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

Loss: 0.3734
Accuracy: 0.9014

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	0.9231	3	0.6830	0.5493
No log	1.8462	6	0.6407	0.6479
No log	2.7692	9	0.6612	0.5352
0.7094	4.0	13	0.6175	0.6901
0.7094	4.9231	16	0.5915	0.6761
0.7094	5.8462	19	0.5677	0.7042
0.6444	6.7692	22	0.5177	0.7746
0.6444	8.0	26	0.4928	0.7887
0.6444	8.9231	29	0.5641	0.6901
0.5574	9.8462	32	0.4372	0.8169
0.5574	10.7692	35	0.5677	0.7606
0.5574	12.0	39	0.5032	0.7606
0.543	12.9231	42	0.4745	0.8028
0.543	13.8462	45	0.4057	0.8310
0.543	14.7692	48	0.4013	0.7746
0.4499	16.0	52	0.3670	0.8310
0.4499	16.9231	55	0.4215	0.8592
0.4499	17.8462	58	0.4862	0.7746
0.3902	18.7692	61	0.5781	0.7324
0.3902	20.0	65	0.5249	0.8028
0.3902	20.9231	68	0.3937	0.8732
0.4029	21.8462	71	0.4132	0.8451
0.4029	22.7692	74	0.4178	0.8451
0.4029	24.0	78	0.7273	0.7183
0.3163	24.9231	81	0.4221	0.8451
0.3163	25.8462	84	0.4086	0.8732
0.3163	26.7692	87	0.3946	0.8732
0.2786	28.0	91	0.5320	0.8028
0.2786	28.9231	94	0.4132	0.8451
0.2786	29.8462	97	0.5542	0.7746
0.2763	30.7692	100	0.3734	0.9014
0.2763	32.0	104	0.4479	0.8310
0.2763	32.9231	107	0.3482	0.8592
0.25	33.8462	110	0.5442	0.7887
0.25	34.7692	113	0.4211	0.8732
0.25	36.0	117	0.4860	0.8592
0.2125	36.9231	120	0.4654	0.8451
0.2125	37.8462	123	0.4779	0.8592
0.2125	38.7692	126	0.5692	0.8310
0.2225	40.0	130	0.4912	0.8451
0.2225	40.9231	133	0.4528	0.8310
0.2225	41.8462	136	0.4470	0.7887
0.2225	42.7692	139	0.4251	0.8028
0.1991	44.0	143	0.4864	0.8028
0.1991	44.9231	146	0.4652	0.8592
0.1991	45.8462	149	0.5949	0.8310
0.164	46.7692	152	1.0035	0.7465
0.164	48.0	156	0.6393	0.8310
0.164	48.9231	159	0.9222	0.6761
0.1974	49.8462	162	1.0633	0.6901
0.1974	50.7692	165	0.6050	0.8310
0.1974	52.0	169	0.7134	0.8310
0.213	52.9231	172	0.6649	0.8310
0.213	53.8462	175	0.7126	0.8310
0.213	54.7692	178	0.6906	0.8169
0.1642	56.0	182	0.6956	0.8310
0.1642	56.9231	185	0.5828	0.8028
0.1642	57.8462	188	0.5866	0.8028
0.1657	58.7692	191	0.6172	0.8169
0.1657	60.0	195	0.7428	0.8310
0.1657	60.9231	198	0.8981	0.8169
0.1347	61.8462	201	0.7168	0.8169
0.1347	62.7692	204	0.8026	0.8451
0.1347	64.0	208	0.8639	0.8451
0.1335	64.9231	211	0.7604	0.8169
0.1335	65.8462	214	0.7993	0.8169
0.1335	66.7692	217	0.8330	0.8592
0.145	68.0	221	0.8143	0.8592
0.145	68.9231	224	0.7520	0.8592
0.145	69.8462	227	0.7216	0.8451
0.1658	70.7692	230	0.7968	0.8592
0.1658	72.0	234	0.7730	0.8732
0.1658	72.9231	237	0.7450	0.8732
0.1381	73.8462	240	0.7855	0.8732
0.1381	74.7692	243	0.8253	0.8592
0.1381	76.0	247	0.8065	0.8592
0.1306	76.9231	250	0.7778	0.8592
0.1306	77.8462	253	0.7814	0.8592
0.1306	78.7692	256	0.7335	0.8310
0.1027	80.0	260	0.7372	0.8451
0.1027	80.9231	263	0.7618	0.8451
0.1027	81.8462	266	0.7891	0.8451
0.1027	82.7692	269	0.8287	0.8592
0.1296	84.0	273	0.8412	0.8451
0.1296	84.9231	276	0.8014	0.8451
0.1296	85.8462	279	0.7530	0.8451
0.1162	86.7692	282	0.7243	0.8451
0.1162	88.0	286	0.7247	0.8451
0.1162	88.9231	289	0.7354	0.8451
0.1166	89.8462	292	0.7391	0.8592
0.1166	90.7692	295	0.7390	0.8592
0.1166	92.0	299	0.7374	0.8592
0.1031	92.3077	300	0.7373	0.8592

Framework versions

Transformers 4.40.2
Pytorch 2.2.1+cu121
Datasets 2.19.1
Tokenizers 0.19.1

BilalMuftuoglu/beit-base-patch16-224-65-fold2

beit-base-patch16-224-65-fold2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

BilalMuftuoglu/
beit-base-patch16-224-65-fold2