hushem_40x_deit_small_sgd_00001_fold1

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: 1.4582
• Accuracy: 0.2444

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
2.0145	1.0	215	1.6167	0.2444
1.9528	2.0	430	1.6051	0.2444
1.8549	3.0	645	1.5941	0.2444
1.9735	4.0	860	1.5837	0.2444
1.8613	5.0	1075	1.5740	0.2444
1.8998	6.0	1290	1.5649	0.2667
1.9805	7.0	1505	1.5563	0.2667
1.7761	8.0	1720	1.5483	0.2444
1.8452	9.0	1935	1.5408	0.2222
1.7743	10.0	2150	1.5338	0.2222
1.7542	11.0	2365	1.5272	0.2222
1.7225	12.0	2580	1.5212	0.2222
1.7844	13.0	2795	1.5156	0.2222
1.7312	14.0	3010	1.5104	0.2444
1.766	15.0	3225	1.5056	0.2222
1.6922	16.0	3440	1.5012	0.2222
1.6889	17.0	3655	1.4971	0.2222
1.6643	18.0	3870	1.4934	0.2
1.6328	19.0	4085	1.4899	0.2222
1.6408	20.0	4300	1.4868	0.2222
1.5927	21.0	4515	1.4839	0.2222
1.6137	22.0	4730	1.4813	0.2
1.623	23.0	4945	1.4789	0.1778
1.5692	24.0	5160	1.4767	0.1778
1.5514	25.0	5375	1.4747	0.1778
1.5483	26.0	5590	1.4729	0.2222
1.571	27.0	5805	1.4713	0.2222
1.5882	28.0	6020	1.4698	0.2222
1.524	29.0	6235	1.4684	0.2222
1.5611	30.0	6450	1.4672	0.2222
1.5511	31.0	6665	1.4661	0.2222
1.5655	32.0	6880	1.4650	0.2222
1.5736	33.0	7095	1.4641	0.2222
1.5317	34.0	7310	1.4633	0.2222
1.5555	35.0	7525	1.4625	0.2222
1.5608	36.0	7740	1.4619	0.2222
1.5011	37.0	7955	1.4612	0.2222
1.5571	38.0	8170	1.4607	0.2444
1.4975	39.0	8385	1.4602	0.2444
1.4908	40.0	8600	1.4598	0.2444
1.5291	41.0	8815	1.4595	0.2444
1.52	42.0	9030	1.4592	0.2444
1.5041	43.0	9245	1.4589	0.2444
1.5102	44.0	9460	1.4587	0.2444
1.5245	45.0	9675	1.4585	0.2444
1.4992	46.0	9890	1.4584	0.2444
1.4976	47.0	10105	1.4583	0.2444
1.5255	48.0	10320	1.4582	0.2444
1.4826	49.0	10535	1.4582	0.2444
1.5224	50.0	10750	1.4582	0.2444

Framework versions

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