smids_1x_deit_tiny_adamax_001_fold1

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: 1.1082
• Accuracy: 0.8581

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
0.7889	1.0	76	0.6069	0.7679
0.5102	2.0	152	0.5610	0.7997
0.4217	3.0	228	0.8352	0.6995
0.4159	4.0	304	0.5497	0.7813
0.3589	5.0	380	0.5194	0.8114
0.276	6.0	456	0.5883	0.8097
0.2664	7.0	532	0.6189	0.7846
0.1835	8.0	608	0.6054	0.8280
0.2342	9.0	684	0.5718	0.7997
0.1616	10.0	760	0.4962	0.8347
0.215	11.0	836	0.5625	0.8397
0.0927	12.0	912	0.7589	0.8114
0.1215	13.0	988	0.6624	0.8280
0.1378	14.0	1064	0.6885	0.8364
0.0728	15.0	1140	0.7416	0.8297
0.0869	16.0	1216	0.6972	0.8514
0.036	17.0	1292	0.9407	0.8214
0.0943	18.0	1368	0.8304	0.8347
0.0241	19.0	1444	0.8350	0.8347
0.0865	20.0	1520	0.8686	0.8214
0.0747	21.0	1596	0.9140	0.8230
0.0326	22.0	1672	0.9287	0.8464
0.0264	23.0	1748	1.1143	0.8531
0.0071	24.0	1824	0.9573	0.8464
0.0572	25.0	1900	0.9057	0.8347
0.0013	26.0	1976	0.9807	0.8548
0.0325	27.0	2052	1.0110	0.8447
0.0062	28.0	2128	1.0638	0.8247
0.0158	29.0	2204	0.9283	0.8464
0.018	30.0	2280	0.7858	0.8564
0.0102	31.0	2356	0.9337	0.8614
0.0111	32.0	2432	0.9445	0.8581
0.0002	33.0	2508	0.9913	0.8514
0.0001	34.0	2584	1.0058	0.8564
0.011	35.0	2660	1.0207	0.8581
0.004	36.0	2736	1.0277	0.8564
0.0081	37.0	2812	1.0540	0.8548
0.0	38.0	2888	1.0713	0.8497
0.0038	39.0	2964	1.0596	0.8548
0.0	40.0	3040	1.0612	0.8581
0.0037	41.0	3116	1.0825	0.8564
0.0	42.0	3192	1.0750	0.8614
0.0051	43.0	3268	1.0851	0.8581
0.0	44.0	3344	1.0866	0.8564
0.0	45.0	3420	1.0922	0.8614
0.0	46.0	3496	1.0974	0.8598
0.0	47.0	3572	1.1025	0.8598
0.0028	48.0	3648	1.1080	0.8598
0.0	49.0	3724	1.1080	0.8581
0.0	50.0	3800	1.1082	0.8581

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0