smids_10x_deit_tiny_adamax_0001_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: 1.0252
• Accuracy: 0.8885

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: 0.0001
• 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
0.2345	1.0	750	0.2862	0.8852
0.1226	2.0	1500	0.3492	0.8819
0.0786	3.0	2250	0.4125	0.8785
0.0821	4.0	3000	0.6690	0.8569
0.0058	5.0	3750	0.6467	0.8869
0.0246	6.0	4500	0.7107	0.8819
0.0249	7.0	5250	0.7670	0.8852
0.0023	8.0	6000	0.8535	0.8952
0.0094	9.0	6750	0.9434	0.8819
0.0202	10.0	7500	1.0037	0.8835
0.0073	11.0	8250	1.0395	0.8802
0.0	12.0	9000	0.9154	0.8935
0.0001	13.0	9750	1.0585	0.8785
0.0048	14.0	10500	0.9392	0.8952
0.0002	15.0	11250	0.9865	0.8885
0.0	16.0	12000	1.1010	0.8885
0.0	17.0	12750	1.0500	0.8968
0.0	18.0	13500	1.0366	0.8968
0.0	19.0	14250	0.9974	0.8735
0.0	20.0	15000	1.0266	0.8935
0.0	21.0	15750	0.9740	0.9018
0.0135	22.0	16500	1.0163	0.8935
0.0062	23.0	17250	1.0796	0.8835
0.0	24.0	18000	1.0547	0.8852
0.0	25.0	18750	1.0544	0.8918
0.0	26.0	19500	0.9809	0.8952
0.0089	27.0	20250	1.0367	0.8918
0.0	28.0	21000	1.0326	0.8835
0.0	29.0	21750	1.0069	0.8935
0.0	30.0	22500	1.0290	0.8968
0.0	31.0	23250	1.0034	0.8935
0.0	32.0	24000	0.9398	0.8985
0.0	33.0	24750	1.0178	0.8935
0.0	34.0	25500	1.0300	0.8902
0.0	35.0	26250	1.0140	0.8918
0.0	36.0	27000	1.0115	0.8902
0.0074	37.0	27750	1.0130	0.8918
0.0	38.0	28500	1.0096	0.8902
0.0	39.0	29250	1.0259	0.8935
0.0	40.0	30000	1.0330	0.8918
0.0	41.0	30750	1.0283	0.8902
0.0	42.0	31500	1.0254	0.8902
0.0	43.0	32250	1.0245	0.8869
0.0	44.0	33000	1.0228	0.8885
0.0024	45.0	33750	1.0249	0.8885
0.0	46.0	34500	1.0240	0.8885
0.0	47.0	35250	1.0247	0.8885
0.0	48.0	36000	1.0252	0.8885
0.0	49.0	36750	1.0256	0.8885
0.0	50.0	37500	1.0252	0.8885

Framework versions

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