hushem_1x_deit_base_rms_001_fold2

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

• Loss: 1.0357
• Accuracy: 0.6444

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	5.4375	0.2667
5.152	2.0	12	2.4176	0.2444
5.152	3.0	18	1.6834	0.2444
2.2029	4.0	24	1.5993	0.2667
1.6431	5.0	30	1.5694	0.2444
1.6431	6.0	36	1.6003	0.2444
1.5667	7.0	42	1.5690	0.2444
1.5667	8.0	48	1.5571	0.2444
1.5065	9.0	54	1.4670	0.2444
1.4556	10.0	60	1.4809	0.2444
1.4556	11.0	66	1.4913	0.2667
1.4366	12.0	72	1.4381	0.2444
1.4366	13.0	78	1.5011	0.2667
1.45	14.0	84	1.6192	0.2667
1.6105	15.0	90	1.3933	0.2667
1.6105	16.0	96	1.3754	0.3556
1.463	17.0	102	1.6119	0.2444
1.463	18.0	108	1.4972	0.2444
1.4133	19.0	114	1.2907	0.3111
1.3552	20.0	120	1.3783	0.2667
1.3552	21.0	126	1.2531	0.4
1.2635	22.0	132	1.2107	0.4222
1.2635	23.0	138	1.2781	0.3778
1.2442	24.0	144	1.1028	0.4222
1.1223	25.0	150	1.1738	0.4444
1.1223	26.0	156	1.1566	0.5111
1.0131	27.0	162	1.0937	0.5111
1.0131	28.0	168	1.0849	0.5556
0.9912	29.0	174	1.3429	0.5111
0.853	30.0	180	0.9919	0.6222
0.853	31.0	186	1.0799	0.5556
0.6912	32.0	192	1.1042	0.5333
0.6912	33.0	198	1.0782	0.5556
0.6669	34.0	204	0.9785	0.5333
0.5453	35.0	210	1.1312	0.6444
0.5453	36.0	216	1.0910	0.5556
0.5668	37.0	222	1.1103	0.6
0.5668	38.0	228	1.1358	0.5778
0.4266	39.0	234	1.0340	0.6222
0.471	40.0	240	1.0428	0.6222
0.471	41.0	246	1.0358	0.6444
0.4178	42.0	252	1.0357	0.6444
0.4178	43.0	258	1.0357	0.6444
0.3636	44.0	264	1.0357	0.6444
0.3974	45.0	270	1.0357	0.6444
0.3974	46.0	276	1.0357	0.6444
0.3949	47.0	282	1.0357	0.6444
0.3949	48.0	288	1.0357	0.6444
0.3754	49.0	294	1.0357	0.6444
0.3739	50.0	300	1.0357	0.6444

Framework versions

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