hushem_1x_deit_base_rms_001_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: 1.5945
• Accuracy: 0.4186

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	5.1656	0.2326
5.6353	2.0	12	2.9961	0.2558
5.6353	3.0	18	1.8729	0.2558
2.0815	4.0	24	2.9243	0.2558
1.6692	5.0	30	1.6813	0.2558
1.6692	6.0	36	1.4288	0.2558
1.5305	7.0	42	1.5132	0.2326
1.5305	8.0	48	1.7063	0.2558
1.5248	9.0	54	1.4498	0.2558
1.47	10.0	60	1.4163	0.2558
1.47	11.0	66	1.5259	0.2558
1.4904	12.0	72	1.3986	0.2326
1.4904	13.0	78	1.4224	0.2558
1.455	14.0	84	1.4163	0.2558
1.5854	15.0	90	1.3942	0.2558
1.5854	16.0	96	1.4547	0.2326
1.4305	17.0	102	1.3943	0.2558
1.4305	18.0	108	1.4560	0.2558
1.3943	19.0	114	1.3964	0.3023
1.4034	20.0	120	1.3547	0.3721
1.4034	21.0	126	2.6056	0.2791
1.3234	22.0	132	1.4424	0.3721
1.3234	23.0	138	1.4761	0.2558
1.2686	24.0	144	1.4102	0.3488
1.2011	25.0	150	1.4342	0.2791
1.2011	26.0	156	1.3674	0.2791
1.1732	27.0	162	2.0106	0.3488
1.1732	28.0	168	1.4114	0.3488
1.1299	29.0	174	1.4639	0.3488
1.1039	30.0	180	1.3928	0.3256
1.1039	31.0	186	1.5567	0.2791
1.099	32.0	192	1.3821	0.3488
1.099	33.0	198	1.4133	0.3023
1.0136	34.0	204	1.5753	0.3721
1.0481	35.0	210	1.4640	0.3953
1.0481	36.0	216	1.4956	0.3023
0.9705	37.0	222	1.4443	0.3488
0.9705	38.0	228	1.4615	0.3256
0.8983	39.0	234	1.4941	0.4186
0.899	40.0	240	1.5259	0.3488
0.899	41.0	246	1.5855	0.4419
0.8181	42.0	252	1.5945	0.4186
0.8181	43.0	258	1.5945	0.4186
0.8111	44.0	264	1.5945	0.4186
0.8316	45.0	270	1.5945	0.4186
0.8316	46.0	276	1.5945	0.4186
0.807	47.0	282	1.5945	0.4186
0.807	48.0	288	1.5945	0.4186
0.8545	49.0	294	1.5945	0.4186
0.798	50.0	300	1.5945	0.4186

Framework versions

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