hushem_5x_beit_base_rms_001_fold1

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: 2.2430
• Accuracy: 0.4444

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
1.5782	1.0	27	1.4061	0.2444
1.4004	2.0	54	1.4559	0.2444
1.3873	3.0	81	1.4120	0.2444
1.3666	4.0	108	1.6275	0.2444
1.3597	5.0	135	1.4398	0.2444
1.2814	6.0	162	1.5328	0.2444
1.2056	7.0	189	1.5389	0.2
1.1635	8.0	216	1.5332	0.2444
1.1235	9.0	243	1.6681	0.2444
1.1484	10.0	270	1.6176	0.2667
1.1757	11.0	297	1.6312	0.2444
1.1297	12.0	324	1.5067	0.2444
1.1448	13.0	351	1.5657	0.2444
1.1725	14.0	378	1.5184	0.1556
1.1591	15.0	405	1.5790	0.2444
1.1549	16.0	432	1.5501	0.2444
1.0865	17.0	459	1.5776	0.2444
1.1351	18.0	486	1.6195	0.3111
1.0974	19.0	513	1.5360	0.2444
1.0992	20.0	540	1.5742	0.3111
1.0894	21.0	567	1.4918	0.3778
1.0557	22.0	594	1.5742	0.2444
1.0574	23.0	621	1.5043	0.4222
1.0148	24.0	648	1.3535	0.4222
1.1133	25.0	675	1.4897	0.4
1.02	26.0	702	1.4554	0.4222
1.0107	27.0	729	1.4238	0.4
0.9307	28.0	756	1.7644	0.3556
0.8335	29.0	783	2.0253	0.3556
0.8203	30.0	810	1.7990	0.3556
0.7263	31.0	837	1.6909	0.3778
0.8387	32.0	864	1.4758	0.4
0.6837	33.0	891	2.1584	0.3556
0.7155	34.0	918	1.7102	0.3778
0.6349	35.0	945	1.1875	0.4667
0.6331	36.0	972	1.9965	0.4222
0.5871	37.0	999	1.7881	0.4
0.595	38.0	1026	1.7629	0.4
0.5266	39.0	1053	1.6720	0.4222
0.4985	40.0	1080	2.3229	0.4222
0.4855	41.0	1107	1.6470	0.4444
0.503	42.0	1134	1.7515	0.4667
0.4432	43.0	1161	2.0538	0.4222
0.3668	44.0	1188	2.1471	0.4444
0.3654	45.0	1215	2.0004	0.4444
0.3317	46.0	1242	2.1973	0.4444
0.2413	47.0	1269	2.2882	0.4444
0.2395	48.0	1296	2.2389	0.4444
0.2502	49.0	1323	2.2430	0.4444
0.237	50.0	1350	2.2430	0.4444

Framework versions

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