smids_3x_deit_base_rms_0001_fold4

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: 1.2420
• Accuracy: 0.8867

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
0.3339	1.0	225	0.3477	0.8767
0.1864	2.0	450	0.4551	0.875
0.1083	3.0	675	0.7108	0.8417
0.1229	4.0	900	0.4781	0.8883
0.0535	5.0	1125	0.6232	0.8733
0.0735	6.0	1350	0.5812	0.88
0.0369	7.0	1575	0.6859	0.8667
0.055	8.0	1800	0.7382	0.8417
0.0533	9.0	2025	0.7001	0.8833
0.0337	10.0	2250	0.7953	0.855
0.0039	11.0	2475	0.6932	0.8783
0.0367	12.0	2700	0.7006	0.8733
0.0008	13.0	2925	0.6228	0.895
0.0352	14.0	3150	0.6483	0.88
0.0536	15.0	3375	0.7338	0.8733
0.0282	16.0	3600	0.7455	0.8667
0.0418	17.0	3825	0.6314	0.8833
0.0045	18.0	4050	0.8581	0.8833
0.0013	19.0	4275	0.8791	0.8717
0.0236	20.0	4500	0.8293	0.8733
0.0003	21.0	4725	0.9177	0.87
0.0214	22.0	4950	0.9227	0.8633
0.0034	23.0	5175	0.7968	0.8867
0.0319	24.0	5400	0.8643	0.885
0.0068	25.0	5625	0.8481	0.88
0.0008	26.0	5850	1.0685	0.8733
0.0	27.0	6075	0.9795	0.88
0.016	28.0	6300	0.9377	0.87
0.0001	29.0	6525	1.0537	0.855
0.0	30.0	6750	1.0268	0.865
0.0184	31.0	6975	0.8036	0.8733
0.0	32.0	7200	0.9837	0.88
0.0	33.0	7425	0.9530	0.8917
0.0	34.0	7650	1.0294	0.8883
0.0	35.0	7875	1.0787	0.8817
0.0	36.0	8100	1.0515	0.885
0.0	37.0	8325	1.0565	0.8867
0.0	38.0	8550	1.0716	0.89
0.0	39.0	8775	1.0979	0.885
0.0	40.0	9000	1.1315	0.8867
0.0	41.0	9225	1.1955	0.8867
0.0034	42.0	9450	1.1923	0.8933
0.0	43.0	9675	1.1858	0.8917
0.0029	44.0	9900	1.2299	0.89
0.0	45.0	10125	1.2340	0.89
0.0	46.0	10350	1.2528	0.885
0.0	47.0	10575	1.2511	0.8867
0.0	48.0	10800	1.2410	0.8867
0.0	49.0	11025	1.2401	0.8867
0.0	50.0	11250	1.2420	0.8867

Framework versions

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