	0.732
1.0085	20.0	4220	0.8828	0.746

Framework versions

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