rtdetr2

This model is a fine-tuned version of PekingU/rtdetr_r50vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 11.8738
• Map: 0.2766
• Map 50: 0.439
• Map 75: 0.2829
• Map Small: 0.1319
• Map Medium: 0.375
• Map Large: 0.4519
• Mar 1: 0.2272
• Mar 10: 0.438
• Mar 100: 0.4559
• Mar Small: 0.2268
• Mar Medium: 0.551
• Mar Large: 0.6909
• Map Person: 0.7159
• Mar 100 Person: 0.8345
• Map Ear: 0.3388
• Mar 100 Ear: 0.4368
• Map Earmuffs: 0.2418
• Mar 100 Earmuffs: 0.5135
• Map Face: 0.4365
• Mar 100 Face: 0.6137
• Map Face-guard: 0.1839
• Mar 100 Face-guard: 0.5625
• Map Face-mask-medical: 0.1733
• Mar 100 Face-mask-medical: 0.2842
• Map Foot: 0.0448
• Mar 100 Foot: 0.2647
• Map Tools: 0.0879
• Mar 100 Tools: 0.284
• Map Glasses: 0.224
• Mar 100 Glasses: 0.3644
• Map Gloves: 0.2994
• Mar 100 Gloves: 0.4226
• Map Helmet: 0.2581
• Mar 100 Helmet: 0.3714
• Map Hands: 0.5167
• Mar 100 Hands: 0.6166
• Map Head: 0.5923
• Mar 100 Head: 0.6739
• Map Medical-suit: 0.1135
• Mar 100 Medical-suit: 0.5556
• Map Shoes: 0.3707
• Mar 100 Shoes: 0.483
• Map Safety-suit: 0.0097
• Mar 100 Safety-suit: 0.315
• Map Safety-vest: 0.0956
• Mar 100 Safety-vest: 0.1532

