smids_3x_deit_base_sgd_0001_fold1

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: 0.6061
• Accuracy: 0.7930

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
1.1207	1.0	226	1.1055	0.3255
1.0758	2.0	452	1.0934	0.3472
1.0718	3.0	678	1.0808	0.3773
1.0624	4.0	904	1.0677	0.4090
1.0382	5.0	1130	1.0537	0.4591
1.0247	6.0	1356	1.0386	0.4925
1.0174	7.0	1582	1.0217	0.5209
0.9921	8.0	1808	1.0035	0.5576
0.9451	9.0	2034	0.9830	0.5910
0.9379	10.0	2260	0.9606	0.6260
0.9321	11.0	2486	0.9380	0.6427
0.8822	12.0	2712	0.9157	0.6661
0.8728	13.0	2938	0.8943	0.6778
0.8476	14.0	3164	0.8736	0.6912
0.8717	15.0	3390	0.8536	0.7062
0.8185	16.0	3616	0.8345	0.7229
0.7787	17.0	3842	0.8163	0.7295
0.8087	18.0	4068	0.7993	0.7396
0.791	19.0	4294	0.7835	0.7429
0.7857	20.0	4520	0.7684	0.7462
0.7846	21.0	4746	0.7543	0.7513
0.7633	22.0	4972	0.7411	0.7513
0.7242	23.0	5198	0.7289	0.7546
0.7569	24.0	5424	0.7173	0.7546
0.7119	25.0	5650	0.7067	0.7546
0.7356	26.0	5876	0.6968	0.7629
0.6807	27.0	6102	0.6877	0.7646
0.7192	28.0	6328	0.6794	0.7696
0.658	29.0	6554	0.6715	0.7713
0.7338	30.0	6780	0.6643	0.7713
0.6572	31.0	7006	0.6576	0.7746
0.6818	32.0	7232	0.6514	0.7796
0.6397	33.0	7458	0.6458	0.7780
0.6265	34.0	7684	0.6406	0.7813
0.6395	35.0	7910	0.6359	0.7846
0.6047	36.0	8136	0.6316	0.7863
0.6592	37.0	8362	0.6277	0.7846
0.6634	38.0	8588	0.6242	0.7863
0.6398	39.0	8814	0.6210	0.7863
0.601	40.0	9040	0.6182	0.7863
0.6061	41.0	9266	0.6157	0.7863
0.6339	42.0	9492	0.6135	0.7880
0.6566	43.0	9718	0.6116	0.7896
0.5859	44.0	9944	0.6100	0.7896
0.6306	45.0	10170	0.6087	0.7896
0.6188	46.0	10396	0.6077	0.7896
0.6096	47.0	10622	0.6069	0.7930
0.6151	48.0	10848	0.6064	0.7930
0.6052	49.0	11074	0.6062	0.7930
0.6333	50.0	11300	0.6061	0.7930

Framework versions

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