hushem_5x_deit_small_adamax_00001_fold2

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: 1.4402
• Accuracy: 0.6889

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: 1e-05
• 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
1.2515	1.0	27	1.2615	0.5333
0.9391	2.0	54	1.2122	0.5556
0.6766	3.0	81	1.1537	0.5333
0.5382	4.0	108	1.1591	0.6
0.3747	5.0	135	1.0974	0.6444
0.2681	6.0	162	1.0815	0.6444
0.1892	7.0	189	1.1005	0.6222
0.1214	8.0	216	1.0974	0.6222
0.0863	9.0	243	1.0922	0.6444
0.0475	10.0	270	1.1018	0.6667
0.0299	11.0	297	1.1054	0.6889
0.0156	12.0	324	1.1555	0.6667
0.0095	13.0	351	1.1847	0.6667
0.0067	14.0	378	1.2033	0.6667
0.0051	15.0	405	1.2483	0.6667
0.004	16.0	432	1.2613	0.6667
0.0032	17.0	459	1.2726	0.6667
0.0028	18.0	486	1.2843	0.6667
0.0026	19.0	513	1.2998	0.6667
0.0021	20.0	540	1.3093	0.6667
0.0019	21.0	567	1.3233	0.6667
0.0018	22.0	594	1.3315	0.6667
0.0015	23.0	621	1.3379	0.6667
0.0014	24.0	648	1.3489	0.6667
0.0014	25.0	675	1.3547	0.6667
0.0013	26.0	702	1.3608	0.6889
0.0012	27.0	729	1.3706	0.6667
0.0011	28.0	756	1.3780	0.6667
0.0012	29.0	783	1.3808	0.6889
0.0011	30.0	810	1.3868	0.6889
0.001	31.0	837	1.3922	0.6889
0.001	32.0	864	1.3991	0.6889
0.0009	33.0	891	1.4019	0.6889
0.0009	34.0	918	1.4078	0.6889
0.0009	35.0	945	1.4120	0.6889
0.0008	36.0	972	1.4161	0.6889
0.0008	37.0	999	1.4179	0.6889
0.0008	38.0	1026	1.4222	0.6889
0.0007	39.0	1053	1.4264	0.6889
0.0007	40.0	1080	1.4282	0.6889
0.0007	41.0	1107	1.4320	0.6889
0.0007	42.0	1134	1.4342	0.6889
0.0007	43.0	1161	1.4365	0.6889
0.0007	44.0	1188	1.4366	0.6889
0.0007	45.0	1215	1.4383	0.6889
0.0007	46.0	1242	1.4394	0.6889
0.0007	47.0	1269	1.4399	0.6889
0.0007	48.0	1296	1.4402	0.6889
0.0007	49.0	1323	1.4402	0.6889
0.0007	50.0	1350	1.4402	0.6889

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0