hushem_5x_deit_base_sgd_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: 1.3841
• 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
1.4207	1.0	28	1.3865	0.3721
1.4337	2.0	56	1.3864	0.3721
1.445	3.0	84	1.3863	0.3721
1.4489	4.0	112	1.3862	0.3721
1.3943	5.0	140	1.3861	0.3721
1.4493	6.0	168	1.3860	0.3721
1.4229	7.0	196	1.3859	0.3721
1.4563	8.0	224	1.3858	0.3721
1.4234	9.0	252	1.3857	0.3721
1.4265	10.0	280	1.3856	0.3721
1.4288	11.0	308	1.3855	0.3721
1.437	12.0	336	1.3855	0.3721
1.4189	13.0	364	1.3854	0.3721
1.4336	14.0	392	1.3853	0.3721
1.3938	15.0	420	1.3852	0.3721
1.4323	16.0	448	1.3851	0.3721
1.4267	17.0	476	1.3851	0.3721
1.4208	18.0	504	1.3850	0.3721
1.4257	19.0	532	1.3849	0.3721
1.4426	20.0	560	1.3849	0.3721
1.4493	21.0	588	1.3848	0.3721
1.4203	22.0	616	1.3848	0.3721
1.4209	23.0	644	1.3847	0.3721
1.4152	24.0	672	1.3847	0.3721
1.4253	25.0	700	1.3846	0.3721
1.4344	26.0	728	1.3846	0.3721
1.4406	27.0	756	1.3845	0.3721
1.435	28.0	784	1.3845	0.3721
1.4128	29.0	812	1.3844	0.3721
1.4483	30.0	840	1.3844	0.3721
1.4308	31.0	868	1.3844	0.3721
1.4319	32.0	896	1.3843	0.3721
1.4115	33.0	924	1.3843	0.3721
1.4269	34.0	952	1.3843	0.3721
1.4112	35.0	980	1.3842	0.3721
1.4513	36.0	1008	1.3842	0.3721
1.4288	37.0	1036	1.3842	0.3721
1.4247	38.0	1064	1.3842	0.3721
1.3988	39.0	1092	1.3842	0.3721
1.4499	40.0	1120	1.3841	0.3721
1.44	41.0	1148	1.3841	0.3721
1.4219	42.0	1176	1.3841	0.3721
1.437	43.0	1204	1.3841	0.3721
1.411	44.0	1232	1.3841	0.3721
1.4061	45.0	1260	1.3841	0.3721
1.4217	46.0	1288	1.3841	0.3721
1.4337	47.0	1316	1.3841	0.3721
1.4341	48.0	1344	1.3841	0.3721
1.415	49.0	1372	1.3841	0.3721
1.4182	50.0	1400	1.3841	0.3721

Framework versions

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