hushem_5x_deit_base_rms_001_fold1

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: 1.9203
• Accuracy: 0.4889

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
2.2898	1.0	27	1.4240	0.2444
1.4014	2.0	54	1.3895	0.1778
1.3975	3.0	81	1.3921	0.2444
1.3985	4.0	108	1.4006	0.2444
1.4125	5.0	135	1.3918	0.2667
1.3889	6.0	162	1.3878	0.2444
1.3949	7.0	189	1.3872	0.2667
1.4192	8.0	216	1.3990	0.2444
1.3939	9.0	243	1.3853	0.2889
1.3926	10.0	270	1.3626	0.4
1.356	11.0	297	1.2435	0.3778
1.2007	12.0	324	1.1323	0.3778
1.1878	13.0	351	1.2381	0.4
1.1084	14.0	378	1.4461	0.4222
1.0442	15.0	405	1.1938	0.4667
0.9336	16.0	432	1.2799	0.4
1.0799	17.0	459	1.1047	0.5333
0.9885	18.0	486	1.2688	0.4444
0.9753	19.0	513	1.2979	0.4444
0.9374	20.0	540	1.4547	0.4
0.8511	21.0	567	1.1517	0.4222
0.9007	22.0	594	1.4045	0.4444
0.8776	23.0	621	1.3447	0.4222
0.8333	24.0	648	1.2229	0.4889
0.8558	25.0	675	1.4746	0.4
0.8456	26.0	702	1.4066	0.4
0.8034	27.0	729	1.6488	0.3333
0.7686	28.0	756	1.4747	0.3778
0.7992	29.0	783	1.6794	0.4222
0.7462	30.0	810	1.9130	0.4
0.6946	31.0	837	1.5285	0.4
0.7435	32.0	864	1.6866	0.3556
0.6803	33.0	891	1.9290	0.3778
0.6227	34.0	918	1.4309	0.4222
0.6076	35.0	945	1.6567	0.4667
0.585	36.0	972	1.7272	0.3778
0.5538	37.0	999	1.6179	0.4
0.5528	38.0	1026	1.7890	0.3778
0.496	39.0	1053	1.6989	0.3556
0.4841	40.0	1080	1.8596	0.4
0.4712	41.0	1107	1.7704	0.4222
0.4626	42.0	1134	1.8917	0.4222
0.5151	43.0	1161	1.8143	0.4444
0.3695	44.0	1188	2.1532	0.3778
0.3767	45.0	1215	2.1009	0.4222
0.3138	46.0	1242	1.9441	0.4667
0.3279	47.0	1269	2.0701	0.4889
0.2542	48.0	1296	1.9192	0.4889
0.2778	49.0	1323	1.9203	0.4889
0.2872	50.0	1350	1.9203	0.4889

Framework versions

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