smids_3x_deit_base_rms_0001_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: 1.0651
• Accuracy: 0.8983

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
0.337	1.0	225	0.2720	0.8967
0.2132	2.0	450	0.3434	0.885
0.1499	3.0	675	0.3688	0.895
0.0712	4.0	900	0.4644	0.8983
0.0713	5.0	1125	0.3775	0.8967
0.0583	6.0	1350	0.4485	0.8967
0.0178	7.0	1575	0.5692	0.8983
0.0543	8.0	1800	0.5124	0.9017
0.0524	9.0	2025	0.5816	0.8967
0.091	10.0	2250	0.7863	0.8483
0.0396	11.0	2475	0.6045	0.8833
0.0486	12.0	2700	0.6519	0.8867
0.018	13.0	2925	0.6027	0.9
0.011	14.0	3150	0.6223	0.9017
0.0027	15.0	3375	0.6635	0.8883
0.0168	16.0	3600	0.7279	0.8967
0.0421	17.0	3825	0.5369	0.9083
0.0321	18.0	4050	0.7204	0.8833
0.0392	19.0	4275	0.6016	0.89
0.043	20.0	4500	0.5463	0.9033
0.0004	21.0	4725	0.8261	0.8933
0.0001	22.0	4950	0.7660	0.8933
0.0229	23.0	5175	0.6989	0.8967
0.0011	24.0	5400	0.8082	0.8867
0.0036	25.0	5625	0.7432	0.905
0.0	26.0	5850	0.7507	0.9033
0.0005	27.0	6075	0.7412	0.8983
0.0156	28.0	6300	0.7887	0.9
0.0479	29.0	6525	0.6286	0.9117
0.0031	30.0	6750	0.7938	0.8883
0.0034	31.0	6975	0.8118	0.8917
0.0001	32.0	7200	0.7433	0.8917
0.0	33.0	7425	0.7678	0.905
0.0001	34.0	7650	0.8245	0.9
0.0	35.0	7875	0.9668	0.8917
0.0176	36.0	8100	0.7443	0.9017
0.0174	37.0	8325	0.8368	0.8883
0.0	38.0	8550	0.8506	0.8983
0.0	39.0	8775	0.8935	0.9017
0.0	40.0	9000	0.8981	0.9
0.0	41.0	9225	0.9362	0.9
0.0	42.0	9450	0.9833	0.8967
0.0	43.0	9675	0.9431	0.905
0.0	44.0	9900	1.0334	0.8967
0.0	45.0	10125	1.0384	0.8967
0.0	46.0	10350	1.0481	0.8967
0.0024	47.0	10575	1.0461	0.9017
0.0	48.0	10800	1.0546	0.9
0.0	49.0	11025	1.0617	0.8983
0.0	50.0	11250	1.0651	0.8983

Framework versions

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