hushem_5x_deit_small_rms_001_fold2

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

• Loss: 3.6556
• Accuracy: 0.3556

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
2.0324	1.0	27	1.4564	0.2444
1.4819	2.0	54	1.4327	0.2444
1.4504	3.0	81	1.4455	0.2667
1.4703	4.0	108	1.5353	0.2444
1.4319	5.0	135	1.4161	0.2444
1.4127	6.0	162	1.4083	0.2444
1.424	7.0	189	1.4264	0.2667
1.3928	8.0	216	1.4087	0.2889
1.4183	9.0	243	1.3797	0.2667
1.2937	10.0	270	1.5479	0.3333
1.444	11.0	297	1.4212	0.2667
1.2489	12.0	324	1.3827	0.3333
1.2092	13.0	351	1.4109	0.3333
1.1924	14.0	378	1.3647	0.3556
1.1322	15.0	405	1.4486	0.4
1.059	16.0	432	1.3236	0.2889
1.007	17.0	459	1.5059	0.3778
1.0396	18.0	486	1.8214	0.3778
0.9935	19.0	513	1.6035	0.2222
0.9595	20.0	540	1.8699	0.3111
0.9315	21.0	567	1.9455	0.2889
0.9127	22.0	594	1.9720	0.1778
0.9141	23.0	621	1.8863	0.4222
0.8941	24.0	648	2.4630	0.2444
0.861	25.0	675	2.3990	0.2
0.8474	26.0	702	2.1204	0.3556
0.7937	27.0	729	2.7394	0.3556
0.7958	28.0	756	2.5648	0.2
0.7373	29.0	783	2.5253	0.3778
0.7358	30.0	810	2.5059	0.3778
0.691	31.0	837	2.3895	0.4222
0.7103	32.0	864	2.5414	0.4222
0.6539	33.0	891	3.0204	0.3333
0.6275	34.0	918	2.6245	0.3778
0.5921	35.0	945	3.2133	0.2667
0.5912	36.0	972	3.5251	0.2667
0.5547	37.0	999	3.3775	0.2889
0.4976	38.0	1026	3.1294	0.4
0.4303	39.0	1053	3.2846	0.3778
0.3956	40.0	1080	3.2354	0.4444
0.3999	41.0	1107	3.0834	0.4667
0.3745	42.0	1134	3.3561	0.3333
0.3219	43.0	1161	3.3246	0.3333
0.2571	44.0	1188	3.4952	0.3556
0.2544	45.0	1215	3.6528	0.3778
0.2048	46.0	1242	3.6814	0.3333
0.2017	47.0	1269	3.5396	0.3778
0.1409	48.0	1296	3.6629	0.3556
0.1528	49.0	1323	3.6556	0.3556
0.122	50.0	1350	3.6556	0.3556

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0