hushem_1x_beit_base_sgd_001_fold3

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: 1.3325
• Accuracy: 0.3023

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	1.5424	0.2558
1.5472	2.0	12	1.5114	0.2326
1.5472	3.0	18	1.4888	0.2791
1.4947	4.0	24	1.4715	0.2558
1.4055	5.0	30	1.4594	0.2791
1.4055	6.0	36	1.4478	0.3256
1.389	7.0	42	1.4378	0.3023
1.389	8.0	48	1.4298	0.3256
1.3593	9.0	54	1.4226	0.3488
1.3527	10.0	60	1.4160	0.3488
1.3527	11.0	66	1.4104	0.3488
1.3352	12.0	72	1.4055	0.3488
1.3352	13.0	78	1.4005	0.3721
1.2989	14.0	84	1.3949	0.3721
1.3078	15.0	90	1.3896	0.3721
1.3078	16.0	96	1.3853	0.3721
1.2945	17.0	102	1.3807	0.3488
1.2945	18.0	108	1.3772	0.3721
1.2862	19.0	114	1.3730	0.3488
1.2665	20.0	120	1.3692	0.3488
1.2665	21.0	126	1.3663	0.3488
1.2571	22.0	132	1.3633	0.3488
1.2571	23.0	138	1.3594	0.3488
1.2478	24.0	144	1.3564	0.3488
1.2356	25.0	150	1.3537	0.3488
1.2356	26.0	156	1.3506	0.3256
1.253	27.0	162	1.3489	0.3256
1.253	28.0	168	1.3460	0.3256
1.2445	29.0	174	1.3442	0.3256
1.208	30.0	180	1.3430	0.3023
1.208	31.0	186	1.3417	0.3256
1.204	32.0	192	1.3396	0.3023
1.204	33.0	198	1.3381	0.3023
1.1994	34.0	204	1.3371	0.3023
1.1991	35.0	210	1.3357	0.3023
1.1991	36.0	216	1.3349	0.3023
1.1819	37.0	222	1.3343	0.3023
1.1819	38.0	228	1.3338	0.3023
1.1973	39.0	234	1.3332	0.3023
1.1899	40.0	240	1.3328	0.3023
1.1899	41.0	246	1.3326	0.3023
1.1779	42.0	252	1.3325	0.3023
1.1779	43.0	258	1.3325	0.3023
1.1962	44.0	264	1.3325	0.3023
1.2031	45.0	270	1.3325	0.3023
1.2031	46.0	276	1.3325	0.3023
1.1985	47.0	282	1.3325	0.3023
1.1985	48.0	288	1.3325	0.3023
1.1942	49.0	294	1.3325	0.3023
1.1653	50.0	300	1.3325	0.3023

Framework versions

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