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: 0.9265
Accuracy: 0.6389

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	1.3178	0.2222
No log	1.7143	3	1.4014	0.2778
No log	2.8571	5	1.3535	0.2778
No log	4.0	7	1.1299	0.3056
No log	4.5714	8	1.0860	0.4722
1.0868	5.7143	10	1.1121	0.3333
1.0868	6.8571	12	1.0691	0.3611
1.0868	8.0	14	1.0270	0.5
1.0868	8.5714	15	1.0360	0.5
1.0868	9.7143	17	1.0385	0.3889
1.0868	10.8571	19	0.9951	0.4167
0.9487	12.0	21	1.0029	0.4444
0.9487	12.5714	22	1.0134	0.4722
0.9487	13.7143	24	0.9599	0.4444
0.9487	14.8571	26	0.9117	0.5278
0.9487	16.0	28	0.8856	0.5278
0.9487	16.5714	29	0.9275	0.4722
0.7942	17.7143	31	0.9041	0.5278
0.7942	18.8571	33	0.8999	0.4722
0.7942	20.0	35	0.8832	0.5833
0.7942	20.5714	36	0.8864	0.5556
0.7942	21.7143	38	0.8551	0.5
0.5911	22.8571	40	0.8242	0.6111
0.5911	24.0	42	0.9265	0.6389
0.5911	24.5714	43	0.8674	0.5833
0.5911	25.7143	45	0.7892	0.5556
0.5911	26.8571	47	0.8005	0.5833
0.5911	28.0	49	0.8302	0.5833
0.4865	28.5714	50	0.8893	0.6111
0.4865	29.7143	52	0.9043	0.6111
0.4865	30.8571	54	0.8433	0.5833
0.4865	32.0	56	0.8677	0.5833
0.4865	32.5714	57	0.9008	0.5833
0.4865	33.7143	59	0.9533	0.6111
0.4007	34.8571	61	0.9175	0.6111
0.4007	36.0	63	0.9090	0.5833
0.4007	36.5714	64	1.0004	0.5
0.4007	37.7143	66	1.0393	0.5
0.4007	38.8571	68	0.9196	0.5833
0.3691	40.0	70	0.9505	0.6389
0.3691	40.5714	71	0.9634	0.6389
0.3691	41.7143	73	0.9718	0.5278
0.3691	42.8571	75	0.9257	0.5278
0.3691	44.0	77	0.9020	0.5
0.3691	44.5714	78	0.9132	0.5556
0.3278	45.7143	80	1.0340	0.5556
0.3278	46.8571	82	1.0933	0.5833
0.3278	48.0	84	1.0231	0.5
0.3278	48.5714	85	0.9826	0.5278
0.3278	49.7143	87	0.9329	0.5278
0.3278	50.8571	89	0.9280	0.5278
0.2909	52.0	91	0.9312	0.5556
0.2909	52.5714	92	0.9359	0.5556
0.2909	53.7143	94	0.9495	0.5833
0.2909	54.8571	96	0.9607	0.5833
0.2909	56.0	98	0.9685	0.5833
0.2909	56.5714	99	0.9703	0.5833
0.2697	57.1429	100	0.9713	0.5833

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-65-fold1

deit-base-distilled-patch16-224-hasta-65-fold1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

BilalMuftuoglu/
deit-base-distilled-patch16-224-hasta-65-fold1