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 300
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map Person	Mar 100 Person	Map Ear	Mar 100 Ear	Map Earmuffs	Mar 100 Earmuffs	Map Face	Mar 100 Face	Map Face-guard	Mar 100 Face-guard	Map Face-mask-medical	Mar 100 Face-mask-medical	Map Foot	Mar 100 Foot	Map Tools	Mar 100 Tools	Map Glasses	Mar 100 Glasses	Map Gloves	Mar 100 Gloves	Map Helmet	Mar 100 Helmet	Map Hands	Mar 100 Hands	Map Head	Mar 100 Head	Map Medical-suit	Mar 100 Medical-suit	Map Shoes	Mar 100 Shoes	Map Safety-suit	Mar 100 Safety-suit	Map Safety-vest	Mar 100 Safety-vest
No log	1.0	459	12.5722	0.1349	0.2143	0.143	0.0728	0.1775	0.1605	0.1436	0.2943	0.3139	0.1692	0.4142	0.5087	0.4573	0.8371	0.1857	0.4252	0.0002	0.1162	0.3926	0.613	0.001	0.2625	0.0572	0.1719	0.0004	0.102	0.0248	0.1706	0.0501	0.361	0.0807	0.1743	0.059	0.2911	0.2425	0.521	0.4662	0.6855	0.0	0.0222	0.2745	0.4592	0.0016	0.095	0.0002	0.0274
26.3979	2.0	918	11.9482	0.208	0.3216	0.2213	0.1016	0.2792	0.3172	0.1846	0.363	0.3799	0.2024	0.4917	0.6211	0.633	0.8297	0.2283	0.335	0.0016	0.3622	0.5135	0.6272	0.0562	0.4875	0.1298	0.2737	0.0051	0.2137	0.0393	0.2209	0.1385	0.3452	0.2407	0.3783	0.1832	0.3482	0.4565	0.6032	0.5955	0.694	0.0002	0.0444	0.3128	0.4551	0.0009	0.17	0.0009	0.0694
18.1729	3.0	1377	11.9715	0.2184	0.3418	0.2313	0.1126	0.3017	0.347	0.1909	0.3883	0.4101	0.2354	0.5085	0.6091	0.6705	0.8279	0.3031	0.4063	0.0117	0.3919	0.4866	0.6111	0.0355	0.5375	0.1147	0.2825	0.0111	0.2784	0.0514	0.2397	0.2188	0.4102	0.2171	0.3757	0.1851	0.3554	0.4664	0.5985	0.5942	0.6886	0.0014	0.2167	0.3408	0.4824	0.001	0.14	0.0037	0.129
16.7823	4.0	1836	11.5629	0.2362	0.3642	0.2521	0.1213	0.3333	0.3732	0.2058	0.3925	0.4068	0.2236	0.5135	0.6145	0.6948	0.8366	0.3441	0.4262	0.0691	0.4216	0.5345	0.6544	0.0409	0.4625	0.1528	0.3053	0.0125	0.2647	0.0651	0.2936	0.1319	0.3079	0.2895	0.4088	0.2312	0.3634	0.4867	0.5981	0.6117	0.6891	0.0104	0.2056	0.3356	0.4696	0.001	0.115	0.0039	0.0935
15.8096	5.0	2295	11.8381	0.2593	0.4074	0.2739	0.1281	0.3589	0.4355	0.2155	0.4214	0.4365	0.2201	0.5394	0.6675	0.6888	0.8275	0.3686	0.4572	0.1986	0.4378	0.5029	0.6389	0.1532	0.475	0.1791	0.3421	0.0229	0.2137	0.0583	0.2428	0.2389	0.4034	0.2943	0.4434	0.2326	0.3607	0.486	0.601	0.5887	0.6697	0.0124	0.4278	0.3458	0.4595	0.0035	0.25	0.0332	0.1694
14.7312	6.0	2754	11.8431	0.2451	0.3926	0.2562	0.1243	0.3434	0.3958	0.2131	0.412	0.4263	0.2187	0.5118	0.6653	0.6987	0.8307	0.3293	0.4215	0.1879	0.4162	0.4487	0.6174	0.0248	0.4875	0.1453	0.3263	0.0302	0.249	0.0799	0.2737	0.2061	0.3531	0.2911	0.4035	0.2357	0.333	0.5077	0.6144	0.5511	0.6344	0.0402	0.5222	0.3495	0.4789	0.0018	0.15	0.0394	0.1355
14.5207	7.0	3213	11.9739	0.2614	0.4212	0.2679	0.1246	0.3646	0.4079	0.2235	0.418	0.4289	0.2175	0.5351	0.617	0.6899	0.8326	0.3247	0.419	0.2156	0.4541	0.4293	0.6056	0.1801	0.4875	0.1518	0.2807	0.0507	0.2294	0.084	0.2649	0.2206	0.3678	0.2696	0.3903	0.2628	0.3625	0.5094	0.612	0.5925	0.6774	0.021	0.4056	0.3423	0.4637	0.0132	0.3	0.0863	0.1387
14.0764	8.0	3672	11.9295	0.2646	0.4292	0.2715	0.1278	0.3629	0.4405	0.2236	0.4189	0.4308	0.2232	0.5261	0.668	0.7004	0.8286	0.3267	0.4195	0.2031	0.4459	0.4416	0.6123	0.188	0.45	0.1675	0.3053	0.0274	0.2373	0.0765	0.2644	0.2267	0.3921	0.321	0.446	0.245	0.367	0.5036	0.6054	0.5888	0.6741	0.0382	0.3722	0.3382	0.4527	0.019	0.305	0.0858	0.1452
13.772	9.0	4131	11.8863	0.2678	0.4293	0.2743	0.1306	0.3694	0.4456	0.2276	0.4405	0.4524	0.2331	0.5595	0.6747	0.7065	0.8322	0.3349	0.432	0.2274	0.5568	0.4314	0.6113	0.1785	0.5625	0.1666	0.3228	0.0354	0.2608	0.0893	0.2812	0.2225	0.3842	0.2942	0.4235	0.2533	0.367	0.5127	0.6124	0.5853	0.6719	0.043	0.4611	0.3693	0.4833	0.0097	0.245	0.092	0.1839
13.5604	10.0	4590	11.8738	0.2766	0.439	0.2829	0.1319	0.375	0.4519	0.2272	0.438	0.4559	0.2268	0.551	0.6909	0.7159	0.8345	0.3388	0.4368	0.2418	0.5135	0.4365	0.6137	0.1839	0.5625	0.1733	0.2842	0.0448	0.2647	0.0879	0.284	0.224	0.3644	0.2994	0.4226	0.2581	0.3714	0.5167	0.6166	0.5923	0.6739	0.1135	0.5556	0.3707	0.483	0.0097	0.315	0.0956	0.1532

Framework versions

• Transformers 4.46.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1