smids_10x_deit_tiny_sgd_0001_fold1

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

• Loss: 0.4530
• Accuracy: 0.8114

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
1.0181	1.0	751	0.9693	0.5359
0.81	2.0	1502	0.8850	0.5993
0.7699	3.0	2253	0.8246	0.6377
0.6601	4.0	3004	0.7789	0.6578
0.653	5.0	3755	0.7391	0.6745
0.6463	6.0	4506	0.7047	0.6912
0.5744	7.0	5257	0.6756	0.7028
0.4963	8.0	6008	0.6490	0.7129
0.5329	9.0	6759	0.6286	0.7195
0.5165	10.0	7510	0.6094	0.7295
0.5717	11.0	8261	0.5949	0.7279
0.4844	12.0	9012	0.5809	0.7396
0.4587	13.0	9763	0.5699	0.7446
0.4195	14.0	10514	0.5589	0.7496
0.4521	15.0	11265	0.5504	0.7579
0.4327	16.0	12016	0.5411	0.7596
0.4611	17.0	12767	0.5341	0.7663
0.4248	18.0	13518	0.5294	0.7746
0.4694	19.0	14269	0.5215	0.7780
0.395	20.0	15020	0.5170	0.7880
0.3437	21.0	15771	0.5117	0.7880
0.4367	22.0	16522	0.5057	0.7947
0.3451	23.0	17273	0.5010	0.7930
0.4413	24.0	18024	0.4962	0.7930
0.3908	25.0	18775	0.4929	0.7930
0.4631	26.0	19526	0.4899	0.7930
0.3779	27.0	20277	0.4860	0.7930
0.4436	28.0	21028	0.4829	0.7963
0.3794	29.0	21779	0.4792	0.7997
0.3732	30.0	22530	0.4775	0.7963
0.3411	31.0	23281	0.4746	0.7980
0.4745	32.0	24032	0.4718	0.7980
0.4263	33.0	24783	0.4692	0.7997
0.3711	34.0	25534	0.4676	0.8030
0.3951	35.0	26285	0.4656	0.8047
0.4026	36.0	27036	0.4635	0.8047
0.4811	37.0	27787	0.4621	0.8063
0.3816	38.0	28538	0.4609	0.8063
0.2904	39.0	29289	0.4596	0.8047
0.4708	40.0	30040	0.4586	0.8097
0.3633	41.0	30791	0.4575	0.8080
0.367	42.0	31542	0.4565	0.8080
0.4048	43.0	32293	0.4557	0.8080
0.3531	44.0	33044	0.4549	0.8080
0.3608	45.0	33795	0.4542	0.8097
0.3794	46.0	34546	0.4538	0.8097
0.3429	47.0	35297	0.4534	0.8114
0.395	48.0	36048	0.4532	0.8114
0.3682	49.0	36799	0.4531	0.8114
0.3927	50.0	37550	0.4530	0.8114

Framework versions

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