smids_1x_deit_tiny_adamax_00001_fold3

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: 0.6352
• Accuracy: 0.8933

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
0.7543	1.0	75	0.6805	0.7367
0.4751	2.0	150	0.5095	0.8067
0.4299	3.0	225	0.4326	0.83
0.3631	4.0	300	0.4020	0.85
0.2821	5.0	375	0.3807	0.8517
0.249	6.0	450	0.3472	0.8733
0.2583	7.0	525	0.3378	0.8767
0.1796	8.0	600	0.3350	0.88
0.19	9.0	675	0.3308	0.88
0.1195	10.0	750	0.3410	0.8717
0.1538	11.0	825	0.3347	0.8767
0.1042	12.0	900	0.3292	0.895
0.1212	13.0	975	0.3308	0.895
0.0747	14.0	1050	0.3402	0.885
0.0423	15.0	1125	0.3519	0.89
0.0318	16.0	1200	0.3697	0.8867
0.0388	17.0	1275	0.3821	0.8883
0.0223	18.0	1350	0.3957	0.885
0.0206	19.0	1425	0.4157	0.885
0.0272	20.0	1500	0.4298	0.8883
0.007	21.0	1575	0.4227	0.8917
0.0064	22.0	1650	0.4518	0.895
0.0167	23.0	1725	0.4704	0.89
0.0018	24.0	1800	0.4600	0.8867
0.0144	25.0	1875	0.4875	0.89
0.0149	26.0	1950	0.5302	0.8817
0.0149	27.0	2025	0.5247	0.8917
0.001	28.0	2100	0.5348	0.8883
0.0008	29.0	2175	0.5323	0.8883
0.0008	30.0	2250	0.5459	0.89
0.0005	31.0	2325	0.5595	0.8883
0.0008	32.0	2400	0.5625	0.8917
0.0049	33.0	2475	0.5790	0.8867
0.0102	34.0	2550	0.5778	0.89
0.0263	35.0	2625	0.6019	0.89
0.01	36.0	2700	0.5907	0.8883
0.0005	37.0	2775	0.6086	0.8867
0.0003	38.0	2850	0.6091	0.8917
0.0002	39.0	2925	0.6105	0.8883
0.0002	40.0	3000	0.6065	0.8933
0.0002	41.0	3075	0.6175	0.8883
0.0165	42.0	3150	0.6281	0.8917
0.0088	43.0	3225	0.6246	0.8883
0.0003	44.0	3300	0.6288	0.89
0.0015	45.0	3375	0.6290	0.89
0.0021	46.0	3450	0.6320	0.89
0.0189	47.0	3525	0.6360	0.8867
0.0002	48.0	3600	0.6334	0.8933
0.0003	49.0	3675	0.6347	0.8933
0.0086	50.0	3750	0.6352	0.8933

Framework versions

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