smids_3x_deit_base_rms_001_fold3

This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.5835
• Accuracy: 0.76

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: 0.001
• 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
0.9577	1.0	225	0.9911	0.4833
0.8945	2.0	450	0.8723	0.515
0.9126	3.0	675	0.9098	0.5583
0.864	4.0	900	0.9590	0.4883
0.7818	5.0	1125	0.8052	0.61
0.8173	6.0	1350	0.8349	0.57
0.8234	7.0	1575	0.8296	0.5633
0.8025	8.0	1800	0.7926	0.6433
0.794	9.0	2025	0.7671	0.6367
0.7364	10.0	2250	0.7623	0.6917
0.7539	11.0	2475	0.7462	0.6567
0.7493	12.0	2700	0.7599	0.6483
0.8066	13.0	2925	0.7984	0.6233
0.7644	14.0	3150	0.7284	0.6767
0.6406	15.0	3375	0.8986	0.6117
0.7539	16.0	3600	0.7380	0.6417
0.7079	17.0	3825	0.7519	0.6483
0.7112	18.0	4050	0.7274	0.6683
0.709	19.0	4275	0.7182	0.6783
0.6627	20.0	4500	0.6933	0.6917
0.62	21.0	4725	0.7192	0.6783
0.6351	22.0	4950	0.6854	0.6967
0.6169	23.0	5175	0.6958	0.6917
0.6173	24.0	5400	0.6916	0.6867
0.6807	25.0	5625	0.6783	0.705
0.6099	26.0	5850	0.6681	0.705
0.5604	27.0	6075	0.7149	0.6767
0.6004	28.0	6300	0.7253	0.6667
0.6392	29.0	6525	0.6891	0.66
0.5659	30.0	6750	0.6273	0.7267
0.5546	31.0	6975	0.6350	0.7317
0.5835	32.0	7200	0.6529	0.6983
0.6237	33.0	7425	0.6048	0.7233
0.5674	34.0	7650	0.6396	0.7167
0.5405	35.0	7875	0.6074	0.7183
0.5745	36.0	8100	0.5947	0.7317
0.5811	37.0	8325	0.5820	0.7383
0.5642	38.0	8550	0.5685	0.7433
0.5332	39.0	8775	0.5891	0.745
0.5278	40.0	9000	0.5919	0.7283
0.5007	41.0	9225	0.5742	0.7567
0.5377	42.0	9450	0.5885	0.76
0.4913	43.0	9675	0.5649	0.755
0.5315	44.0	9900	0.5703	0.74
0.4857	45.0	10125	0.5619	0.765
0.4747	46.0	10350	0.5832	0.7533
0.5553	47.0	10575	0.5734	0.755
0.452	48.0	10800	0.5866	0.7617
0.4761	49.0	11025	0.5792	0.7567
0.4755	50.0	11250	0.5835	0.76

Framework versions

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