hushem_40x_deit_small_sgd_00001_fold2

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.4445
• Accuracy: 0.2889

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
1.9592	1.0	215	1.6375	0.2444
1.9589	2.0	430	1.6244	0.2444
1.9405	3.0	645	1.6119	0.2444
1.8111	4.0	860	1.6000	0.2444
1.7854	5.0	1075	1.5888	0.2444
1.7837	6.0	1290	1.5782	0.2444
1.8329	7.0	1505	1.5681	0.2444
1.808	8.0	1720	1.5587	0.2444
1.8135	9.0	1935	1.5499	0.2444
1.7532	10.0	2150	1.5416	0.2444
1.7679	11.0	2365	1.5338	0.2667
1.6698	12.0	2580	1.5266	0.2667
1.7071	13.0	2795	1.5198	0.2667
1.7064	14.0	3010	1.5135	0.2667
1.6474	15.0	3225	1.5077	0.2667
1.6795	16.0	3440	1.5023	0.2667
1.6509	17.0	3655	1.4973	0.2444
1.6571	18.0	3870	1.4927	0.2444
1.6903	19.0	4085	1.4885	0.2444
1.594	20.0	4300	1.4845	0.2444
1.6043	21.0	4515	1.4809	0.2444
1.5781	22.0	4730	1.4775	0.2667
1.6042	23.0	4945	1.4744	0.2667
1.6585	24.0	5160	1.4716	0.2667
1.6133	25.0	5375	1.4689	0.2667
1.5503	26.0	5590	1.4665	0.2667
1.559	27.0	5805	1.4642	0.2889
1.5271	28.0	6020	1.4621	0.3111
1.5368	29.0	6235	1.4602	0.3111
1.5411	30.0	6450	1.4584	0.3111
1.6163	31.0	6665	1.4568	0.3111
1.5496	32.0	6880	1.4553	0.3111
1.5517	33.0	7095	1.4539	0.3111
1.5789	34.0	7310	1.4526	0.3111
1.5768	35.0	7525	1.4515	0.3111
1.5496	36.0	7740	1.4505	0.3111
1.5074	37.0	7955	1.4495	0.3111
1.5918	38.0	8170	1.4487	0.3111
1.5751	39.0	8385	1.4479	0.3111
1.5533	40.0	8600	1.4472	0.3111
1.5217	41.0	8815	1.4467	0.3111
1.5477	42.0	9030	1.4461	0.3111
1.5219	43.0	9245	1.4457	0.3111
1.5414	44.0	9460	1.4453	0.3111
1.5697	45.0	9675	1.4450	0.3111
1.5331	46.0	9890	1.4448	0.3111
1.5195	47.0	10105	1.4446	0.3111
1.5072	48.0	10320	1.4445	0.2889
1.5469	49.0	10535	1.4445	0.2889
1.5351	50.0	10750	1.4445	0.2889

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2