smids_3x_beit_base_adamax_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: 1.2272
• Accuracy: 0.8733

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
0.3102	1.0	225	0.3679	0.8633
0.1053	2.0	450	0.3867	0.87
0.0809	3.0	675	0.4978	0.8617
0.1418	4.0	900	0.5585	0.8717
0.0152	5.0	1125	0.6419	0.885
0.0232	6.0	1350	0.6902	0.8717
0.0119	7.0	1575	0.8503	0.8633
0.0116	8.0	1800	0.8413	0.8667
0.0484	9.0	2025	0.9018	0.8683
0.0101	10.0	2250	0.9930	0.855
0.0039	11.0	2475	1.0769	0.8733
0.0004	12.0	2700	1.0602	0.8717
0.0292	13.0	2925	1.1584	0.875
0.0029	14.0	3150	1.2271	0.8583
0.01	15.0	3375	1.1632	0.8733
0.0001	16.0	3600	1.1832	0.8633
0.0	17.0	3825	1.2281	0.86
0.004	18.0	4050	1.0844	0.8783
0.0003	19.0	4275	1.2463	0.8683
0.0112	20.0	4500	1.2122	0.8733
0.0013	21.0	4725	1.2444	0.8617
0.0002	22.0	4950	1.2159	0.86
0.0002	23.0	5175	1.2215	0.8667
0.0	24.0	5400	1.2014	0.8733
0.0007	25.0	5625	1.1844	0.875
0.0	26.0	5850	1.3054	0.8683
0.0	27.0	6075	1.3588	0.8583
0.0332	28.0	6300	1.2029	0.875
0.0	29.0	6525	1.2414	0.87
0.0001	30.0	6750	1.2400	0.8783
0.0005	31.0	6975	1.1861	0.87
0.0	32.0	7200	1.1528	0.8767
0.0	33.0	7425	1.2071	0.8783
0.0002	34.0	7650	1.2652	0.875
0.0	35.0	7875	1.2647	0.8783
0.0	36.0	8100	1.3389	0.865
0.0	37.0	8325	1.3158	0.8683
0.0	38.0	8550	1.2845	0.8717
0.0	39.0	8775	1.2211	0.8783
0.0383	40.0	9000	1.3005	0.865
0.0001	41.0	9225	1.3129	0.8567
0.0025	42.0	9450	1.2924	0.865
0.0	43.0	9675	1.2393	0.8667
0.0021	44.0	9900	1.2861	0.87
0.0	45.0	10125	1.2626	0.8717
0.0	46.0	10350	1.2383	0.8733
0.0	47.0	10575	1.2652	0.8717
0.0	48.0	10800	1.2466	0.8733
0.0	49.0	11025	1.2259	0.875
0.0	50.0	11250	1.2272	0.8733

Framework versions

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