hushem_5x_deit_base_adamax_0001_fold1

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

• Loss: 1.1576
• Accuracy: 0.7111

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: 1e-05
• 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
1.238	1.0	27	1.3398	0.2667
0.9269	2.0	54	1.2685	0.4444
0.6591	3.0	81	1.1740	0.5333
0.5112	4.0	108	1.1289	0.5556
0.3169	5.0	135	1.0720	0.5778
0.2415	6.0	162	0.9458	0.6
0.1769	7.0	189	0.9250	0.6
0.0983	8.0	216	0.8893	0.6667
0.0567	9.0	243	0.8959	0.7111
0.0295	10.0	270	1.0130	0.5778
0.0202	11.0	297	0.9509	0.6889
0.0113	12.0	324	0.9586	0.7111
0.0094	13.0	351	0.9844	0.6889
0.0072	14.0	378	0.9965	0.7333
0.0063	15.0	405	1.0087	0.7111
0.005	16.0	432	1.0089	0.6889
0.0041	17.0	459	1.0347	0.6889
0.004	18.0	486	1.0569	0.6889
0.0034	19.0	513	1.0522	0.6889
0.0031	20.0	540	1.0681	0.6889
0.0027	21.0	567	1.0686	0.6889
0.0026	22.0	594	1.0745	0.6889
0.0023	23.0	621	1.0948	0.6889
0.0022	24.0	648	1.0979	0.6889
0.0021	25.0	675	1.0958	0.6889
0.0021	26.0	702	1.1008	0.6889
0.0018	27.0	729	1.1079	0.6889
0.0017	28.0	756	1.1114	0.6889
0.0019	29.0	783	1.1187	0.6889
0.0017	30.0	810	1.1246	0.6889
0.0016	31.0	837	1.1229	0.6889
0.0016	32.0	864	1.1290	0.6889
0.0014	33.0	891	1.1312	0.6889
0.0015	34.0	918	1.1349	0.6889
0.0014	35.0	945	1.1402	0.6889
0.0013	36.0	972	1.1442	0.6889
0.0013	37.0	999	1.1434	0.6889
0.0012	38.0	1026	1.1425	0.7111
0.0012	39.0	1053	1.1512	0.6889
0.0011	40.0	1080	1.1497	0.6889
0.0012	41.0	1107	1.1525	0.6889
0.0012	42.0	1134	1.1548	0.6889
0.0012	43.0	1161	1.1577	0.6889
0.0011	44.0	1188	1.1573	0.6889
0.0011	45.0	1215	1.1575	0.6889
0.0011	46.0	1242	1.1575	0.7111
0.0011	47.0	1269	1.1575	0.7111
0.0011	48.0	1296	1.1576	0.7111
0.0011	49.0	1323	1.1576	0.7111
0.0011	50.0	1350	1.1576	0.7111

Framework versions

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