hushem_conflu_deneme_fold3

This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.9617
• Accuracy: 0.6279

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: 0.001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• 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
No log	1.0	6	1.6871	0.2558
1.9835	2.0	12	1.3632	0.2326
1.9835	3.0	18	1.4109	0.3256
1.294	4.0	24	1.3794	0.4186
1.2341	5.0	30	1.2119	0.4651
1.2341	6.0	36	1.4964	0.4419
1.0897	7.0	42	1.2398	0.4651
1.0897	8.0	48	1.0532	0.5349
0.9835	9.0	54	1.1022	0.5116
0.9034	10.0	60	0.9784	0.6279
0.9034	11.0	66	1.5952	0.5116
0.8061	12.0	72	0.9828	0.5581
0.8061	13.0	78	0.9199	0.7209
0.765	14.0	84	1.0672	0.5581
0.6513	15.0	90	1.0129	0.6744
0.6513	16.0	96	0.9247	0.6977
0.4919	17.0	102	0.9617	0.6279
0.4919	18.0	108	0.9617	0.6279
0.4742	19.0	114	0.9617	0.6279
0.4695	20.0	120	0.9617	0.6279

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1