hushem_1x_deit_base_sgd_0001_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: 1.3860
• Accuracy: 0.3721

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	1.3866	0.3721
1.4431	2.0	12	1.3866	0.3721
1.4431	3.0	18	1.3866	0.3721
1.4376	4.0	24	1.3865	0.3721
1.4767	5.0	30	1.3865	0.3721
1.4767	6.0	36	1.3865	0.3721
1.4576	7.0	42	1.3864	0.3721
1.4576	8.0	48	1.3864	0.3721
1.4525	9.0	54	1.3864	0.3721
1.436	10.0	60	1.3864	0.3721
1.436	11.0	66	1.3863	0.3721
1.4515	12.0	72	1.3863	0.3721
1.4515	13.0	78	1.3863	0.3721
1.4484	14.0	84	1.3863	0.3721
1.4554	15.0	90	1.3862	0.3721
1.4554	16.0	96	1.3862	0.3721
1.4506	17.0	102	1.3862	0.3721
1.4506	18.0	108	1.3862	0.3721
1.4516	19.0	114	1.3862	0.3721
1.4451	20.0	120	1.3861	0.3721
1.4451	21.0	126	1.3861	0.3721
1.4507	22.0	132	1.3861	0.3721
1.4507	23.0	138	1.3861	0.3721
1.4465	24.0	144	1.3861	0.3721
1.4556	25.0	150	1.3861	0.3721
1.4556	26.0	156	1.3861	0.3721
1.4543	27.0	162	1.3861	0.3721
1.4543	28.0	168	1.3860	0.3721
1.4297	29.0	174	1.3860	0.3721
1.4518	30.0	180	1.3860	0.3721
1.4518	31.0	186	1.3860	0.3721
1.4478	32.0	192	1.3860	0.3721
1.4478	33.0	198	1.3860	0.3721
1.4514	34.0	204	1.3860	0.3721
1.4485	35.0	210	1.3860	0.3721
1.4485	36.0	216	1.3860	0.3721
1.4495	37.0	222	1.3860	0.3721
1.4495	38.0	228	1.3860	0.3721
1.4541	39.0	234	1.3860	0.3721
1.4436	40.0	240	1.3860	0.3721
1.4436	41.0	246	1.3860	0.3721
1.4487	42.0	252	1.3860	0.3721
1.4487	43.0	258	1.3860	0.3721
1.4524	44.0	264	1.3860	0.3721
1.437	45.0	270	1.3860	0.3721
1.437	46.0	276	1.3860	0.3721
1.4494	47.0	282	1.3860	0.3721
1.4494	48.0	288	1.3860	0.3721
1.4556	49.0	294	1.3860	0.3721
1.4483	50.0	300	1.3860	0.3721

Framework versions

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