smids_1x_deit_tiny_rms_00001_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: 0.8725
• Accuracy: 0.8852

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
0.4335	1.0	75	0.3629	0.8519
0.2445	2.0	150	0.3163	0.8752
0.1892	3.0	225	0.3276	0.8719
0.1722	4.0	300	0.3320	0.8769
0.1627	5.0	375	0.4052	0.8336
0.0774	6.0	450	0.4177	0.8636
0.0654	7.0	525	0.4298	0.8835
0.0585	8.0	600	0.5091	0.8586
0.0287	9.0	675	0.5515	0.8785
0.0409	10.0	750	0.6043	0.8735
0.0028	11.0	825	0.6415	0.8852
0.015	12.0	900	0.7011	0.8819
0.0186	13.0	975	0.7682	0.8669
0.0337	14.0	1050	0.7658	0.8669
0.0071	15.0	1125	0.7841	0.8636
0.0427	16.0	1200	0.8228	0.8702
0.0064	17.0	1275	0.8036	0.8752
0.0151	18.0	1350	0.8626	0.8719
0.0002	19.0	1425	0.8177	0.8819
0.0112	20.0	1500	0.8351	0.8636
0.0001	21.0	1575	0.8049	0.8869
0.0032	22.0	1650	0.8268	0.8735
0.0001	23.0	1725	0.8681	0.8752
0.0001	24.0	1800	0.8420	0.8686
0.0	25.0	1875	0.8252	0.8686
0.0001	26.0	1950	0.8504	0.8719
0.0079	27.0	2025	0.8787	0.8769
0.0047	28.0	2100	0.8232	0.8769
0.0019	29.0	2175	0.8813	0.8752
0.0048	30.0	2250	0.8417	0.8652
0.0115	31.0	2325	0.8526	0.8702
0.0	32.0	2400	0.8276	0.8835
0.0001	33.0	2475	0.8214	0.8835
0.0001	34.0	2550	0.8377	0.8802
0.0182	35.0	2625	0.8794	0.8735
0.0	36.0	2700	0.8604	0.8802
0.0001	37.0	2775	0.8420	0.8802
0.0	38.0	2850	0.9013	0.8802
0.0008	39.0	2925	0.8537	0.8819
0.0	40.0	3000	0.8672	0.8819
0.0029	41.0	3075	0.8548	0.8852
0.0	42.0	3150	0.8641	0.8835
0.002	43.0	3225	0.8645	0.8835
0.0052	44.0	3300	0.8649	0.8835
0.0	45.0	3375	0.8672	0.8835
0.0	46.0	3450	0.8698	0.8852
0.0018	47.0	3525	0.8729	0.8852
0.0	48.0	3600	0.8731	0.8852
0.0032	49.0	3675	0.8719	0.8852
0.0011	50.0	3750	0.8725	0.8852

Framework versions

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