hushem_1x_beit_base_rms_0001_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: 3.2060
• Accuracy: 0.5116

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
No log	1.0	6	1.4345	0.2326
2.0035	2.0	12	1.4013	0.2558
2.0035	3.0	18	1.4133	0.2558
1.4085	4.0	24	1.4118	0.2558
1.3752	5.0	30	1.3892	0.4419
1.3752	6.0	36	1.3795	0.2791
1.3345	7.0	42	1.3859	0.3256
1.3345	8.0	48	1.3535	0.3023
1.2957	9.0	54	1.3373	0.4419
1.2266	10.0	60	1.3000	0.4651
1.2266	11.0	66	1.2541	0.4651
1.2119	12.0	72	1.3081	0.3023
1.2119	13.0	78	1.3255	0.4186
1.1642	14.0	84	1.2598	0.4419
1.0863	15.0	90	1.3634	0.4651
1.0863	16.0	96	1.2765	0.4419
1.0739	17.0	102	1.2557	0.4651
1.0739	18.0	108	1.3482	0.4651
0.9189	19.0	114	1.2441	0.5814
0.9333	20.0	120	1.3137	0.5116
0.9333	21.0	126	1.4928	0.5116
0.7984	22.0	132	1.4587	0.4419
0.7984	23.0	138	1.4263	0.4884
0.7474	24.0	144	1.3937	0.5116
0.6261	25.0	150	1.7138	0.4651
0.6261	26.0	156	1.9139	0.3488
0.6149	27.0	162	2.2211	0.4419
0.6149	28.0	168	2.6636	0.3953
0.5568	29.0	174	2.0456	0.4419
0.5749	30.0	180	2.1341	0.3488
0.5749	31.0	186	2.6940	0.4651
0.5955	32.0	192	2.3824	0.4419
0.5955	33.0	198	2.3420	0.4419
0.4884	34.0	204	2.5519	0.5116
0.4591	35.0	210	2.4344	0.4186
0.4591	36.0	216	2.3412	0.5116
0.3989	37.0	222	2.6657	0.5116
0.3989	38.0	228	3.0833	0.5116
0.2794	39.0	234	2.9344	0.5349
0.252	40.0	240	3.0820	0.5349
0.252	41.0	246	3.2011	0.5116
0.2307	42.0	252	3.2060	0.5116
0.2307	43.0	258	3.2060	0.5116
0.2027	44.0	264	3.2060	0.5116
0.2023	45.0	270	3.2060	0.5116
0.2023	46.0	276	3.2060	0.5116
0.2295	47.0	282	3.2060	0.5116
0.2295	48.0	288	3.2060	0.5116
0.216	49.0	294	3.2060	0.5116
0.2238	50.0	300	3.2060	0.5116

Framework versions

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