smids_10x_deit_small_adamax_001_fold5

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

• Loss: 0.9313
• Accuracy: 0.905

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
0.3135	1.0	750	0.2998	0.8733
0.2768	2.0	1500	0.4449	0.8417
0.2302	3.0	2250	0.3152	0.885
0.1705	4.0	3000	0.4052	0.8717
0.1542	5.0	3750	0.3463	0.9017
0.1492	6.0	4500	0.3620	0.8967
0.0963	7.0	5250	0.3599	0.895
0.0922	8.0	6000	0.4809	0.8883
0.1185	9.0	6750	0.5269	0.9017
0.0382	10.0	7500	0.5699	0.8983
0.0415	11.0	8250	0.4670	0.9033
0.0358	12.0	9000	0.4714	0.9083
0.0301	13.0	9750	0.4780	0.915
0.0178	14.0	10500	0.5327	0.9033
0.0172	15.0	11250	0.6375	0.8983
0.0148	16.0	12000	0.5566	0.8967
0.0235	17.0	12750	0.5739	0.9067
0.0298	18.0	13500	0.7210	0.8983
0.0012	19.0	14250	0.7611	0.8883
0.0391	20.0	15000	0.8089	0.8917
0.0002	21.0	15750	0.6460	0.8983
0.0095	22.0	16500	0.6954	0.9067
0.0251	23.0	17250	0.6718	0.9017
0.0021	24.0	18000	0.6374	0.9067
0.0001	25.0	18750	0.6533	0.905
0.0001	26.0	19500	0.7022	0.91
0.003	27.0	20250	0.8113	0.9
0.0277	28.0	21000	0.7402	0.8983
0.0056	29.0	21750	0.7949	0.8967
0.007	30.0	22500	0.8055	0.8967
0.0001	31.0	23250	0.8426	0.9083
0.0	32.0	24000	0.8618	0.905
0.0	33.0	24750	0.8392	0.9083
0.0	34.0	25500	0.8019	0.9067
0.0	35.0	26250	0.8163	0.9067
0.0	36.0	27000	0.8994	0.895
0.0037	37.0	27750	0.8599	0.9067
0.0	38.0	28500	0.8721	0.905
0.0	39.0	29250	0.8612	0.9067
0.0	40.0	30000	0.9150	0.9033
0.0	41.0	30750	0.8804	0.91
0.0	42.0	31500	0.8814	0.9067
0.0	43.0	32250	0.8966	0.9083
0.0	44.0	33000	0.9028	0.9083
0.0	45.0	33750	0.9087	0.9083
0.0	46.0	34500	0.9131	0.9083
0.0	47.0	35250	0.9195	0.9067
0.0	48.0	36000	0.9269	0.9067
0.0	49.0	36750	0.9312	0.9067
0.0	50.0	37500	0.9313	0.905

Framework versions

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