smids_10x_deit_small_sgd_00001_fold2

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: 0.9349
• Accuracy: 0.5591

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: 1e-05
• 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
1.0693	1.0	750	1.0692	0.4293
1.0584	2.0	1500	1.0648	0.4343
1.0342	3.0	2250	1.0600	0.4376
1.0374	4.0	3000	1.0551	0.4443
1.028	5.0	3750	1.0500	0.4459
1.0131	6.0	4500	1.0451	0.4476
1.022	7.0	5250	1.0402	0.4459
1.0192	8.0	6000	1.0354	0.4526
1.0168	9.0	6750	1.0306	0.4576
0.9985	10.0	7500	1.0259	0.4592
0.9898	11.0	8250	1.0213	0.4609
1.0116	12.0	9000	1.0168	0.4642
0.9986	13.0	9750	1.0125	0.4659
0.9818	14.0	10500	1.0083	0.4759
0.9837	15.0	11250	1.0041	0.4809
0.9601	16.0	12000	1.0001	0.4809
0.9572	17.0	12750	0.9961	0.4809
0.9406	18.0	13500	0.9923	0.4859
0.9621	19.0	14250	0.9887	0.4892
0.9467	20.0	15000	0.9850	0.4925
0.9691	21.0	15750	0.9816	0.4992
0.9406	22.0	16500	0.9782	0.5008
0.9223	23.0	17250	0.9750	0.5058
0.9127	24.0	18000	0.9718	0.5075
0.9371	25.0	18750	0.9688	0.5141
0.9589	26.0	19500	0.9659	0.5175
0.9189	27.0	20250	0.9631	0.5208
0.9249	28.0	21000	0.9605	0.5258
0.927	29.0	21750	0.9580	0.5275
0.9378	30.0	22500	0.9556	0.5308
0.8829	31.0	23250	0.9533	0.5308
0.931	32.0	24000	0.9512	0.5341
0.9197	33.0	24750	0.9492	0.5374
0.9032	34.0	25500	0.9474	0.5374
0.9	35.0	26250	0.9457	0.5391
0.8939	36.0	27000	0.9442	0.5441
0.9276	37.0	27750	0.9427	0.5458
0.8712	38.0	28500	0.9414	0.5458
0.9222	39.0	29250	0.9402	0.5458
0.8913	40.0	30000	0.9392	0.5474
0.8879	41.0	30750	0.9383	0.5474
0.8851	42.0	31500	0.9375	0.5541
0.8777	43.0	32250	0.9368	0.5541
0.8945	44.0	33000	0.9362	0.5541
0.8708	45.0	33750	0.9358	0.5574
0.9082	46.0	34500	0.9354	0.5591
0.9028	47.0	35250	0.9352	0.5591
0.8903	48.0	36000	0.9350	0.5591
0.8994	49.0	36750	0.9349	0.5591
0.9183	50.0	37500	0.9349	0.5591

Framework versions

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