hushem_5x_deit_base_sgd_0001_fold2

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: 1.4239
• Accuracy: 0.1333

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: 0.0001
• 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
1.43	1.0	27	1.4766	0.1333
1.4102	2.0	54	1.4737	0.1333
1.4007	3.0	81	1.4709	0.1333
1.4122	4.0	108	1.4682	0.1333
1.4031	5.0	135	1.4658	0.1333
1.4094	6.0	162	1.4633	0.1333
1.3941	7.0	189	1.4611	0.1333
1.4105	8.0	216	1.4588	0.1333
1.4006	9.0	243	1.4567	0.1333
1.3895	10.0	270	1.4547	0.1333
1.3922	11.0	297	1.4528	0.1333
1.3661	12.0	324	1.4510	0.1333
1.397	13.0	351	1.4492	0.1333
1.3778	14.0	378	1.4476	0.1333
1.3888	15.0	405	1.4461	0.1333
1.3865	16.0	432	1.4446	0.1333
1.3782	17.0	459	1.4432	0.1333
1.3766	18.0	486	1.4418	0.1333
1.3767	19.0	513	1.4404	0.1333
1.3782	20.0	540	1.4392	0.1333
1.3664	21.0	567	1.4381	0.1333
1.3644	22.0	594	1.4370	0.1333
1.386	23.0	621	1.4359	0.1333
1.3679	24.0	648	1.4349	0.1333
1.3604	25.0	675	1.4339	0.1333
1.3727	26.0	702	1.4330	0.1333
1.3624	27.0	729	1.4321	0.1333
1.3512	28.0	756	1.4313	0.1333
1.3641	29.0	783	1.4305	0.1333
1.3697	30.0	810	1.4298	0.1333
1.3661	31.0	837	1.4292	0.1333
1.3762	32.0	864	1.4286	0.1333
1.3653	33.0	891	1.4280	0.1333
1.3526	34.0	918	1.4274	0.1333
1.3565	35.0	945	1.4269	0.1333
1.3671	36.0	972	1.4265	0.1333
1.3721	37.0	999	1.4261	0.1333
1.3579	38.0	1026	1.4257	0.1333
1.3662	39.0	1053	1.4254	0.1333
1.3491	40.0	1080	1.4250	0.1333
1.3508	41.0	1107	1.4248	0.1333
1.3555	42.0	1134	1.4245	0.1333
1.3427	43.0	1161	1.4244	0.1333
1.3543	44.0	1188	1.4242	0.1333
1.3592	45.0	1215	1.4241	0.1333
1.3632	46.0	1242	1.4240	0.1333
1.3606	47.0	1269	1.4239	0.1333
1.3593	48.0	1296	1.4239	0.1333
1.3726	49.0	1323	1.4239	0.1333
1.3608	50.0	1350	1.4239	0.1333

Framework versions

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