hushem_5x_deit_base_adamax_0001_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: 1.3720
• Accuracy: 0.6667

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
1.2573	1.0	27	1.3177	0.3778
0.9474	2.0	54	1.2698	0.4667
0.6743	3.0	81	1.1709	0.5333
0.53	4.0	108	1.1238	0.6
0.3327	5.0	135	1.1060	0.6
0.2187	6.0	162	1.0991	0.6444
0.1497	7.0	189	1.1072	0.6444
0.086	8.0	216	1.1220	0.6444
0.0449	9.0	243	1.1215	0.6444
0.0257	10.0	270	1.1368	0.6667
0.0174	11.0	297	1.1587	0.6667
0.0102	12.0	324	1.1715	0.6889
0.0083	13.0	351	1.2117	0.6889
0.0067	14.0	378	1.2042	0.6889
0.0061	15.0	405	1.2320	0.6889
0.0048	16.0	432	1.2396	0.6889
0.0043	17.0	459	1.2501	0.6889
0.0039	18.0	486	1.2585	0.6667
0.0034	19.0	513	1.2714	0.6889
0.0031	20.0	540	1.2786	0.6889
0.0029	21.0	567	1.2831	0.6667
0.0026	22.0	594	1.2886	0.6667
0.0022	23.0	621	1.2985	0.6667
0.0022	24.0	648	1.3036	0.6667
0.002	25.0	675	1.3071	0.6667
0.002	26.0	702	1.3150	0.6667
0.0017	27.0	729	1.3222	0.6667
0.0018	28.0	756	1.3235	0.6667
0.0018	29.0	783	1.3294	0.6667
0.0017	30.0	810	1.3351	0.6667
0.0015	31.0	837	1.3358	0.6667
0.0016	32.0	864	1.3406	0.6667
0.0015	33.0	891	1.3434	0.6667
0.0014	34.0	918	1.3481	0.6667
0.0013	35.0	945	1.3523	0.6667
0.0013	36.0	972	1.3535	0.6667
0.0013	37.0	999	1.3558	0.6667
0.0012	38.0	1026	1.3590	0.6667
0.0012	39.0	1053	1.3619	0.6667
0.0011	40.0	1080	1.3634	0.6667
0.0012	41.0	1107	1.3657	0.6667
0.0011	42.0	1134	1.3669	0.6667
0.0011	43.0	1161	1.3696	0.6667
0.0011	44.0	1188	1.3699	0.6667
0.0011	45.0	1215	1.3707	0.6667
0.0011	46.0	1242	1.3712	0.6667
0.0011	47.0	1269	1.3718	0.6667
0.0011	48.0	1296	1.3720	0.6667
0.0011	49.0	1323	1.3720	0.6667
0.0011	50.0	1350	1.3720	0.6667

Framework versions

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