hushem_5x_deit_base_rms_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: 5.1764
• Accuracy: 0.5333

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
1.4732	1.0	27	1.5871	0.2667
1.4137	2.0	54	1.4271	0.2667
1.462	3.0	81	1.4098	0.2667
1.4423	4.0	108	1.4316	0.2444
1.4677	5.0	135	1.1736	0.6
1.1753	6.0	162	1.3090	0.4889
1.0628	7.0	189	1.1008	0.4
0.8856	8.0	216	1.3194	0.4667
0.7266	9.0	243	1.5517	0.4667
0.7206	10.0	270	1.5964	0.4222
0.6825	11.0	297	1.9511	0.5333
0.6024	12.0	324	1.1289	0.5111
0.7093	13.0	351	1.6051	0.4667
0.5446	14.0	378	1.0604	0.5333
0.4716	15.0	405	2.6293	0.5778
0.4728	16.0	432	3.2908	0.4889
0.5099	17.0	459	2.0246	0.5333
0.4809	18.0	486	3.4545	0.5333
0.3484	19.0	513	2.2451	0.5111
0.352	20.0	540	2.8572	0.4889
0.3258	21.0	567	3.5970	0.5556
0.2785	22.0	594	3.6404	0.5556
0.3005	23.0	621	3.6333	0.5111
0.2089	24.0	648	4.2561	0.5333
0.1996	25.0	675	3.8526	0.5111
0.1044	26.0	702	4.1245	0.5333
0.2042	27.0	729	3.9154	0.5556
0.1371	28.0	756	3.3906	0.5556
0.1014	29.0	783	4.2534	0.5556
0.0761	30.0	810	3.8328	0.5778
0.0321	31.0	837	4.5117	0.5556
0.1194	32.0	864	4.5296	0.5333
0.0072	33.0	891	4.9299	0.5333
0.0276	34.0	918	5.0433	0.5111
0.0121	35.0	945	4.9519	0.5333
0.0051	36.0	972	4.9546	0.5333
0.0001	37.0	999	4.9700	0.5111
0.0001	38.0	1026	4.9962	0.5111
0.0	39.0	1053	5.0319	0.5111
0.0	40.0	1080	5.0566	0.5111
0.0001	41.0	1107	5.0812	0.5333
0.0	42.0	1134	5.1051	0.5333
0.0	43.0	1161	5.1228	0.5333
0.0	44.0	1188	5.1393	0.5333
0.0	45.0	1215	5.1531	0.5333
0.0	46.0	1242	5.1647	0.5333
0.0	47.0	1269	5.1724	0.5333
0.0	48.0	1296	5.1763	0.5333
0.0	49.0	1323	5.1764	0.5333
0.0	50.0	1350	5.1764	0.5333

Framework versions

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