hushem_40x_beit_large_adamax_001_fold2

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

• Loss: 4.4537
• Accuracy: 0.6889

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
0.3184	1.0	215	1.1233	0.7333
0.1436	2.0	430	1.6198	0.6889
0.047	3.0	645	1.8592	0.6667
0.0527	4.0	860	1.5842	0.7111
0.0463	5.0	1075	2.7619	0.6889
0.0137	6.0	1290	1.4680	0.7556
0.0665	7.0	1505	2.2491	0.6889
0.0121	8.0	1720	1.7706	0.7556
0.0005	9.0	1935	2.1035	0.7556
0.0192	10.0	2150	3.0002	0.6889
0.02	11.0	2365	2.1406	0.6667
0.011	12.0	2580	2.2828	0.6667
0.0346	13.0	2795	2.5178	0.6667
0.0045	14.0	3010	2.0578	0.7333
0.0021	15.0	3225	1.4918	0.7556
0.0002	16.0	3440	2.6023	0.7111
0.0007	17.0	3655	2.4242	0.7111
0.0019	18.0	3870	2.8391	0.6667
0.0005	19.0	4085	2.9921	0.7556
0.0	20.0	4300	3.1529	0.6667
0.0	21.0	4515	2.7412	0.7556
0.0	22.0	4730	2.8583	0.7333
0.0	23.0	4945	2.9971	0.7333
0.0	24.0	5160	3.0142	0.7556
0.0	25.0	5375	3.0328	0.7556
0.0	26.0	5590	3.0307	0.7778
0.0	27.0	5805	3.2285	0.7556
0.0	28.0	6020	3.2719	0.7111
0.0	29.0	6235	2.7270	0.7778
0.0	30.0	6450	3.4979	0.7111
0.0	31.0	6665	3.4752	0.7333
0.0	32.0	6880	3.4952	0.7333
0.0	33.0	7095	3.5111	0.7333
0.0	34.0	7310	3.5230	0.7333
0.0	35.0	7525	3.5422	0.7333
0.0	36.0	7740	3.5606	0.7333
0.0	37.0	7955	3.5754	0.7333
0.0	38.0	8170	3.5859	0.7333
0.0	39.0	8385	3.5773	0.7333
0.0	40.0	8600	4.7039	0.6
0.0	41.0	8815	4.7831	0.6
0.0	42.0	9030	4.4812	0.6667
0.0	43.0	9245	4.4224	0.6889
0.0	44.0	9460	4.4294	0.6889
0.0	45.0	9675	4.4285	0.6889
0.0	46.0	9890	4.4304	0.6889
0.0	47.0	10105	4.4476	0.6889
0.0	48.0	10320	4.4513	0.6889
0.0	49.0	10535	4.4531	0.6889
0.0	50.0	10750	4.4537	0.6889

Framework versions

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