hushem_1x_beit_base_rms_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: 1.6554
• 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	4.7780	0.2444
5.2713	2.0	12	1.6054	0.2444
5.2713	3.0	18	1.4082	0.2444
1.584	4.0	24	1.4460	0.2444
1.4611	5.0	30	1.3995	0.2444
1.4611	6.0	36	1.4702	0.2444
1.426	7.0	42	1.4146	0.2889
1.426	8.0	48	1.4135	0.2667
1.4231	9.0	54	1.3923	0.2444
1.3996	10.0	60	1.3952	0.3333
1.3996	11.0	66	1.3447	0.3111
1.3678	12.0	72	1.2012	0.4667
1.3678	13.0	78	1.2676	0.3111
1.3823	14.0	84	1.6735	0.2444
1.3117	15.0	90	1.4063	0.2889
1.3117	16.0	96	1.1481	0.5111
1.2447	17.0	102	1.1199	0.4444
1.2447	18.0	108	1.4527	0.3556
1.159	19.0	114	1.2632	0.4
1.1248	20.0	120	1.8823	0.3111
1.1248	21.0	126	1.3002	0.3556
1.076	22.0	132	1.3851	0.3333
1.076	23.0	138	1.5013	0.4444
1.017	24.0	144	1.6658	0.3556
1.0181	25.0	150	1.6317	0.3556
1.0181	26.0	156	2.0640	0.3556
0.9565	27.0	162	1.8453	0.3778
0.9565	28.0	168	1.5546	0.3556
0.9059	29.0	174	1.7001	0.4222
0.8651	30.0	180	1.6735	0.4444
0.8651	31.0	186	1.6947	0.4667
0.8326	32.0	192	1.8722	0.3333
0.8326	33.0	198	1.5166	0.4667
0.7811	34.0	204	1.5728	0.4222
0.7343	35.0	210	1.5820	0.4444
0.7343	36.0	216	1.5827	0.4444
0.7013	37.0	222	1.8267	0.4
0.7013	38.0	228	1.7339	0.3778
0.6091	39.0	234	1.6279	0.4222
0.6541	40.0	240	1.6395	0.4444
0.6541	41.0	246	1.6688	0.4
0.5842	42.0	252	1.6554	0.4222
0.5842	43.0	258	1.6554	0.4222
0.591	44.0	264	1.6554	0.4222
0.5639	45.0	270	1.6554	0.4222
0.5639	46.0	276	1.6554	0.4222
0.5746	47.0	282	1.6554	0.4222
0.5746	48.0	288	1.6554	0.4222
0.5561	49.0	294	1.6554	0.4222
0.5875	50.0	300	1.6554	0.4222

Framework versions

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