hushem_5x_deit_base_sgd_00001_fold4

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: 1.4123
• Accuracy: 0.1667

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
1.4346	1.0	28	1.4153	0.1667
1.4109	2.0	56	1.4151	0.1429
1.4175	3.0	84	1.4150	0.1429
1.4323	4.0	112	1.4149	0.1429
1.4583	5.0	140	1.4148	0.1429
1.4358	6.0	168	1.4146	0.1429
1.4494	7.0	196	1.4145	0.1429
1.4285	8.0	224	1.4144	0.1429
1.4244	9.0	252	1.4143	0.1429
1.4228	10.0	280	1.4142	0.1429
1.4286	11.0	308	1.4141	0.1429
1.4283	12.0	336	1.4140	0.1429
1.4138	13.0	364	1.4139	0.1429
1.4095	14.0	392	1.4138	0.1429
1.4337	15.0	420	1.4137	0.1429
1.4551	16.0	448	1.4136	0.1429
1.447	17.0	476	1.4135	0.1429
1.4176	18.0	504	1.4134	0.1667
1.4339	19.0	532	1.4134	0.1667
1.4259	20.0	560	1.4133	0.1667
1.4112	21.0	588	1.4132	0.1667
1.4108	22.0	616	1.4131	0.1667
1.4336	23.0	644	1.4131	0.1667
1.4239	24.0	672	1.4130	0.1667
1.4464	25.0	700	1.4130	0.1667
1.4393	26.0	728	1.4129	0.1667
1.4057	27.0	756	1.4129	0.1667
1.433	28.0	784	1.4128	0.1667
1.4342	29.0	812	1.4128	0.1667
1.4798	30.0	840	1.4127	0.1667
1.4269	31.0	868	1.4127	0.1667
1.432	32.0	896	1.4126	0.1667
1.4028	33.0	924	1.4126	0.1667
1.4121	34.0	952	1.4125	0.1667
1.4327	35.0	980	1.4125	0.1667
1.4094	36.0	1008	1.4125	0.1667
1.4345	37.0	1036	1.4125	0.1667
1.431	38.0	1064	1.4124	0.1667
1.4318	39.0	1092	1.4124	0.1667
1.4429	40.0	1120	1.4124	0.1667
1.4245	41.0	1148	1.4124	0.1667
1.4194	42.0	1176	1.4124	0.1667
1.4301	43.0	1204	1.4123	0.1667
1.4429	44.0	1232	1.4123	0.1667
1.4185	45.0	1260	1.4123	0.1667
1.4323	46.0	1288	1.4123	0.1667
1.4214	47.0	1316	1.4123	0.1667
1.3979	48.0	1344	1.4123	0.1667
1.4318	49.0	1372	1.4123	0.1667
1.4427	50.0	1400	1.4123	0.1667

Framework versions

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