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

Loss: 5.8123
Accuracy: 0.4222

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: 0.001
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.3956	1.0	27	1.9737	0.3111
0.888	2.0	54	1.5910	0.3556
1.1183	3.0	81	1.4091	0.3778
0.7709	4.0	108	1.4706	0.3778
0.9892	5.0	135	1.4916	0.4222
0.5847	6.0	162	2.0869	0.3778
0.6569	7.0	189	1.9470	0.3556
0.5263	8.0	216	1.6436	0.4
0.46	9.0	243	2.3342	0.3556
0.4825	10.0	270	1.8564	0.4222
0.3607	11.0	297	2.1004	0.4222
0.2444	12.0	324	2.4392	0.4222
0.3872	13.0	351	1.8032	0.4444
0.3209	14.0	378	2.9763	0.4
0.1884	15.0	405	2.8695	0.4667
0.1329	16.0	432	3.4787	0.4
0.2021	17.0	459	2.9858	0.4
0.1653	18.0	486	3.6825	0.4667
0.0813	19.0	513	3.2825	0.4444
0.1467	20.0	540	3.0809	0.4889
0.0538	21.0	567	3.9816	0.4222
0.1511	22.0	594	3.9404	0.4444
0.0505	23.0	621	4.4773	0.4667
0.0602	24.0	648	3.6484	0.4222
0.0403	25.0	675	4.0392	0.4444
0.005	26.0	702	3.8791	0.5556
0.0725	27.0	729	5.2091	0.4222
0.0084	28.0	756	4.7587	0.4222
0.0002	29.0	783	5.6091	0.3778
0.0001	30.0	810	5.5834	0.4222
0.0004	31.0	837	5.1075	0.4
0.0002	32.0	864	5.0938	0.4667
0.0014	33.0	891	5.4645	0.4667
0.0	34.0	918	5.9402	0.4222
0.0	35.0	945	5.8799	0.4222
0.0	36.0	972	5.8415	0.4222
0.0	37.0	999	5.8263	0.4222
0.0	38.0	1026	5.8129	0.4222
0.0	39.0	1053	5.8088	0.4222
0.0	40.0	1080	5.8085	0.4222
0.0	41.0	1107	5.8075	0.4222
0.0	42.0	1134	5.8084	0.4222
0.0	43.0	1161	5.8094	0.4222
0.0	44.0	1188	5.8109	0.4222
0.0	45.0	1215	5.8113	0.4222
0.0	46.0	1242	5.8120	0.4222
0.0	47.0	1269	5.8119	0.4222
0.0	48.0	1296	5.8123	0.4222
0.0	49.0	1323	5.8123	0.4222
0.0	50.0	1350	5.8123	0.4222

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.15.0

hkivancoral/hushem_5x_deit_tiny_adamax_001_fold1

hushem_5x_deit_tiny_adamax_001_fold1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

hkivancoral/
hushem_5x_deit_tiny_adamax_001_fold1