hushem_conflu_deneme_fold4

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: 0.8165
• Accuracy: 0.7381

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	1.7088	0.2381
1.9076	2.0	12	1.4617	0.2381
1.9076	3.0	18	1.4512	0.2619
1.4689	4.0	24	1.3283	0.2381
1.3599	5.0	30	1.0112	0.6667
1.3599	6.0	36	1.1598	0.3810
1.2233	7.0	42	1.4323	0.4524
1.2233	8.0	48	0.9658	0.6667
1.0502	9.0	54	0.9166	0.6429
0.8636	10.0	60	0.8181	0.6190
0.8636	11.0	66	1.2729	0.5238
0.8856	12.0	72	0.7434	0.7381
0.8856	13.0	78	0.6840	0.7143
0.6672	14.0	84	0.9596	0.5238
0.5861	15.0	90	0.7243	0.7381
0.5861	16.0	96	0.8378	0.7143
0.4357	17.0	102	0.8165	0.7381
0.4357	18.0	108	0.8165	0.7381
0.4614	19.0	114	0.8165	0.7381
0.431	20.0	120	0.8165	0.7381

Framework versions

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