smids_3x_beit_base_rms_0001_fold4

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

• Loss: 1.8372
• Accuracy: 0.81

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
0.787	1.0	225	0.7726	0.5917
0.7085	2.0	450	0.7059	0.6467
0.6145	3.0	675	0.6591	0.6983
0.6005	4.0	900	0.5422	0.7817
0.572	5.0	1125	0.5970	0.7567
0.4372	6.0	1350	0.5610	0.785
0.3918	7.0	1575	0.5957	0.7917
0.4058	8.0	1800	0.5296	0.7933
0.3971	9.0	2025	0.6041	0.7833
0.3274	10.0	2250	0.5347	0.8
0.2417	11.0	2475	0.6768	0.785
0.1989	12.0	2700	0.6501	0.8133
0.2222	13.0	2925	0.6337	0.7933
0.1654	14.0	3150	0.7865	0.7867
0.1241	15.0	3375	0.7840	0.8033
0.1208	16.0	3600	0.9856	0.795
0.0877	17.0	3825	1.0442	0.7767
0.1165	18.0	4050	0.9465	0.8117
0.1328	19.0	4275	0.8299	0.81
0.0427	20.0	4500	1.1880	0.7917
0.0826	21.0	4725	1.0665	0.8083
0.0679	22.0	4950	1.2201	0.7917
0.1018	23.0	5175	1.1824	0.8
0.0255	24.0	5400	1.2359	0.8117
0.0956	25.0	5625	1.2156	0.805
0.0725	26.0	5850	1.3671	0.81
0.0849	27.0	6075	1.3399	0.7917
0.068	28.0	6300	1.3279	0.8117
0.0512	29.0	6525	1.1460	0.82
0.0439	30.0	6750	1.4730	0.8017
0.0414	31.0	6975	1.2224	0.8067
0.0174	32.0	7200	1.6967	0.7983
0.0407	33.0	7425	1.5401	0.7983
0.0316	34.0	7650	1.2844	0.8017
0.0008	35.0	7875	1.7477	0.805
0.0104	36.0	8100	1.5173	0.8167
0.0005	37.0	8325	1.6340	0.7967
0.0286	38.0	8550	1.4323	0.7983
0.0292	39.0	8775	1.4953	0.805
0.0108	40.0	9000	1.6930	0.8183
0.022	41.0	9225	1.7083	0.8033
0.0101	42.0	9450	1.8030	0.8083
0.0122	43.0	9675	1.8925	0.8133
0.0071	44.0	9900	1.7250	0.815
0.0004	45.0	10125	1.7937	0.8017
0.0008	46.0	10350	1.9056	0.8067
0.0003	47.0	10575	1.8311	0.8083
0.0001	48.0	10800	1.9401	0.8033
0.0001	49.0	11025	1.8499	0.8083
0.0	50.0	11250	1.8372	0.81

Framework versions

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