hushem_1x_deit_base_sgd_0001_fold5

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: 1.3737
• Accuracy: 0.1951

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	1.3746	0.2195
1.4522	2.0	12	1.3746	0.2195
1.4522	3.0	18	1.3745	0.2195
1.4442	4.0	24	1.3745	0.2195
1.4519	5.0	30	1.3744	0.2195
1.4519	6.0	36	1.3744	0.2195
1.44	7.0	42	1.3744	0.1951
1.44	8.0	48	1.3743	0.1951
1.4684	9.0	54	1.3743	0.1951
1.4551	10.0	60	1.3742	0.1951
1.4551	11.0	66	1.3742	0.1951
1.4522	12.0	72	1.3742	0.1951
1.4522	13.0	78	1.3741	0.1951
1.4475	14.0	84	1.3741	0.1951
1.4669	15.0	90	1.3741	0.1951
1.4669	16.0	96	1.3740	0.1951
1.4488	17.0	102	1.3740	0.1951
1.4488	18.0	108	1.3740	0.1951
1.4465	19.0	114	1.3739	0.1951
1.4496	20.0	120	1.3739	0.1951
1.4496	21.0	126	1.3739	0.1951
1.462	22.0	132	1.3739	0.1951
1.462	23.0	138	1.3739	0.1951
1.4476	24.0	144	1.3738	0.1951
1.4485	25.0	150	1.3738	0.1951
1.4485	26.0	156	1.3738	0.1951
1.4497	27.0	162	1.3738	0.1951
1.4497	28.0	168	1.3738	0.1951
1.4444	29.0	174	1.3737	0.1951
1.4564	30.0	180	1.3737	0.1951
1.4564	31.0	186	1.3737	0.1951
1.4483	32.0	192	1.3737	0.1951
1.4483	33.0	198	1.3737	0.1951
1.4659	34.0	204	1.3737	0.1951
1.459	35.0	210	1.3737	0.1951
1.459	36.0	216	1.3737	0.1951
1.4493	37.0	222	1.3737	0.1951
1.4493	38.0	228	1.3737	0.1951
1.455	39.0	234	1.3737	0.1951
1.4497	40.0	240	1.3737	0.1951
1.4497	41.0	246	1.3737	0.1951
1.4461	42.0	252	1.3737	0.1951
1.4461	43.0	258	1.3737	0.1951
1.4457	44.0	264	1.3737	0.1951
1.446	45.0	270	1.3737	0.1951
1.446	46.0	276	1.3737	0.1951
1.4334	47.0	282	1.3737	0.1951
1.4334	48.0	288	1.3737	0.1951
1.466	49.0	294	1.3737	0.1951
1.4481	50.0	300	1.3737	0.1951

Framework versions

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