smids_3x_deit_base_rms_001_fold2

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: 0.6036
• Accuracy: 0.8020

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
1.1186	1.0	225	0.8982	0.5075
0.9141	2.0	450	0.8195	0.5557
0.8927	3.0	675	0.8189	0.5441
0.8323	4.0	900	0.8095	0.5541
0.8947	5.0	1125	0.7623	0.5757
0.7287	6.0	1350	0.8273	0.5591
0.7585	7.0	1575	0.7770	0.5973
0.8103	8.0	1800	0.7290	0.6106
0.7335	9.0	2025	0.7908	0.5807
0.7359	10.0	2250	0.7312	0.5874
0.8194	11.0	2475	0.9398	0.5557
0.7512	12.0	2700	0.7107	0.5923
0.7169	13.0	2925	0.7015	0.6639
0.6759	14.0	3150	0.6767	0.6672
0.7072	15.0	3375	0.6493	0.6955
0.6502	16.0	3600	0.6076	0.7404
0.6691	17.0	3825	0.6396	0.6855
0.6248	18.0	4050	0.5525	0.7621
0.5977	19.0	4275	0.7766	0.6373
0.582	20.0	4500	0.5758	0.7438
0.5383	21.0	4725	0.5521	0.7554
0.6208	22.0	4950	0.5508	0.7521
0.6018	23.0	5175	0.5519	0.7604
0.5417	24.0	5400	0.5813	0.7471
0.6149	25.0	5625	0.5077	0.7820
0.5061	26.0	5850	0.5197	0.7804
0.5327	27.0	6075	0.5610	0.7454
0.487	28.0	6300	0.5448	0.7654
0.5248	29.0	6525	0.5394	0.7704
0.4978	30.0	6750	0.5209	0.7804
0.523	31.0	6975	0.5417	0.7604
0.502	32.0	7200	0.5080	0.7770
0.4674	33.0	7425	0.5071	0.7820
0.4329	34.0	7650	0.4947	0.8003
0.4583	35.0	7875	0.5207	0.7854
0.4868	36.0	8100	0.4819	0.8087
0.4542	37.0	8325	0.4836	0.7987
0.4328	38.0	8550	0.5050	0.7937
0.3395	39.0	8775	0.5073	0.7953
0.339	40.0	9000	0.5849	0.7870
0.3908	41.0	9225	0.5523	0.7820
0.4049	42.0	9450	0.5288	0.7920
0.3295	43.0	9675	0.5405	0.8053
0.3742	44.0	9900	0.5541	0.8020
0.3832	45.0	10125	0.5567	0.7953
0.3742	46.0	10350	0.5578	0.7920
0.3317	47.0	10575	0.5698	0.8103
0.2873	48.0	10800	0.5859	0.8037
0.3255	49.0	11025	0.6015	0.7970
0.3175	50.0	11250	0.6036	0.8020

Framework versions

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