hushem_1x_deit_small_rms_001_fold1

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.5587
• Accuracy: 0.3556

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	5.6616	0.2444
4.5403	2.0	12	1.9139	0.2444
4.5403	3.0	18	1.7372	0.2444
1.8724	4.0	24	1.4323	0.2667
1.5505	5.0	30	1.5541	0.2444
1.5505	6.0	36	1.5305	0.2444
1.4992	7.0	42	1.5286	0.2444
1.4992	8.0	48	1.5617	0.2444
1.4899	9.0	54	1.4717	0.2444
1.4501	10.0	60	1.4440	0.2444
1.4501	11.0	66	1.4155	0.2667
1.4052	12.0	72	1.3606	0.2444
1.4052	13.0	78	1.4215	0.3333
1.4555	14.0	84	1.3356	0.3333
1.4209	15.0	90	1.4688	0.2667
1.4209	16.0	96	1.2956	0.4444
1.4079	17.0	102	1.4012	0.2444
1.4079	18.0	108	1.4817	0.2444
1.4101	19.0	114	1.4296	0.2667
1.6129	20.0	120	1.5601	0.2444
1.6129	21.0	126	1.8216	0.2667
1.5349	22.0	132	1.6109	0.2667
1.5349	23.0	138	1.6663	0.2444
1.4443	24.0	144	1.4166	0.2444
1.3949	25.0	150	1.5159	0.2444
1.3949	26.0	156	1.5557	0.2444
1.2549	27.0	162	1.2710	0.3333
1.2549	28.0	168	1.4661	0.3333
1.2756	29.0	174	1.3759	0.3111
1.2244	30.0	180	1.3243	0.4222
1.2244	31.0	186	1.1877	0.4222
1.1482	32.0	192	1.1943	0.4667
1.1482	33.0	198	1.3644	0.3111
1.0904	34.0	204	1.3812	0.3778
1.051	35.0	210	1.3131	0.4444
1.051	36.0	216	1.7518	0.2667
1.0583	37.0	222	1.8440	0.3556
1.0583	38.0	228	1.7450	0.2889
0.8766	39.0	234	1.5767	0.3556
0.9084	40.0	240	1.5052	0.3778
0.9084	41.0	246	1.5534	0.3556
0.8553	42.0	252	1.5587	0.3556
0.8553	43.0	258	1.5587	0.3556
0.8404	44.0	264	1.5587	0.3556
0.8432	45.0	270	1.5587	0.3556
0.8432	46.0	276	1.5587	0.3556
0.8133	47.0	282	1.5587	0.3556
0.8133	48.0	288	1.5587	0.3556
0.8467	49.0	294	1.5587	0.3556
0.8396	50.0	300	1.5587	0.3556

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1