hushem_1x_beit_base_sgd_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: 1.3898
• 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	1.5634	0.2889
1.5431	2.0	12	1.5296	0.2667
1.5431	3.0	18	1.5055	0.2889
1.4519	4.0	24	1.4860	0.2889
1.4616	5.0	30	1.4682	0.3111
1.4616	6.0	36	1.4542	0.3333
1.4129	7.0	42	1.4431	0.3333
1.4129	8.0	48	1.4346	0.3333
1.3729	9.0	54	1.4281	0.3333
1.3344	10.0	60	1.4232	0.3333
1.3344	11.0	66	1.4155	0.3333
1.3342	12.0	72	1.4113	0.3556
1.3342	13.0	78	1.4084	0.3556
1.3011	14.0	84	1.4054	0.3556
1.2926	15.0	90	1.4035	0.3556
1.2926	16.0	96	1.4007	0.3556
1.2856	17.0	102	1.3978	0.3778
1.2856	18.0	108	1.3952	0.3778
1.2766	19.0	114	1.3942	0.3778
1.2702	20.0	120	1.3946	0.3778
1.2702	21.0	126	1.3932	0.4222
1.2243	22.0	132	1.3923	0.4222
1.2243	23.0	138	1.3923	0.4
1.2345	24.0	144	1.3921	0.3778
1.2029	25.0	150	1.3905	0.3778
1.2029	26.0	156	1.3901	0.3778
1.2078	27.0	162	1.3891	0.4
1.2078	28.0	168	1.3898	0.4
1.189	29.0	174	1.3895	0.4
1.1978	30.0	180	1.3896	0.4
1.1978	31.0	186	1.3902	0.4
1.1767	32.0	192	1.3905	0.4222
1.1767	33.0	198	1.3902	0.4222
1.1816	34.0	204	1.3893	0.4222
1.1968	35.0	210	1.3893	0.4222
1.1968	36.0	216	1.3899	0.4222
1.1681	37.0	222	1.3900	0.4222
1.1681	38.0	228	1.3901	0.4222
1.1768	39.0	234	1.3901	0.4222
1.1628	40.0	240	1.3901	0.4222
1.1628	41.0	246	1.3899	0.4222
1.1656	42.0	252	1.3898	0.4222
1.1656	43.0	258	1.3898	0.4222
1.1713	44.0	264	1.3898	0.4222
1.1727	45.0	270	1.3898	0.4222
1.1727	46.0	276	1.3898	0.4222
1.1615	47.0	282	1.3898	0.4222
1.1615	48.0	288	1.3898	0.4222
1.1754	49.0	294	1.3898	0.4222
1.1668	50.0	300	1.3898	0.4222

Framework versions

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