hushem_1x_beit_base_rms_001_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: 1.5076
• 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
No log	1.0	6	3.8156	0.2444
3.6272	2.0	12	1.5830	0.2444
3.6272	3.0	18	1.3920	0.3556
1.5643	4.0	24	1.3724	0.4667
1.4312	5.0	30	1.3703	0.4
1.4312	6.0	36	1.5314	0.2444
1.3956	7.0	42	1.3355	0.3111
1.3956	8.0	48	1.4411	0.3111
1.3396	9.0	54	1.4040	0.2667
1.3553	10.0	60	1.2838	0.5111
1.3553	11.0	66	1.2321	0.4889
1.3209	12.0	72	1.1729	0.4667
1.3209	13.0	78	1.2333	0.3556
1.2531	14.0	84	1.5215	0.3111
1.2528	15.0	90	1.2938	0.3778
1.2528	16.0	96	1.3032	0.3333
1.2233	17.0	102	1.2266	0.4667
1.2233	18.0	108	1.4102	0.3778
1.176	19.0	114	1.5633	0.2889
1.185	20.0	120	1.2459	0.4444
1.185	21.0	126	1.2338	0.4444
1.1231	22.0	132	1.2463	0.3333
1.1231	23.0	138	1.1738	0.5333
1.1314	24.0	144	1.3614	0.3778
1.083	25.0	150	1.1802	0.5556
1.083	26.0	156	1.2064	0.4444
1.0558	27.0	162	1.2033	0.5111
1.0558	28.0	168	1.2614	0.4444
1.0238	29.0	174	1.1933	0.4667
1.0431	30.0	180	1.2411	0.3333
1.0431	31.0	186	1.1551	0.5111
0.9703	32.0	192	1.2204	0.5778
0.9703	33.0	198	1.2367	0.4667
0.9502	34.0	204	1.2552	0.4222
0.8685	35.0	210	1.2938	0.4889
0.8685	36.0	216	1.3314	0.4889
0.8441	37.0	222	1.4307	0.5111
0.8441	38.0	228	1.3667	0.5111
0.7558	39.0	234	1.4572	0.4444
0.7537	40.0	240	1.4773	0.4667
0.7537	41.0	246	1.5046	0.4222
0.7017	42.0	252	1.5076	0.4222
0.7017	43.0	258	1.5076	0.4222
0.7145	44.0	264	1.5076	0.4222
0.7028	45.0	270	1.5076	0.4222
0.7028	46.0	276	1.5076	0.4222
0.6738	47.0	282	1.5076	0.4222
0.6738	48.0	288	1.5076	0.4222
0.7171	49.0	294	1.5076	0.4222
0.6462	50.0	300	1.5076	0.4222

Framework versions

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