hushem_5x_deit_base_sgd_00001_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: 1.3712
• Accuracy: 0.1707

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
1.4308	1.0	28	1.3745	0.2195
1.4381	2.0	56	1.3744	0.1951
1.4303	3.0	84	1.3742	0.1951
1.4196	4.0	112	1.3741	0.1951
1.4462	5.0	140	1.3739	0.1951
1.4289	6.0	168	1.3738	0.1951
1.4513	7.0	196	1.3737	0.1951
1.454	8.0	224	1.3736	0.1951
1.4395	9.0	252	1.3734	0.1951
1.4245	10.0	280	1.3733	0.1951
1.4477	11.0	308	1.3732	0.1951
1.4479	12.0	336	1.3731	0.1951
1.4264	13.0	364	1.3730	0.1951
1.4179	14.0	392	1.3729	0.1951
1.4497	15.0	420	1.3728	0.1951
1.4379	16.0	448	1.3727	0.1951
1.4414	17.0	476	1.3726	0.1951
1.452	18.0	504	1.3725	0.1951
1.4605	19.0	532	1.3724	0.1951
1.4508	20.0	560	1.3723	0.1951
1.4355	21.0	588	1.3722	0.1951
1.4232	22.0	616	1.3721	0.1951
1.4314	23.0	644	1.3721	0.1951
1.4464	24.0	672	1.3720	0.1951
1.4347	25.0	700	1.3719	0.1951
1.4331	26.0	728	1.3719	0.1707
1.4315	27.0	756	1.3718	0.1707
1.4463	28.0	784	1.3717	0.1707
1.4461	29.0	812	1.3717	0.1707
1.4576	30.0	840	1.3716	0.1707
1.4346	31.0	868	1.3716	0.1707
1.4439	32.0	896	1.3715	0.1707
1.4382	33.0	924	1.3715	0.1707
1.4458	34.0	952	1.3715	0.1707
1.4323	35.0	980	1.3714	0.1707
1.4333	36.0	1008	1.3714	0.1707
1.4238	37.0	1036	1.3714	0.1707
1.4188	38.0	1064	1.3713	0.1707
1.4355	39.0	1092	1.3713	0.1707
1.4499	40.0	1120	1.3713	0.1707
1.4289	41.0	1148	1.3713	0.1707
1.4376	42.0	1176	1.3712	0.1707
1.4427	43.0	1204	1.3712	0.1707
1.4421	44.0	1232	1.3712	0.1707
1.4536	45.0	1260	1.3712	0.1707
1.4126	46.0	1288	1.3712	0.1707
1.4355	47.0	1316	1.3712	0.1707
1.4384	48.0	1344	1.3712	0.1707
1.431	49.0	1372	1.3712	0.1707
1.4525	50.0	1400	1.3712	0.1707

Framework versions

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