hushem_5x_beit_base_rms_0001_fold2

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: 6.9592
• Accuracy: 0.5111

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: 0.0001
• 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
1.3984	1.0	27	1.3816	0.2889
1.3107	2.0	54	1.6609	0.2889
1.229	3.0	81	1.5449	0.2889
1.3814	4.0	108	1.6341	0.2889
1.2031	5.0	135	1.4184	0.2667
1.1619	6.0	162	1.4603	0.2889
1.1757	7.0	189	1.4200	0.2889
1.1575	8.0	216	1.3581	0.2889
1.0419	9.0	243	1.5164	0.4
1.0334	10.0	270	1.3939	0.4889
0.799	11.0	297	1.4216	0.5333
0.7589	12.0	324	1.5018	0.5111
0.7466	13.0	351	1.2714	0.3778
0.7077	14.0	378	1.2899	0.4
0.7022	15.0	405	1.4427	0.3333
0.6019	16.0	432	1.5793	0.4
0.6413	17.0	459	1.5251	0.3111
0.6003	18.0	486	2.0148	0.4889
0.5924	19.0	513	2.2670	0.4889
0.5357	20.0	540	2.0323	0.3556
0.5196	21.0	567	2.5285	0.4889
0.5137	22.0	594	3.7709	0.4222
0.4488	23.0	621	3.1001	0.5111
0.4667	24.0	648	2.9452	0.4
0.3277	25.0	675	2.8861	0.4667
0.3619	26.0	702	3.3939	0.5111
0.3379	27.0	729	3.5247	0.5333
0.2572	28.0	756	4.2104	0.5111
0.2257	29.0	783	3.4821	0.4889
0.2189	30.0	810	3.8860	0.4667
0.1431	31.0	837	5.2772	0.4667
0.2402	32.0	864	6.2470	0.4222
0.122	33.0	891	5.2693	0.4
0.2017	34.0	918	6.0732	0.5111
0.0844	35.0	945	6.0091	0.5556
0.1316	36.0	972	6.1584	0.4889
0.0377	37.0	999	7.3245	0.4889
0.1128	38.0	1026	6.6950	0.4444
0.0551	39.0	1053	7.0821	0.5111
0.0382	40.0	1080	7.5961	0.4889
0.0547	41.0	1107	6.2914	0.5111
0.0128	42.0	1134	6.4101	0.4889
0.0359	43.0	1161	6.6377	0.5111
0.004	44.0	1188	6.6707	0.4889
0.0224	45.0	1215	7.0078	0.4889
0.0292	46.0	1242	6.9800	0.4889
0.0156	47.0	1269	6.9010	0.4889
0.0096	48.0	1296	6.9583	0.5111
0.0108	49.0	1323	6.9592	0.5111
0.0394	50.0	1350	6.9592	0.5111

Framework versions

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