smids_5x_deit_tiny_rms_001_fold5

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: 0.6094
• Accuracy: 0.7433

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
0.8822	1.0	375	0.8719	0.52
0.8936	2.0	750	0.8470	0.535
0.8252	3.0	1125	0.8071	0.595
0.8333	4.0	1500	0.7970	0.6017
0.8046	5.0	1875	0.8070	0.5633
0.8082	6.0	2250	0.9208	0.5167
0.7481	7.0	2625	0.7984	0.5633
0.8409	8.0	3000	0.7900	0.5783
0.7673	9.0	3375	0.7551	0.62
0.7321	10.0	3750	0.7485	0.6133
0.8282	11.0	4125	0.7517	0.6083
0.7206	12.0	4500	0.7745	0.6
0.6841	13.0	4875	0.8307	0.5917
0.7738	14.0	5250	0.7274	0.6683
0.8416	15.0	5625	0.7353	0.67
0.704	16.0	6000	0.7258	0.65
0.6873	17.0	6375	0.7174	0.68
0.714	18.0	6750	0.7557	0.6483
0.7105	19.0	7125	0.6868	0.6917
0.6559	20.0	7500	0.6845	0.6783
0.6717	21.0	7875	0.7043	0.67
0.7139	22.0	8250	0.6944	0.68
0.6633	23.0	8625	0.7071	0.6667
0.6888	24.0	9000	0.6979	0.6883
0.6621	25.0	9375	0.6468	0.7117
0.6157	26.0	9750	0.6767	0.6833
0.6777	27.0	10125	0.7097	0.67
0.7108	28.0	10500	0.6811	0.6917
0.8139	29.0	10875	0.6750	0.7067
0.6291	30.0	11250	0.6415	0.7133
0.5725	31.0	11625	0.6769	0.6833
0.6243	32.0	12000	0.6733	0.7267
0.6311	33.0	12375	0.6227	0.7217
0.6254	34.0	12750	0.6222	0.72
0.567	35.0	13125	0.6040	0.735
0.5363	36.0	13500	0.5935	0.7533
0.6308	37.0	13875	0.6047	0.7267
0.5334	38.0	14250	0.6481	0.7217
0.5951	39.0	14625	0.6059	0.7317
0.6325	40.0	15000	0.6172	0.735
0.5905	41.0	15375	0.6255	0.7233
0.6095	42.0	15750	0.5896	0.7433
0.49	43.0	16125	0.5925	0.7367
0.4891	44.0	16500	0.5937	0.7367
0.4867	45.0	16875	0.5918	0.7583
0.5178	46.0	17250	0.6030	0.735
0.561	47.0	17625	0.6183	0.74
0.4632	48.0	18000	0.5943	0.7517
0.4666	49.0	18375	0.6107	0.7417
0.4901	50.0	18750	0.6094	0.7433

Framework versions

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