smids_3x_deit_base_rms_001_fold5

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: 0.6853
• Accuracy: 0.8

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
0.9132	1.0	225	0.8802	0.5617
0.8136	2.0	450	0.8157	0.5617
0.8155	3.0	675	0.8250	0.5317
0.7604	4.0	900	0.7965	0.565
0.7705	5.0	1125	0.7610	0.6317
0.7485	6.0	1350	0.7507	0.6383
0.8596	7.0	1575	0.7348	0.645
0.7401	8.0	1800	0.7401	0.6367
0.7129	9.0	2025	0.7139	0.6667
0.7473	10.0	2250	0.7399	0.6617
0.7702	11.0	2475	0.6996	0.6483
0.707	12.0	2700	0.6871	0.69
0.7396	13.0	2925	0.6860	0.6933
0.7208	14.0	3150	0.6741	0.6817
0.6449	15.0	3375	0.6795	0.6867
0.6507	16.0	3600	0.6654	0.7367
0.6374	17.0	3825	0.6227	0.7183
0.6647	18.0	4050	0.6295	0.7133
0.6463	19.0	4275	0.6181	0.73
0.5869	20.0	4500	0.6227	0.7233
0.5936	21.0	4725	0.5964	0.73
0.6054	22.0	4950	0.6202	0.705
0.6046	23.0	5175	0.5370	0.7667
0.6069	24.0	5400	0.5762	0.75
0.5874	25.0	5625	0.5468	0.7667
0.6566	26.0	5850	0.5470	0.745
0.5832	27.0	6075	0.5273	0.7717
0.5576	28.0	6300	0.5735	0.715
0.5323	29.0	6525	0.5824	0.74
0.531	30.0	6750	0.5710	0.7467
0.4798	31.0	6975	0.5333	0.7667
0.4348	32.0	7200	0.5654	0.765
0.441	33.0	7425	0.5506	0.7683
0.4964	34.0	7650	0.5395	0.7817
0.5026	35.0	7875	0.5510	0.7783
0.4124	36.0	8100	0.5446	0.78
0.5129	37.0	8325	0.5421	0.7883
0.4253	38.0	8550	0.4973	0.795
0.4373	39.0	8775	0.5360	0.79
0.415	40.0	9000	0.5258	0.7983
0.4173	41.0	9225	0.5340	0.7867
0.4858	42.0	9450	0.5549	0.8067
0.3971	43.0	9675	0.5638	0.8117
0.456	44.0	9900	0.5513	0.815
0.3511	45.0	10125	0.5715	0.8067
0.2767	46.0	10350	0.5835	0.7967
0.3597	47.0	10575	0.6128	0.805
0.264	48.0	10800	0.6285	0.815
0.3062	49.0	11025	0.6718	0.8017
0.2147	50.0	11250	0.6853	0.8

Framework versions

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