hushem_1x_beit_base_adamax_00001_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: 1.3147
• 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: 1e-05
• 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	1.3169	0.4222
1.2993	2.0	12	1.2806	0.4667
1.2993	3.0	18	1.2579	0.4889
0.9161	4.0	24	1.2167	0.4889
0.7083	5.0	30	1.1786	0.5556
0.7083	6.0	36	1.1207	0.5778
0.5194	7.0	42	1.1521	0.4667
0.5194	8.0	48	1.1594	0.4889
0.4029	9.0	54	1.0967	0.5333
0.2875	10.0	60	1.1131	0.5333
0.2875	11.0	66	1.0617	0.5556
0.2225	12.0	72	1.0697	0.5778
0.2225	13.0	78	1.1375	0.5333
0.1628	14.0	84	1.1927	0.5111
0.1384	15.0	90	1.1635	0.5556
0.1384	16.0	96	1.2191	0.5111
0.1154	17.0	102	1.1629	0.5556
0.1154	18.0	108	1.1505	0.5556
0.0953	19.0	114	1.2116	0.5333
0.0771	20.0	120	1.1885	0.5333
0.0771	21.0	126	1.2166	0.5111
0.0552	22.0	132	1.2643	0.5111
0.0552	23.0	138	1.2478	0.5111
0.0544	24.0	144	1.1768	0.5556
0.0597	25.0	150	1.1020	0.5778
0.0597	26.0	156	1.1318	0.5556
0.0518	27.0	162	1.1807	0.5333
0.0518	28.0	168	1.2523	0.5333
0.0452	29.0	174	1.2755	0.5333
0.0366	30.0	180	1.2279	0.5333
0.0366	31.0	186	1.2089	0.5333
0.037	32.0	192	1.2327	0.5333
0.037	33.0	198	1.2492	0.5111
0.0339	34.0	204	1.2215	0.5333
0.0349	35.0	210	1.2052	0.5556
0.0349	36.0	216	1.2247	0.5111
0.0299	37.0	222	1.2589	0.5111
0.0299	38.0	228	1.2861	0.5111
0.0453	39.0	234	1.2942	0.5111
0.0256	40.0	240	1.3064	0.5111
0.0256	41.0	246	1.3149	0.5111
0.0285	42.0	252	1.3147	0.5111
0.0285	43.0	258	1.3147	0.5111
0.0329	44.0	264	1.3147	0.5111
0.0329	45.0	270	1.3147	0.5111
0.0329	46.0	276	1.3147	0.5111
0.0305	47.0	282	1.3147	0.5111
0.0305	48.0	288	1.3147	0.5111
0.0327	49.0	294	1.3147	0.5111
0.0227	50.0	300	1.3147	0.5111

Framework versions

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