cvt-13-finetuned-AL

This model is a fine-tuned version of microsoft/cvt-13 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 2.6319
• Accuracy: 0.2619

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use 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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.2095	1.0	11	2.6973	0.1905
2.1537	2.0	22	2.6885	0.1667
2.3281	3.0	33	2.6838	0.1429
2.3004	4.0	44	2.6759	0.1667
2.3531	5.0	55	2.6765	0.1905
2.4045	6.0	66	2.6647	0.1905
2.3842	7.0	77	2.6552	0.2143
2.4049	8.0	88	2.6505	0.2381
2.3972	9.0	99	2.6470	0.2143
2.4238	10.0	110	2.6428	0.2143
2.4359	11.0	121	2.6429	0.1905
2.4042	12.0	132	2.6383	0.2619
2.4737	13.0	143	2.6364	0.2381
2.4003	14.0	154	2.6280	0.2143
2.382	15.0	165	2.6319	0.2619

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0