smids_1x_deit_small_adamax_00001_fold4

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: 0.9148
• Accuracy: 0.86

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
0.5803	1.0	75	0.4968	0.8167
0.4066	2.0	150	0.3965	0.8517
0.3494	3.0	225	0.3679	0.8583
0.257	4.0	300	0.3627	0.8583
0.1968	5.0	375	0.3612	0.8567
0.1309	6.0	450	0.3609	0.865
0.1744	7.0	525	0.3526	0.8667
0.1066	8.0	600	0.3650	0.8733
0.0701	9.0	675	0.3803	0.87
0.058	10.0	750	0.3887	0.8683
0.0585	11.0	825	0.4227	0.8667
0.0507	12.0	900	0.4565	0.8667
0.0443	13.0	975	0.4751	0.8667
0.023	14.0	1050	0.5029	0.875
0.0067	15.0	1125	0.5522	0.8667
0.0046	16.0	1200	0.5758	0.8683
0.0072	17.0	1275	0.6012	0.8667
0.0186	18.0	1350	0.6185	0.8667
0.0049	19.0	1425	0.6452	0.8633
0.0012	20.0	1500	0.6704	0.8633
0.0009	21.0	1575	0.6922	0.8633
0.0009	22.0	1650	0.7205	0.8617
0.0007	23.0	1725	0.7357	0.8617
0.0194	24.0	1800	0.7622	0.86
0.0038	25.0	1875	0.7720	0.8583
0.0005	26.0	1950	0.7827	0.86
0.0003	27.0	2025	0.7974	0.8583
0.0054	28.0	2100	0.8004	0.8583
0.019	29.0	2175	0.8026	0.8633
0.0003	30.0	2250	0.8285	0.86
0.0002	31.0	2325	0.8245	0.8617
0.0002	32.0	2400	0.8349	0.86
0.0002	33.0	2475	0.8577	0.8617
0.0002	34.0	2550	0.8568	0.86
0.0002	35.0	2625	0.8651	0.8583
0.0002	36.0	2700	0.8693	0.86
0.0161	37.0	2775	0.8692	0.8633
0.0002	38.0	2850	0.8782	0.8583
0.0002	39.0	2925	0.8858	0.86
0.0001	40.0	3000	0.8886	0.8583
0.0154	41.0	3075	0.8970	0.86
0.0001	42.0	3150	0.8973	0.86
0.0001	43.0	3225	0.9034	0.86
0.0001	44.0	3300	0.9094	0.8617
0.0001	45.0	3375	0.9094	0.86
0.0001	46.0	3450	0.9101	0.86
0.0001	47.0	3525	0.9123	0.86
0.0001	48.0	3600	0.9135	0.86
0.0001	49.0	3675	0.9142	0.86
0.0001	50.0	3750	0.9148	0.86

Framework versions

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