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: 0.8625
Accuracy: 0.8228

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	0.8571	3	0.8664	0.3797
No log	2.0	7	0.8144	0.4810
0.7345	2.8571	10	0.6451	0.5823
0.7345	4.0	14	0.7268	0.5190
0.7345	4.8571	17	0.5731	0.7342
0.6447	6.0	21	0.6948	0.5316
0.6447	6.8571	24	0.5315	0.6962
0.6447	8.0	28	0.5292	0.6835
0.5582	8.8571	31	0.6226	0.5823
0.5582	10.0	35	0.5479	0.6329
0.5582	10.8571	38	0.5230	0.7975
0.4807	12.0	42	0.6146	0.6709
0.4807	12.8571	45	0.6346	0.6709
0.4807	14.0	49	1.0584	0.5823
0.4441	14.8571	52	0.4716	0.7848
0.4441	16.0	56	0.4454	0.7595
0.4441	16.8571	59	0.5688	0.7595
0.3829	18.0	63	0.5233	0.7975
0.3829	18.8571	66	0.5291	0.7975
0.3352	20.0	70	0.4981	0.7848
0.3352	20.8571	73	0.8590	0.7468
0.3352	22.0	77	0.5201	0.7975
0.341	22.8571	80	0.6530	0.7468
0.341	24.0	84	0.5732	0.7848
0.341	24.8571	87	0.6532	0.7975
0.2554	26.0	91	0.5656	0.7722
0.2554	26.8571	94	0.8085	0.7595
0.2554	28.0	98	0.4826	0.7848
0.325	28.8571	101	0.8227	0.7468
0.325	30.0	105	0.5349	0.7975
0.325	30.8571	108	0.7056	0.7722
0.2572	32.0	112	0.5415	0.7848
0.2572	32.8571	115	0.5953	0.7848
0.2572	34.0	119	0.5884	0.7848
0.1994	34.8571	122	0.5971	0.7975
0.1994	36.0	126	0.8481	0.7468
0.1994	36.8571	129	0.6107	0.7595
0.209	38.0	133	0.9242	0.7848
0.209	38.8571	136	0.7106	0.7722
0.188	40.0	140	0.9581	0.7722
0.188	40.8571	143	0.9282	0.7722
0.188	42.0	147	1.0447	0.7848
0.1982	42.8571	150	0.9748	0.7722
0.1982	44.0	154	0.7132	0.7848
0.1982	44.8571	157	0.6957	0.7848
0.1799	46.0	161	0.8674	0.7848
0.1799	46.8571	164	0.9036	0.7848
0.1799	48.0	168	0.9910	0.7848
0.1502	48.8571	171	1.0821	0.7722
0.1502	50.0	175	0.8117	0.7848
0.1502	50.8571	178	0.9037	0.7722
0.1591	52.0	182	0.8272	0.7722
0.1591	52.8571	185	1.0109	0.7848
0.1591	54.0	189	0.8490	0.7722
0.1588	54.8571	192	0.8973	0.7722
0.1588	56.0	196	0.9084	0.7722
0.1588	56.8571	199	0.8034	0.7595
0.1353	58.0	203	0.8885	0.7848
0.1353	58.8571	206	1.0837	0.7848
0.1308	60.0	210	0.8627	0.7975
0.1308	60.8571	213	0.9842	0.8101
0.1308	62.0	217	0.9966	0.7975
0.1402	62.8571	220	0.9639	0.7848
0.1402	64.0	224	1.0104	0.7848
0.1402	64.8571	227	0.8940	0.7722
0.1301	66.0	231	0.9349	0.7848
0.1301	66.8571	234	1.1485	0.7848
0.1301	68.0	238	1.0021	0.7848
0.1399	68.8571	241	0.8971	0.7975
0.1399	70.0	245	0.8604	0.7975
0.1399	70.8571	248	0.8625	0.8228
0.1203	72.0	252	1.0301	0.7975
0.1203	72.8571	255	1.2355	0.7722
0.1203	74.0	259	1.0716	0.7848
0.1385	74.8571	262	0.9598	0.7975
0.1385	76.0	266	0.9965	0.7975
0.1385	76.8571	269	1.1018	0.7975
0.115	78.0	273	1.0895	0.7975
0.115	78.8571	276	1.0261	0.7848
0.1227	80.0	280	0.9800	0.7975
0.1227	80.8571	283	0.9960	0.7848
0.1227	82.0	287	1.0251	0.7848
0.1044	82.8571	290	1.0236	0.7848
0.1044	84.0	294	1.0139	0.7848
0.1044	84.8571	297	1.0073	0.7848
0.1246	85.7143	300	1.0066	0.7848

Framework versions

Transformers 4.40.2
Pytorch 2.2.1+cu121
Datasets 2.19.1
Tokenizers 0.19.1

BilalMuftuoglu/beit-base-patch16-224-55-fold4

beit-base-patch16-224-fold4

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

BilalMuftuoglu/
beit-base-patch16-224-55-fold4