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

Loss: 0.8865
Accuracy: 0.6944

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.5714	1	1.3455	0.3611
No log	1.7143	3	1.1600	0.3889
No log	2.8571	5	1.1451	0.3889
No log	4.0	7	1.1113	0.4167
No log	4.5714	8	1.1294	0.3611
1.114	5.7143	10	1.1000	0.4722
1.114	6.8571	12	1.0526	0.5
1.114	8.0	14	1.0222	0.5
1.114	8.5714	15	1.0203	0.4444
1.114	9.7143	17	0.9883	0.5556
1.114	10.8571	19	0.9686	0.5
0.928	12.0	21	0.9622	0.5556
0.928	12.5714	22	0.9396	0.5556
0.928	13.7143	24	0.9678	0.5556
0.928	14.8571	26	0.9045	0.6389
0.928	16.0	28	0.8842	0.6667
0.928	16.5714	29	0.8690	0.6389
0.7807	17.7143	31	0.8539	0.6389
0.7807	18.8571	33	0.9446	0.6389
0.7807	20.0	35	0.8785	0.6667
0.7807	20.5714	36	0.8500	0.6389
0.7807	21.7143	38	0.9317	0.6389
0.6419	22.8571	40	0.9105	0.6111
0.6419	24.0	42	0.9513	0.6111
0.6419	24.5714	43	0.9155	0.6111
0.6419	25.7143	45	0.8752	0.6389
0.6419	26.8571	47	0.8911	0.6111
0.6419	28.0	49	0.8772	0.5833
0.5401	28.5714	50	0.8496	0.5833
0.5401	29.7143	52	0.8562	0.6667
0.5401	30.8571	54	0.8377	0.6111
0.5401	32.0	56	0.9969	0.6111
0.5401	32.5714	57	1.0985	0.5833
0.5401	33.7143	59	0.9632	0.5833
0.4658	34.8571	61	0.8651	0.6389
0.4658	36.0	63	0.8731	0.6111
0.4658	36.5714	64	0.9148	0.5556
0.4658	37.7143	66	1.0383	0.6111
0.4658	38.8571	68	0.9203	0.6111
0.3816	40.0	70	0.8720	0.6389
0.3816	40.5714	71	0.8789	0.6389
0.3816	41.7143	73	0.8742	0.6667
0.3816	42.8571	75	0.8865	0.6944
0.3816	44.0	77	0.8931	0.6667
0.3816	44.5714	78	0.9036	0.6667
0.337	45.7143	80	0.9182	0.6389
0.337	46.8571	82	0.9406	0.6389
0.337	48.0	84	0.9587	0.6667
0.337	48.5714	85	0.9697	0.6389
0.337	49.7143	87	0.9818	0.6667
0.337	50.8571	89	0.9692	0.6389
0.2958	52.0	91	0.9426	0.6389
0.2958	52.5714	92	0.9374	0.6667
0.2958	53.7143	94	0.9338	0.6389
0.2958	54.8571	96	0.9337	0.6389
0.2958	56.0	98	0.9361	0.6389
0.2958	56.5714	99	0.9377	0.6389
0.2914	57.1429	100	0.9385	0.6389

Framework versions

Transformers 4.41.0
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1

BilalMuftuoglu/deit-base-distilled-patch16-224-hasta-55-fold3

deit-base-distilled-patch16-224-hasta-55-fold3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

BilalMuftuoglu/
deit-base-distilled-patch16-224-hasta-55-fold3