smids_3x_beit_base_rms_0001_fold5

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: 1.9458
• Accuracy: 0.8133

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
0.8091	1.0	225	0.7888	0.5967
0.6825	2.0	450	0.6663	0.7
0.6331	3.0	675	0.6179	0.7233
0.545	4.0	900	0.5461	0.7667
0.4511	5.0	1125	0.5082	0.785
0.5168	6.0	1350	0.4960	0.7867
0.498	7.0	1575	0.4691	0.8133
0.4445	8.0	1800	0.5042	0.7883
0.3476	9.0	2025	0.5424	0.7883
0.3599	10.0	2250	0.4586	0.8183
0.2707	11.0	2475	0.5539	0.8133
0.3169	12.0	2700	0.5413	0.805
0.2566	13.0	2925	0.6318	0.7933
0.1731	14.0	3150	0.6551	0.805
0.1033	15.0	3375	0.7315	0.7967
0.1418	16.0	3600	0.6489	0.8217
0.1165	17.0	3825	0.8690	0.815
0.1219	18.0	4050	0.8038	0.7883
0.1084	19.0	4275	0.9593	0.795
0.0893	20.0	4500	0.9469	0.8
0.0691	21.0	4725	1.0155	0.8183
0.155	22.0	4950	1.0808	0.8033
0.068	23.0	5175	1.2932	0.8067
0.0871	24.0	5400	1.0549	0.7933
0.0614	25.0	5625	1.2073	0.8183
0.0426	26.0	5850	1.1147	0.8167
0.0501	27.0	6075	1.1794	0.8067
0.0496	28.0	6300	1.2384	0.8033
0.0951	29.0	6525	1.1319	0.795
0.0433	30.0	6750	1.1451	0.8217
0.0293	31.0	6975	1.3635	0.815
0.0095	32.0	7200	1.4313	0.8067
0.0383	33.0	7425	1.2822	0.8217
0.0173	34.0	7650	1.4012	0.8217
0.0213	35.0	7875	1.5178	0.8117
0.0022	36.0	8100	1.6408	0.8283
0.0311	37.0	8325	1.7017	0.8117
0.0285	38.0	8550	1.5903	0.8183
0.0027	39.0	8775	1.6460	0.815
0.0036	40.0	9000	1.6806	0.7983
0.0136	41.0	9225	1.5975	0.8067
0.0063	42.0	9450	1.8013	0.8217
0.0155	43.0	9675	1.6944	0.8083
0.0003	44.0	9900	1.5360	0.8283
0.0001	45.0	10125	1.7589	0.8283
0.0203	46.0	10350	1.7247	0.8267
0.0004	47.0	10575	1.7343	0.8217
0.0069	48.0	10800	1.8685	0.8183
0.0001	49.0	11025	1.9172	0.8133
0.0	50.0	11250	1.9458	0.8133

Framework versions

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