smids_10x_deit_tiny_adamax_001_fold2

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

• Loss: 1.0691
• Accuracy: 0.9002

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
0.3362	1.0	750	0.3519	0.8652
0.2971	2.0	1500	0.3131	0.8918
0.1771	3.0	2250	0.2717	0.8885
0.2985	4.0	3000	0.3652	0.8652
0.1399	5.0	3750	0.3216	0.9018
0.1317	6.0	4500	0.3948	0.8802
0.1309	7.0	5250	0.3860	0.8902
0.1165	8.0	6000	0.4557	0.8852
0.0308	9.0	6750	0.5032	0.8686
0.0315	10.0	7500	0.4981	0.8769
0.0974	11.0	8250	0.6363	0.8769
0.1017	12.0	9000	0.5021	0.8869
0.0475	13.0	9750	0.5896	0.8885
0.0086	14.0	10500	0.6931	0.8918
0.0301	15.0	11250	0.6531	0.8902
0.0049	16.0	12000	0.7157	0.8819
0.0307	17.0	12750	0.7054	0.8935
0.0113	18.0	13500	0.7646	0.8869
0.0492	19.0	14250	0.7424	0.8885
0.0093	20.0	15000	0.6366	0.8952
0.011	21.0	15750	0.8426	0.8885
0.0191	22.0	16500	0.7557	0.8952
0.0047	23.0	17250	0.7578	0.8885
0.0163	24.0	18000	0.8275	0.8902
0.0001	25.0	18750	0.8176	0.8935
0.0023	26.0	19500	0.8054	0.8968
0.0181	27.0	20250	0.8270	0.8952
0.0	28.0	21000	0.8173	0.9035
0.0001	29.0	21750	0.8348	0.9018
0.0	30.0	22500	0.8105	0.9101
0.0	31.0	23250	0.7837	0.9118
0.0	32.0	24000	0.9929	0.8935
0.0	33.0	24750	0.8103	0.9085
0.0	34.0	25500	0.8769	0.9035
0.0	35.0	26250	0.8987	0.8985
0.0	36.0	27000	1.0129	0.9002
0.0053	37.0	27750	0.9506	0.9068
0.0	38.0	28500	1.0495	0.8935
0.0	39.0	29250	0.9869	0.9018
0.0	40.0	30000	1.0087	0.8968
0.0	41.0	30750	1.0348	0.8985
0.0	42.0	31500	1.0299	0.8985
0.0	43.0	32250	1.0437	0.8968
0.0	44.0	33000	1.0468	0.8985
0.0028	45.0	33750	1.0539	0.9002
0.0	46.0	34500	1.0588	0.9002
0.0	47.0	35250	1.0567	0.9002
0.0	48.0	36000	1.0631	0.9002
0.0	49.0	36750	1.0673	0.9002
0.0	50.0	37500	1.0691	0.9002

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2