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: 0.9542
Accuracy: 0.7273

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	1.0	1	2.0908	0.0
No log	2.0	2	1.8864	0.0
No log	3.0	3	1.5313	0.0
No log	4.0	4	1.1520	0.3636
No log	5.0	5	0.9542	0.7273
No log	6.0	6	1.0324	0.7273
No log	7.0	7	1.1792	0.7273
No log	8.0	8	1.2680	0.7273
No log	9.0	9	1.2078	0.7273
0.4805	10.0	10	1.1594	0.7273
0.4805	11.0	11	1.1604	0.7273
0.4805	12.0	12	1.1110	0.7273
0.4805	13.0	13	1.1074	0.7273
0.4805	14.0	14	1.1010	0.7273
0.4805	15.0	15	1.0594	0.7273
0.4805	16.0	16	1.0675	0.7273
0.4805	17.0	17	1.1689	0.7273
0.4805	18.0	18	1.2630	0.7273
0.4805	19.0	19	1.3017	0.7273
0.1904	20.0	20	1.3083	0.7273
0.1904	21.0	21	1.3764	0.7273
0.1904	22.0	22	1.4229	0.7273
0.1904	23.0	23	1.4830	0.7273
0.1904	24.0	24	1.4698	0.7273
0.1904	25.0	25	1.4947	0.7273
0.1904	26.0	26	1.5839	0.7273
0.1904	27.0	27	1.7219	0.7273
0.1904	28.0	28	1.7474	0.7273
0.1904	29.0	29	1.6645	0.7273
0.1472	30.0	30	1.5988	0.7273
0.1472	31.0	31	1.5703	0.7273
0.1472	32.0	32	1.5965	0.7273
0.1472	33.0	33	1.6618	0.7273
0.1472	34.0	34	1.6432	0.7273
0.1472	35.0	35	1.5392	0.7273
0.1472	36.0	36	1.4368	0.7273
0.1472	37.0	37	1.4019	0.7273
0.1472	38.0	38	1.4714	0.7273
0.1472	39.0	39	1.7523	0.7273
0.0741	40.0	40	1.8653	0.7273
0.0741	41.0	41	1.8966	0.7273
0.0741	42.0	42	1.8252	0.7273
0.0741	43.0	43	1.8433	0.7273
0.0741	44.0	44	1.7327	0.7273
0.0741	45.0	45	1.7456	0.7273
0.0741	46.0	46	1.8948	0.7273
0.0741	47.0	47	1.8811	0.7273
0.0741	48.0	48	1.8930	0.7273
0.0741	49.0	49	1.8065	0.7273
0.0506	50.0	50	1.7629	0.7273
0.0506	51.0	51	1.7893	0.7273
0.0506	52.0	52	1.8770	0.7273
0.0506	53.0	53	1.9144	0.7273
0.0506	54.0	54	2.0357	0.7273
0.0506	55.0	55	2.2103	0.7273
0.0506	56.0	56	2.2568	0.7273
0.0506	57.0	57	2.2401	0.7273
0.0506	58.0	58	2.1406	0.7273
0.0506	59.0	59	2.0280	0.7273
0.0355	60.0	60	1.9517	0.7273
0.0355	61.0	61	1.8787	0.7273
0.0355	62.0	62	1.8958	0.7273
0.0355	63.0	63	1.9893	0.7273
0.0355	64.0	64	2.1195	0.7273
0.0355	65.0	65	2.2715	0.7273
0.0355	66.0	66	2.3640	0.7273
0.0355	67.0	67	2.4391	0.7273
0.0355	68.0	68	2.4633	0.7273
0.0355	69.0	69	2.4489	0.7273
0.021	70.0	70	2.4177	0.7273
0.021	71.0	71	2.3674	0.7273
0.021	72.0	72	2.3065	0.7273
0.021	73.0	73	2.2978	0.7273
0.021	74.0	74	2.3590	0.7273
0.021	75.0	75	2.4277	0.7273
0.021	76.0	76	2.5220	0.7273
0.021	77.0	77	2.6248	0.7273
0.021	78.0	78	2.6925	0.7273
0.021	79.0	79	2.7284	0.7273
0.0346	80.0	80	2.7409	0.7273
0.0346	81.0	81	2.7218	0.7273
0.0346	82.0	82	2.6717	0.7273
0.0346	83.0	83	2.6142	0.7273
0.0346	84.0	84	2.5196	0.7273
0.0346	85.0	85	2.4438	0.7273
0.0346	86.0	86	2.3812	0.7273
0.0346	87.0	87	2.3026	0.7273
0.0346	88.0	88	2.2691	0.7273
0.0346	89.0	89	2.2665	0.7273
0.0219	90.0	90	2.2659	0.7273
0.0219	91.0	91	2.2735	0.7273
0.0219	92.0	92	2.3000	0.7273
0.0219	93.0	93	2.3136	0.7273
0.0219	94.0	94	2.3247	0.7273
0.0219	95.0	95	2.3388	0.7273
0.0219	96.0	96	2.3597	0.7273
0.0219	97.0	97	2.3746	0.7273
0.0219	98.0	98	2.3864	0.7273
0.0219	99.0	99	2.3913	0.7273
0.0286	100.0	100	2.3933	0.7273

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-85-fold4

deit-base-distilled-patch16-224-hasta-85-fold4

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

BilalMuftuoglu/
deit-base-distilled-patch16-224-hasta-85-fold4