smids_1x_deit_tiny_rms_001_fold2

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: 1.2712
• Accuracy: 0.7554

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: 0.001
• 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
1.1905	1.0	75	1.0709	0.4526
1.1189	2.0	150	0.9655	0.5241
0.9756	3.0	225	0.9009	0.4626
0.9398	4.0	300	0.8912	0.5757
0.9083	5.0	375	0.8740	0.5191
0.9033	6.0	450	0.8827	0.5092
0.8824	7.0	525	0.7939	0.5824
0.9323	8.0	600	0.8247	0.5524
0.8419	9.0	675	0.8138	0.5807
0.8164	10.0	750	0.8013	0.5674
0.859	11.0	825	0.8565	0.5458
0.7125	12.0	900	0.8056	0.6173
0.803	13.0	975	0.8062	0.6190
0.7687	14.0	1050	0.7699	0.6040
0.7356	15.0	1125	0.7299	0.6589
0.7325	16.0	1200	0.7012	0.6722
0.6848	17.0	1275	0.7155	0.6423
0.6885	18.0	1350	0.6752	0.6689
0.6698	19.0	1425	0.6669	0.6872
0.7037	20.0	1500	0.6890	0.6789
0.6983	21.0	1575	0.7382	0.6423
0.6127	22.0	1650	0.6732	0.7038
0.6242	23.0	1725	0.6106	0.7304
0.658	24.0	1800	0.6268	0.7121
0.5546	25.0	1875	0.6631	0.7088
0.5765	26.0	1950	0.6682	0.6988
0.6162	27.0	2025	0.6203	0.7304
0.5296	28.0	2100	0.6174	0.7438
0.5276	29.0	2175	0.5823	0.7371
0.4954	30.0	2250	0.7129	0.6922
0.5509	31.0	2325	0.6075	0.7404
0.4629	32.0	2400	0.6387	0.7488
0.4323	33.0	2475	0.6167	0.7421
0.4094	34.0	2550	0.6489	0.7637
0.419	35.0	2625	0.6362	0.7371
0.444	36.0	2700	0.6255	0.7621
0.4294	37.0	2775	0.6272	0.7604
0.3866	38.0	2850	0.6218	0.7770
0.3776	39.0	2925	0.6660	0.7637
0.3382	40.0	3000	0.7027	0.7720
0.3406	41.0	3075	0.7627	0.7770
0.3115	42.0	3150	0.7813	0.7737
0.2146	43.0	3225	0.8652	0.7521
0.2529	44.0	3300	0.9528	0.7504
0.1582	45.0	3375	0.9733	0.7704
0.1575	46.0	3450	1.0460	0.7704
0.1254	47.0	3525	1.1356	0.7737
0.076	48.0	3600	1.2020	0.7604
0.0433	49.0	3675	1.2652	0.7571
0.0823	50.0	3750	1.2712	0.7554

Framework versions

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