smids_3x_deit_small_rms_0001_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: 1.3428
• Accuracy: 0.875

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: 0.0001
• 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
0.3711	1.0	225	0.3570	0.87
0.1764	2.0	450	0.4501	0.86
0.1125	3.0	675	0.7262	0.8017
0.1192	4.0	900	0.4185	0.8867
0.0724	5.0	1125	0.7207	0.8333
0.0693	6.0	1350	0.5544	0.88
0.012	7.0	1575	0.7617	0.865
0.1104	8.0	1800	0.8266	0.8633
0.0494	9.0	2025	0.7952	0.8633
0.043	10.0	2250	1.1043	0.8333
0.0404	11.0	2475	0.7760	0.8583
0.047	12.0	2700	0.9765	0.8483
0.0545	13.0	2925	0.8078	0.885
0.0635	14.0	3150	0.7741	0.8717
0.0334	15.0	3375	0.9921	0.86
0.0026	16.0	3600	0.9584	0.8717
0.0226	17.0	3825	1.0326	0.8633
0.0021	18.0	4050	1.0274	0.8567
0.0114	19.0	4275	1.0650	0.8467
0.0091	20.0	4500	1.1426	0.835
0.0087	21.0	4725	1.0992	0.8533
0.0141	22.0	4950	1.0907	0.8517
0.0005	23.0	5175	1.0030	0.85
0.0006	24.0	5400	1.1052	0.8483
0.0255	25.0	5625	1.0962	0.8633
0.0002	26.0	5850	1.0258	0.8817
0.0001	27.0	6075	0.8746	0.8883
0.0	28.0	6300	0.9827	0.88
0.0	29.0	6525	0.9372	0.8883
0.0004	30.0	6750	0.9400	0.875
0.0	31.0	6975	0.9451	0.8933
0.0	32.0	7200	1.0223	0.88
0.0007	33.0	7425	1.1451	0.88
0.0	34.0	7650	1.0754	0.875
0.0	35.0	7875	1.1610	0.8683
0.0	36.0	8100	1.1706	0.8717
0.0	37.0	8325	1.0772	0.8733
0.0	38.0	8550	1.1612	0.87
0.0001	39.0	8775	1.1977	0.8583
0.0	40.0	9000	1.2371	0.8667
0.0	41.0	9225	1.2466	0.8717
0.0034	42.0	9450	1.2736	0.8733
0.0	43.0	9675	1.2213	0.88
0.0027	44.0	9900	1.3864	0.8617
0.0	45.0	10125	1.3338	0.8717
0.0	46.0	10350	1.3311	0.8683
0.0	47.0	10575	1.3334	0.8717
0.0	48.0	10800	1.3390	0.8733
0.0	49.0	11025	1.3422	0.875
0.0	50.0	11250	1.3428	0.875

Framework versions

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