hushem_1x_deit_base_adamax_00001_fold3

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: 0.6539
• Accuracy: 0.7907

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	1.3331	0.3953
1.3593	2.0	12	1.2859	0.5581
1.3593	3.0	18	1.2417	0.5581
1.2039	4.0	24	1.1932	0.5581
1.0598	5.0	30	1.1339	0.5581
1.0598	6.0	36	1.0752	0.5814
0.8965	7.0	42	1.0271	0.6977
0.8965	8.0	48	0.9915	0.6744
0.7657	9.0	54	0.9413	0.7209
0.6159	10.0	60	0.8991	0.7209
0.6159	11.0	66	0.8637	0.6744
0.485	12.0	72	0.8373	0.6977
0.485	13.0	78	0.8208	0.6977
0.3863	14.0	84	0.7902	0.7442
0.3144	15.0	90	0.7725	0.7907
0.3144	16.0	96	0.7576	0.7907
0.2438	17.0	102	0.7436	0.7907
0.2438	18.0	108	0.7294	0.8140
0.193	19.0	114	0.7217	0.8140
0.1581	20.0	120	0.7141	0.7907
0.1581	21.0	126	0.6979	0.8140
0.1306	22.0	132	0.6874	0.8140
0.1306	23.0	138	0.6970	0.8140
0.1005	24.0	144	0.6887	0.8140
0.0911	25.0	150	0.6790	0.8140
0.0911	26.0	156	0.6739	0.8140
0.0714	27.0	162	0.6752	0.7907
0.0714	28.0	168	0.6709	0.7907
0.0651	29.0	174	0.6638	0.8140
0.053	30.0	180	0.6587	0.7907
0.053	31.0	186	0.6642	0.7907
0.0459	32.0	192	0.6649	0.7907
0.0459	33.0	198	0.6606	0.7907
0.0416	34.0	204	0.6572	0.7907
0.0382	35.0	210	0.6545	0.7907
0.0382	36.0	216	0.6526	0.7907
0.0358	37.0	222	0.6540	0.7907
0.0358	38.0	228	0.6543	0.7907
0.0336	39.0	234	0.6546	0.7907
0.0331	40.0	240	0.6538	0.7907
0.0331	41.0	246	0.6539	0.7907
0.0325	42.0	252	0.6539	0.7907
0.0325	43.0	258	0.6539	0.7907
0.0333	44.0	264	0.6539	0.7907
0.0317	45.0	270	0.6539	0.7907
0.0317	46.0	276	0.6539	0.7907
0.0319	47.0	282	0.6539	0.7907
0.0319	48.0	288	0.6539	0.7907
0.033	49.0	294	0.6539	0.7907
0.0315	50.0	300	0.6539	0.7907

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.15.0