smids_10x_beit_large_adamax_00001_fold5

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: 0.8705
• Accuracy: 0.9183

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: 1e-05
• 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
0.151	1.0	750	0.2341	0.9117
0.085	2.0	1500	0.2729	0.9117
0.0389	3.0	2250	0.3555	0.9183
0.0354	4.0	3000	0.4728	0.92
0.0161	5.0	3750	0.5494	0.9117
0.0006	6.0	4500	0.5920	0.9167
0.0191	7.0	5250	0.7177	0.9083
0.0025	8.0	6000	0.7193	0.9183
0.0296	9.0	6750	0.7219	0.9183
0.0071	10.0	7500	0.7346	0.9067
0.0001	11.0	8250	0.8516	0.9133
0.0012	12.0	9000	0.7790	0.9217
0.0009	13.0	9750	0.7769	0.9117
0.0	14.0	10500	0.8050	0.92
0.0	15.0	11250	0.7869	0.9167
0.0001	16.0	12000	0.8102	0.9133
0.0588	17.0	12750	0.7913	0.9183
0.0	18.0	13500	0.9080	0.9117
0.0	19.0	14250	0.7883	0.915
0.0	20.0	15000	0.8588	0.9183
0.0001	21.0	15750	0.8772	0.9167
0.0001	22.0	16500	0.8747	0.9133
0.0001	23.0	17250	0.7911	0.9217
0.0	24.0	18000	0.7828	0.9217
0.0	25.0	18750	0.7802	0.9233
0.0	26.0	19500	0.8237	0.92
0.0	27.0	20250	0.8003	0.9217
0.0	28.0	21000	0.8936	0.9133
0.0009	29.0	21750	0.8831	0.915
0.0181	30.0	22500	0.8036	0.9217
0.0	31.0	23250	0.7557	0.9267
0.0	32.0	24000	0.8859	0.92
0.0	33.0	24750	0.8754	0.92
0.0001	34.0	25500	0.8554	0.9117
0.0	35.0	26250	0.8615	0.9167
0.0	36.0	27000	0.8299	0.9217
0.0035	37.0	27750	0.8816	0.9167
0.0	38.0	28500	0.8681	0.9233
0.0	39.0	29250	0.8281	0.92
0.0	40.0	30000	0.8247	0.9183
0.0008	41.0	30750	0.8595	0.9183
0.0	42.0	31500	0.8563	0.92
0.0038	43.0	32250	0.8322	0.925
0.0	44.0	33000	0.8334	0.9183
0.0	45.0	33750	0.8475	0.9183
0.0	46.0	34500	0.8657	0.92
0.0	47.0	35250	0.8614	0.9183
0.0	48.0	36000	0.8662	0.92
0.0	49.0	36750	0.8708	0.9183
0.0	50.0	37500	0.8705	0.9183

Framework versions

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